Prof. Dr. Waqar ul Qounain
Address: Admin Block, Punjab University College of Information Technology,
University of The Punjab,
Allama Iqbal Campus, Shahrah-e-Quaid-e-Azam (The Mall), 54000 Lahore, Pakistan.
Phone: +92.42.111923923-ext-524 Fax: +92.42.92120905
E-mail: s w jaffry at pucit dot edu dot pk
Internet Footprint:
LinkedIn,
ResearchGate,
The DBLP Computer Science Bibliography,
Google Scholar,
Microsoft Academic Research,
Nature Network,
Mendeley,
Introduction:
Dr. Waqar is the Director of Artificial Intelligence and Multidisciplinary (AIM) Resaerch Lab,
Agent-based Computational Modelling Lab,
National Centre of Artificial Intelligence, University of the Punjab,
and Professor / Chairman of Department of Information Technology,
University of The Punjab Lahore, Pakistan.
He has completed his PhD from Vrije University, Amsterdam, the Netherlands
in the field of Artificial Intelligence during 2007 - 2011. His PhD research focused on the Analysis and Validation of Models for Trust Dynamics.
This research is applied in the EU funded project SOCIONICAL.
His current research interests include Computational Modelling, Agent Based Systems, Social Simulations, Data Mining and Text Analytics.
For more details see CV updated 31-03-2021.
Projects and Funding:
- 2019 - 2022: Establishment of National Centre of Artificial Intelligence, Higher Education Commission, Pakistan
- 2018 - 2021: Design of Humanistic Driver Agent, National Research Program for Universities (NRPU) Higher Education Commission, Pakistan
- 2017 - 2018: Information Extraction from Research Articles, Ignite, National ICT RnD Fund, Pakistan
- 2017 - 2018: Design of Virtual Reality Simulator, Ignite, National ICT RnD Fund, Pakistan
- 2017 - 2018: Evaluation of Autonomous Vehicle using Virtual Reality, Ignite, National ICT RnD Fund, Pakistan
- 2017 - 2018: Design of Software for Ergonomic Hazard Detection to Computer Users, Faculty Research Grant, University of the Punjab, Lahore, Pakistan
- 2015 - 2016: Electroid - Smart Homes, Ignite, National ICT RnD Fund, Pakistan
- 2015 - 2016: Prediction System Based on Library Data, Ignite, National ICT RnD Fund, Pakistan
- 2014 - 2015: Anlysis of Social Media Agent, Faculty Research Grant, University of the Punjab, Lahore, Pakistan
- 2014 - 2015: Learning Optimal Driver Personality Profile, Ignite, National ICT RnD Fund, Pakistan
- 2013 - 2014: Human Acceptance of Autonomous Social Media Agent, Ignite, National ICT RnD Fund, Pakistan
- 2013 - 2014: Empirical Validation of Model for Human Decision Making, Faculty Research Grant, University of the Punjab, Lahore, Pakistan
- 2012 - 2013: Human Resource Development for Data Mining and Text Analytics, PolyVista Inc., Houston, TX 77057, USA
- 2012 - 2013: Validation of Trust based Human Decision Making, Ignite, National ICT RnD Fund, Pakistan
- 2012 - 2013: Unsupervised Deception Detection in Text, Ignite, National ICT RnD Fund, Pakistan
- 2012 - 2013: Automated Scientific Literature Acquisition for Systematic Literature Review, Faculty Research Grant, University of the Punjab, Lahore, Pakistan
Teaching:
- Computational Modeling, PUCIT, University of the Punjab, Lahore, Pakistan
- Advance Computational Modeling, PUCIT, University of the Punjab, Lahore, Pakistan
- Crowd Simulation , PUCIT, University of the Punjab, Lahore, Pakistan
- Topics in Artificial Intelligence, PUCIT, University of the Punjab, Lahore, Pakistan
- Bioinformatics and Artificial Intelligence, Institute of Biochemistry and Biotechnology, University of the Punjab, Lahore, Pakistan
- Data Mining and Information Retrieval , PUCIT, University of the Punjab, Lahore, Pakistan
- Natural Language Processing, PUCIT, University of the Punjab, Lahore, Pakistan
- Text Mining, PUCIT, University of the Punjab, Lahore, Pakistan
- Research Methods in Information and Computational Sciences, PUCIT, University of the Punjab, Lahore, Pakistan
- Mathematics for Computer Science, PUCIT, University of the Punjab, Lahore, Pakistan
- Managing Information, Institute of Communication Studies, University of the Punjab, Lahore, Pakistan
- Behavior Dynamics, Department of Computer Science, VU University, Amsterdam, the Netherlands
- Introduction to Computing , PUCIT, University of the Punjab, Lahore, Pakistan
- Introduction to Programming Concepts, PUCIT, University of the Punjab, Lahore, Pakistan
- Object Oriented Programming, PUCIT, University of the Punjab, Lahore, Pakistan
- Data Structures and Algorithms, PUCIT, University of the Punjab, Lahore, Pakistan
- Advance Computer Programming, Department of Computer Science, VU University, Amsterdam, the Netherlands
- System Programming, PUCIT, University of the Punjab, Lahore, Pakistan
- Introduction to Artificial Intelligence, PUCIT, University of the Punjab, Lahore, Pakistan
- Artificial Intelligence, Information Technology University (ITU), Lahore, Pakistan
Graduate Supervision:
- Mufrah Naveed, Modelling of Group Formation Behavior of Pedestrians in Panic Situation.
- Bilal Shahzad, Design and Validation of Performance Evaluation Model for Scrum.
- Muhammad Haris, Analysis of Architectural Definitions of Driver Agents.
- Aroosa Siddique, Modeling of Driver Behavior Under Secondary Task.
- Imran Haider, Evaluation of Agent- based Traffic Demand Model in Subcontinental Urban Context.
- Hifsa Tanveer, Integrating Human Panic Factor in Traffic Flow Models.
- Abdur Rafay, Evolutionary Artificial Intelligence Inspired Building Energy Optimization.
- Khadija Saleem, Computational Modeling of Driver Behavior under Mixed Traffic Conditions.
- Celal Ahmed, Computational Modeling of Driver Behavior under Environmental Perturbation.
- Asif Sohail, Adaptive Blocking and Parameter Configuration for Record Linkage.
- Mian Muhammad Mubasher, Design and Evaluation of Humanistic Driver Agent.
- Zara Nasar, Information Extraction from Full Length Scientific Articles.
- Muhammad Furqan Khan, Design of Virtual Reality based Human-in-the-loop Traffic Simulator.
- Abrar Sabir, Named Entity Recognition from Reported Civil Judgments.
- Junaid Abdullah, An Investigation of Computational Procedures for Building Layout Optimization.
- Qura-tul-Ain Nafees, Modelling of Driver's Honking Behavior.
- Bisma Arshad, Modeling of Driver's Perception using Fuzzy Modeling.
- Alisha Younas, Design of Hierarchical Human Driver Behavior Model.
- Faiza Javed, Modeling of Driver Behavior in Anger.
- Wajiha Batool, Incorporation and Analysis of Task Difficulty in Microscopic Driver Behavior Model.
- Aqsa Chaudhary, Incorporation of Human Factors in Mixed Traffic Model.
- Maria Khurshid, Multi-agent Modeling and Simulation for Pedestrian Movement in an Urban Environment.
- Muhammad Waqas Riaz, Implementation and Validation of Social Agent.
- Shahmin Sharafat, Named Entity Extraction from Law Proceeding of Pakistan.
- Usama Nawaz, Implementation and Validation of Framework for Location Based Social Network.
- Afnan Iftakhar, Mining Information from Criminal Proceedings of the Lahore High Court Pakistan.
- Mufrah Naveed, Modeling of Neurotic Personality Trait of Pedestrians in Panic Situation.
- Maria Javiad, A Generalized Framework for Modeling of Collective Intelligence.
- Ainee Ejaz Mirza, Modeling of Text Complexity in Science Textbooks of Primary Classes.
- Zara Nasar, Information Mining from Research Papers.
- Ali Hussain Khan, Design of a Creative Agent.
- Hafiza Baseerat Fatima, Design and Analysis of a Social Agent.
- Huma Shoaib, Associating Sentiments with Images.
- Anzar Ahmad, An Investigation on Cast Based Agent Oriented Programming (CBAOP) Language Design and Implementation.
- Bilal Shahzad, Data Visualization for Social Networks.
- Imran Latif, Fostering Readers Trust on Wikipedia through Behaviour Analysis of Wikipedia Editors.
- Mian Muhammad Mubasher, A Multi-agent Model for Realistic Urban Traffic Flow Simulation.
- Muhammad Haris, Design of a Conceptual Framework for Location Based Social Networking.
Publications:
Names of the authors are in alphabetical order and all can be regarded as having made a comparable contribution, exceptions are marked with *
-
Arts, M., Avital, M., Bennamar, K., Besselink, T., Boog, W., Collard, P., Colton, S., Corneille, S., Germans, D., Gielink, S., Hanson, A., Hanson, D., Hekkert, P., Hoekendijk, C., Hoorn, J. F., Jaffry, S. W., Jensenius, H., Krabbendam, D., Kresin, F., Kuipers, A., Kuipers, T., Lee, M., Lindenberg, V., Lowe, J., Straffon, L. Miller, A. I., Molella, A., Olthof, T., Ott, C., Otte, M., Poelman, W., Pontier, M., Postma, A., Roosendaal, L., Rump, V., Runco, M., Smeulders, A., Taylor, R., van den Herik, J., van der Kooij, K., van der Sloot, J., van der Veen, O., van Dijk, G., van Gorsel, P., van Rosmalen, B., van Uden, J., Werner, J., Organic Creativity and Physics Within, John Benjamins Publishing, 2013.
online
-
Title: "Organic Creativity and Physics Within."
Authors: Arts, M., Avital, M., Bennamar, K., Besselink, T., Boog, W., Collard, P., Colton, S., Corneille, S., Germans, D., Gielink, S., Hanson, A., Hanson, D., Hekkert, P., Hoekendijk, C., Hoorn, J. F., Jaffry, S. W., Jensenius, H., Krabbendam, D., Kresin, F., Kuipers, A., Kuipers, T., Lee, M., Lindenberg, V., Lowe, J., Straffon, L. Miller, A. I., Molella, A., Olthof, T., Ott, C., Otte, M., Poelman, W., Pontier, M., Postma, A., Roosendaal, L., Rump, V., Runco, M., Smeulders, A., Taylor, R., van den Herik, J., van der Kooij, K., van der Sloot, J., van der Veen, O., van Dijk, G., van Gorsel, P., van Rosmalen, B., van Uden, J., Werner, J.
Abstract
A group of international top scientists from a diversity of disciplines sat together for five days with artists, designers, and entrepreneurs to develop a trans-disciplinary theory of creativity. Organic Creativity and the Physics Within assumes that creativity is a quality of nature visible in physics as well as in psychology, its basis being combinatorics, coincidence, complementarity, and fractal emergence. The authors prompt a mechanism cutting through particle physics, perception, psychology, and culminating into playfulness. Organic Creativity and the Physics Within connects us to the universality of nature's creativeness.
-
Title: "Organic Creativity and Physics Within."
-
Jaffry, S. W., Analysis and Validation of Models for Trust Dynamics, PhD Thesis, VU University Amsterdam , The Netherlands, 2011.
online
-
Title: "Analysis and Validation of Models for Trust Dynamics."
Authors: Jaffry, S. W.
Abstract
Technological advancements in the last century have opened numerous venues and great challenges. The use of machines and devices in everyday human life has assigned them a new societal role. Envisioning the machine to act as an educator, helper, supporter, mediator, negotiator, moderator, doctor, and a daily companion is the debate of the day. Emergence of such a techno-human society has brought many challenges to technologists and social scientists. A main question that still stands is whether such a societal setup will be stable. One of the crucial factors for such a setup to be successful is the humans’ levels of trust in machines. To make such artifacts more human-aware with respect to their trust, dynamical models of trust should be designed, verified, validated and embedded into them. These models will enable the machine to estimate human trust and adapt accordingly. In existing literature there are several computational models of trust which silently assume a rational basis of trust. This performance oriented, system-theoretic view of trust is not the true representation of human behaviour. As reported in many recent studies humans usually do not behave rationally under the influence of personal perception, feeling and biases. This notion of trust is called human-based trust in this dissertation. Both system-theoretic and human-based trust models have their own domain of applications in the current socio-technological world. In these days interactions in the socio-technological world can be classified into three types, namely, human-human, human-machine and machine-machine. System-theoretic trust has a wide application for cases when two autonomous machines communicate or interact with each other. These machines do not have human aspects. Hence, the performance-oriented view of utility, past experiences and institutions enforcements can provide them enough to trust on each other or not. Different from a machine-machine view of interaction, human-based trust has an application in both human-machine and human-human interactions. Hence, understanding of human trust dynamics in these perspectives is necessary for progress and effective utilization of current socio-technological advancements. This dissertation deals with modelling of human-based trust and validation of these models, which is an essential component for human-human and human-machine interaction posed under the challenges of the postmodern human society.
-
Title: "Analysis and Validation of Models for Trust Dynamics."
-
Nasar, Z., Jaffry, S. W., Malik, M. K., Tagging Assistant for Scientific Articles. In: Communications in Computer and Information Science (CCIS) Series, Springer Verlag, 2019, vol. 932, pp. 351-362.
DOI (preliminary version)
-
Title: "Tagging Assistant for Scientific Articles."
Authors: Nasar, Z., Jaffry, S. W., Malik, M. K.
Abstract
With the advent of World Wide Web (WWW), world is being over-loaded with huge data. This huge data carries potential information that once extracted, can be used for betterment of humanity. Information from this data can be extracted using manual and automatic analysis. Manual analysis is not scalable and efficient, whereas, the automatic analysis involves computing mechanisms that aid in automatic information extraction over huge amount of data. WWW has also affected overall growth in scientific literature that makes the process of literature review quite laborious, time consuming and cumber-some job for researchers. Hence a dire need is felt to automatically extract po-tential information out of immense set of scientific articles in order to automate the process of literature review. Such service would require machine learning models to train. Whereas, such model in turn require training dataset. To con-struct a quality dataset often involves employment of annotation tools. There exist wide variety of annotation tools, but none are tailored to assist annotation of scientific articles. Hence in this study, web-based annotation tool for scien-tific articles is developed using Python language. The developed assistant em-ploys state of the art machine learning models to extract metadata from scien-tific articles as well as to process article’s text. It provides various filters in or-der to assist annotators. An article is divided into various textual constructs in-cluding sections, paragraphs, sentences, tokens and lemmas. This division can help annotators by addressing their information need in an efficient manner. Hence, this annotation tool can significantly reduce time while preparing da-taset for full-text scientific articles.
-
Title: "Tagging Assistant for Scientific Articles."
-
*Sharaft, S., Nasar, Z., Jaffry, S. W., Legal Data Mining from Civil Judgments. In: Communications in Computer and Information Science (CCIS) Series, Springer Verlag, 2019, vol. 932, pp. 426-436.
DOI (preliminary version)
-
Title: "Legal Data Mining from Civil Judgments."
Authors: Sharaft, S., Nasar, Z., Jaffry, S. W.,
Abstract
Due to advent of computing, content digitization and its processing is being widely performed across the globe. Legal domain is amongst many of those areas that provide various opportunities for innovation and betterment by means of computational advancements. In Pakistan, since last couple of years, courts have been reporting judgments for public consumption. This reported data is of great importance for judges, lawyers and civilians in various aspects. As this data is growing at rapid rate, there is dire need to process this huge amount of data to better address the need of respective stakeholders. Therefore, in this study, our aim is to develop a machine learning system that can automatically extract infor-mation out of public reported judgments of Lahore High Court. This information, once extracted, can be utilized in betterment for society and policy making in Pakistan. This study takes the first step to achieve this goal by means of extracting various entities from legal judgments. Total ten entities are being extracted that include dates, case numbers, reference cases, person names, respondent names etc. In order to automatically extract these entities, primary requirement was to construct dataset using legal judgments. Hence, firstly annotation guidelines are prepared followed by preparation of annotated dataset for entity extraction. Fi-nally, various algorithms including Markov models and Conditional Random Fields are applied on annotated dataset. Experiments show that these approaches achieve reasonable well results for legal data extraction. Primary contribution of this study is development of annotated dataset on civil judgments followed by training of various machine learning models to extract the potential information from a judgment.
-
Title: "Legal Data Mining from Civil Judgments."
-
Jaffry, S.W., and Treur, J., Modelling Trust for Communicating Agents: Agent-Based and Population-Based Perspectives. In: Lecture Notes in Artifical Intelligence, vol. 6922. Springer Verlag, 2011, pp. 366-377.
DOI (preliminary version)
-
Title: "Modelling Trust for Communicating Agents: Agent-Based and Population-Based Perspectives."
Authors: Jaffry, S.W., and Treur, J.,
Abstract
This paper presents an exploration of the differences between agent-based and population-based models for trust dynamics. This exploration is based on both a large variety of simulation experiments and a mathematical analysis of the equilibria of the two types of models. The outcomes show that the differences between the models are not very substantial, and become less for larger numbers of agents.
-
Title: "Modelling Trust for Communicating Agents: Agent-Based and Population-Based Perspectives."
-
Hoogendoorn, M., Jaffry, S.W., and Maanen, P.-P. van., Validation and Verification of Agent Models for Trust: Independent compared to Relative Trust. In: Advances in Information and Communication Technology Series, Springer Verlag, 2011, vol. 358, pp. 35-50.
-
Title: "Validation and Verification of Agent Models for Trust: Independent compared to Relative Trust."
Authors: Hoogendoorn, M., Jaffry, S.W., and Maanen, P.-P. van.,
Abstract
In this paper, the results of a validation experiment for two existing computational trust models describing human trust are reported. One model uses experiences of performance in order to estimate the trust in different trustees. The second model carries the notion of relative trust. The idea of relative trust is that trust in a certain trustee not solely depends on the experiences with that trustee, but also on trustees that are considered competitors of that trustee. In order to validate the models, parameter adaptation has been used to tailor the models towards human behavior. A comparison between the two models has also been made to see whether the notion of relative trust describes human trust behavior in a more accurate way. The results show that taking trust relativity into account indeed leads to a higher accuracy of the trust model. Finally, a number of assumptions underlying the two models are verified using an automated verification tool.
-
Title: "Validation and Verification of Agent Models for Trust: Independent compared to Relative Trust."
-
Hoogendoorn, M., Jaffry, S.W., and Treur, J., Incorporating Interdependency of Trust Values in Existing Models for Trust Dynamics. In: Nishigaki, M., Josang, A., Murayama, Y., Marsh, S. (eds.), Trust Management IV, Advances in Information and Communication Technology, vol. 321. Springer Verlag, 2010, pp. 263-276.
DOI (preliminary version)
-
Title: "Incorporating Interdependency of Trust Values in Existing Models for Trust Dynamics."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.,
Abstract
Many models of trust consider the trust an agent has in another agent (the trustee) as the result of experiences with that specific agent in combination with certain personality attributes. For the case of multiple trustees, there might however be dependencies between the trust levels in different trustees. In this paper, two alternatives are described to model such dependencies: (1) development of a new trust model which incorporates dependencies explicitly, and (2) an extension of existing trust models that is able to express these interdependencies using a translation mechanism from objective experiences to subjective ones. For the latter, placing the interdependencies in the experiences enables the reuse of existing trust models that typically are based upon certain experiences over time as input. Simulation runs are performed using the two approaches, showing that both are able to generate realistic patterns of interdependent trust values.
-
Title: "Incorporating Interdependency of Trust Values in Existing Models for Trust Dynamics."
-
Both, F., Hoogendoorn, M., Jaffry. S.W., Lambalgen, R. van, Oorburg, R., Sharpanskykh, A., Treur, J., and Vos, M. de, Adaptation and Validation of an Agent Model of Functional State and Performance for Individuals. In: Yang, J.-J.; Yokoo, M.; Ito, T.; Jin, Z.; Scerri, P. (eds.), Lecture Notes in Artificial Intelligence, vol. 5925, Springer Verlag, 2009, pp. 595-607.
DOI (preliminary version)
-
Title: "Adaptation and Validation of an Agent Model of Functional State and Performance for Individuals."
Authors: Both, F., Hoogendoorn, M., Jaffry. S.W., Lambalgen, R. van, Oorburg, R., Sharpanskykh, A., Treur, J., and Vos, M. de,
Abstract
Human performance can seriously degrade under demanding tasks. To improve performance, agents can reason about the current state of the human, and give the most appropriate and effective support. To enable this, the agent needs a model of a specific person's functional state and performance, which should be valid, as the agent might otherwise give inappropriate advice and even worsen performance. This paper concerns the adaptation of the parameters of the existing functional state model to the individual and validation of the resulting model. First, human experiments have been designed and conducted, whereby measurements related to the model have been performed. Next, this data has been used to obtain appropriate parameter settings for the model, describing the specific subject. Finally, the model, with the tailored parameter settings, has been used to predict human behavior to investigate predictive capabilities of the model. The results have been analyzed using formal verification.
-
Title: "Adaptation and Validation of an Agent Model of Functional State and Performance for Individuals."
-
Jaffry, S.W., and Treur, J., Comparing a Cognitive and a Neural Model for Relative Trust Dynamics. In: Leung, C.S., Lee, M., and Chan, J.H. (eds.), Lecture Notes in Computer Science, vol. 5863 Part I, Springer Verlag, 2009, pp. 72-83.
DOI (preliminary version)
-
Title: "Comparing a Cognitive and a Neural Model for Relative Trust Dynamics."
Authors: Jaffry, S.W., and Treur, J.,
Abstract
Trust dynamics can be modelled in relation to experiences. Both cognitive and neural models for trust dynamics in relation to experiences are available, but were not yet related or compared in more detail. This paper presents a comparison between a cognitive and a neural model. As each of the models has its own specific set of parameters, with values that depend on the type of person modelled, such a comparison is nontrivial. In this paper a comparison approach is presented that is based on mutual mirroring of the models in each other. More specifically, for given parameter values set for one model, by automated parameter estimation processes the most optimal values for the parameter values of the other model are determined to show the same behavior. Roughly spoken the results are that the models can mirror each other up to an accuracy of around 90%.
-
Title: "Comparing a Cognitive and a Neural Model for Relative Trust Dynamics."
-
Hoogendoorn, M., Jaffry, S.W., and Treur, J., Modelling Trust Dynamics from a Neurological Perspective. In: Wang, R., Gu, F. (eds.), Advances in Cognitive Neurodynamics II, Springer Verlag, 2011, pp. 523-536.
DOI (preliminary version)
-
Title: "Modelling Trust Dynamics from a Neurological Perspective."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.,
Abstract
Trust is often assumed to depend on experiences. Models for the dynamics of trust in relation to experiences usually have a cognitive nature, leaving affective aspects out of consideration. However, neurological findings show more and more how in mental processes cognitive and affective aspects are intertwined. In this paper, by adopting neurological theories on the role of emotions and feelings, a model for trust dynamics is introduced incorporating the relation between trust and feeling. The model makes use of a Hebbian learning principle and describes how trust does not only depend on experiences viewed as information obtained over time, but also on emotional responses and feelings related to experiences.
-
Title: "Modelling Trust Dynamics from a Neurological Perspective."
-
Hoogendoorn, M., Jaffry, S.W., and Treur, J., Modeling Dynamics of Relative Trust of Competitive Information Agents. In:Klusch, M., Pechoucek, M., Polleres, A. (eds.), Lecture Notes in Artificial Intelligence, vol. 5180, Springer Verlag, 2008, pp. 55-70.
DOI (preliminary version)
-
Title: "Modeling Dynamics of Relative Trust of Competitive Information Agents."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.
Abstract
In order for personal assistant agents in an ambient intelligence context to provide good recommendations, or pro-actively support humans in task allocation, a good model of what the human prefers is essential. One aspect that can be considered to tailor this support to the preferences of humans is trust. This measurement of trust should incorporate the notion of relativeness since a personal assistant agent typically has a choice of advising substitutable options. In this paper such a model for relative trust is presented, whereby a number of parameters can be set that represent characteristics of a human.
International Journal Papers -
Title: "Modeling Dynamics of Relative Trust of Competitive Information Agents."
-
Nasar, Z., Jaffry, S. W., Malik, M. K., Named Entity Recognition and Relation Extraction: State of the Art. In: ACM Computing Surveys. (ACMCS), vol. 54, no. 1, ACM, 2021, page 36.
DOI (preliminary version)
-
Title: "Nasar, Z., Jaffry, S. W., Malik, M. K., Named Entity Recognition and Relation Extraction: State of the Art."
Authors: Nasar, Z., Jaffry, S. W., Malik, M. K.
Abstract
With the advent of Web 2.0, there exist many online platforms that result in massive textual-data production. With ever-increasing textual data at hand, it is of immense importance to extract information nuggets from this data. One approach towards effective harnessing of this unstructured textual data could be its transformation into structured text. Hence, this study aims to present an overview of approaches that can be applied to extract key insights from textual data in a structured way. For this, Named Entity Recognition and Relation Extraction are being majorly addressed in this review study. The former deals with identification of named entities, and the latter deals with problem of extracting relation between set of entities. This study covers early approaches as well as the developments made up till now using machine learning models. Survey findings conclude that deep-learning-based hybrid and joint models are currently governing the state-of-the-art. It is also observed that annotated benchmark datasets for various textual-data generators such as Twitter and other social forums are not available. This scarcity of dataset has resulted into relatively less progress in these domains. Additionally, the majority of the state-of-the-art techniques are offline and computationally expensive. Last, with increasing focus on deep-learning frameworks, there is need to understand and explain the under-going processes in deep architectures.
-
Title: "Nasar, Z., Jaffry, S. W., Malik, M. K., Named Entity Recognition and Relation Extraction: State of the Art."
-
Mubasher, M. M., Jaffry, S. W., Modeling of Individual Differences in Driver Behavior, In: Journal of Ambient Intelligence and Humanized Computing, Springer International Publishing, (JAIHC), Springer International Publishing, 2020, vol. 21, pp. 705-718.
DOI, (preliminary version)
-
Title: "Modeling of Individual Differences in Driver Behavior."
Authors: Mubasher, M. M., Jaffry, S. W.
Abstract
Computational transportation is a scientific discipline which uses traffic flow simulation for intervention design and analysis. Realistic traffic flow simulation depends on realistic computational modeling of individual agents such as drivers. Whereas, realistic agent model relies on realistic modeling of microscopic driver behaviors. IDM and MOBIL are considered de-facto car-following and lane change models respectively. All the prominent microscopic models have been developed with engineering perspective i.e. to reproduce perfect driving behavior. Whereas human driving behavior exhibit individual difference and is prone to risks and errors. This study focuses on development of a personality-based model of driver behavior. The parameters that have been modeled to represent driving preferences have been identified. In their existing forms, model parameters could be assigned arbitrary values from a prescribed range to define different driver profiles. This way, theoretically, infinite driver profiles could be created, many of which does not exist in real. Whereas, literature of traffic psychology suggests that there are few prevalent classes of drivers which exhibit certain behavioral patterns. These classes are characterized with the help of human personality. In proposed study, a relationship between personality traits and model parameters have been modeled. This enhancement may reproduce individual differences in driving behaviors.
-
Title: "Modeling of Individual Differences in Driver Behavior."
-
Title: "Reinforcement Learning Based Adaptive Duty Cycling in LR-WPANs."
Authors: Sarwar, S., Sirhindi, R., Aslam, L., Ghulam, M., Yousaf, M. M., Jaffry, S. W.
Abstract
For conserving energy, duty cycle is defined by setting up the active and sleep periods of network nodes. In beacon enabled networks, to provide support for duty cycle, the IEEE 802.15.4 standard uses optional super-frame structure. This duty cycle is usually fixed and does not consider the topology changes that often occur in dynamic sensor networks. In this paper, existing energy conserving duty cycling approaches for 802.15.4 networks especially the adaptive duty cycling techniques for wireless sensor networks are summed up. Also, this paper highlights the shortcomings of the proposals in the literature, such as induced additional latency, so that they may not support the practical scenarios of Internet of Things (IoT). Further, this study highlights a gross shortcoming that relative performance comparison of RL-based proposals cannot be performed without using a benchmarking framework and real test-bed environment. In this paper, we have presented the future research directions that would lay the foundation for successful development of energy efficient RL-based duty-cycling techniques.
-
Title: "Sharafat, S., Nasar, Z., Jaffry, S. W., Data Mining for Smart Legal Systems."
Authors: Sharafat, S., Nasar, Z., Jaffry, S. W.
Abstract
Smart legal systems carry immense potential to provide legal community and public with valuable insights using legal data. These systems can consequently help in analyzing and mitigating various social issues. In Pakistan, since last couple of years, courts have been reporting judgments online for public consumption. This public data, once processed, can be utilized for betterment of society and policy making in Pakistan. This study takes the first step to realize smart legal system by extracting various entities such as dates, case numbers, reference cases, person names, etc. from legal judgments. To automatically extract these entities, the primary requirement is to construct dataset using legal judgments. Hence, firstly annotation guidelines are prepared followed by preparation of annotated dataset for extraction of various legal entities. Experiments conducted using variety of datasets, multiple algorithms and annotation schemes, resulted into maximum F1-score of 91.51% using Conditional Random Fields.
-
Title: "Textual Keyword Extraction and Summarization: State of the Art."
Authors: Nasar, Z., Jaffry, S. W., Malik, M. K.
Abstract
With the advent of Web 2.0, there exist many online platforms that results in massive textual data production such as social networks, online blogs, magazines etc. This textual data carries information that can be used for betterment of humanity. Hence, there is a dire need to extract potential information out of it. This study aims to present an overview of approaches that can be applied to extract and later present these valuable information nuggets residing within text in brief, clear and concise way. In this regard, two major tasks of automatic keyword extraction and text summarization are being reviewed. To compile the literature, scientific articles were collected using major digital computing research repositories. In the light of acquired literature, survey study covers early approaches up to all the way till recent advancements using machine learning solutions. Survey findings conclude that annotated benchmark datasets for various textual data-generators such as twitter and social forms are not available. This scarcity of dataset has resulted into relatively less progress in many domains. Also, applications of deep learning techniques for the task of automatic keyword extraction are relatively unaddressed. Hence, impact of various deep architectures stands as an open research direction. For text summarization task, deep learning techniques are applied after advent of word vectors, and are currently governing state-of-the-art for abstractive summarization. Currently, one of the major challenges in these tasks is semantic aware evaluation of generated results.
-
Title: "NvPD: Novel Parallel Edit Distance Algorithm, Correctness, and Performance Evaluation"
Authors: Sadiq, U., Yousaf, M. M., Aslam, L., Aleem, M., Sarwar, S., Jaffry, S. W.
Abstract
dit distance has applications in many domains such as bioinformatics for genome analysis, spell checking, plagiarism checking, database query optimization, speech recognition, and data mining for finding patterns of interest in large sequences. Traditionally, edit distance is computed by dynamic programming based solution that works in a sequential manner which becomes infeasible for large problems. In this paper, we introduce NvPD, a novel algorithm for parallel edit distance computation by resolving dependencies in the conventional dynamic programming based solution. We also establish the correctness of modified dependencies. NvPD exhibits certain characteristics such as balanced workload among processors, less synchronization overhead, maximum utilization of resources and it can exploit spatial locality. It requires min(m, n) steps to complete as compared to diagonal based approach that completes in max(m, n). Experimental evaluation using variety of random and real life data sets over shared memory multi-core systems and graphic processing units (GPUs) show that NvPD outperforms state-of-the-art parallel edit distance algorithms.
-
Title: "A Smart Integrated Environment for Vehicular Traffic Simulation."
Authors: Mubasher, M. M., Jaffry, S. W., Yousaf, M. M., Bajwa, I., Sarwar, S., Aslam, L.
Abstract
Globally, urbanization has increased due to the availability of the facilities, jobs, and better standard of living in cities. Increase in urban population has caused increase in urban commute. The increased pressure on urban settlements has made design and optimization of urban civil infrastructure very challenging especially road transport infrastructure. Domain experts such as transportation managers, engineers, scientist, and academics need effective, easy, and readily usable computational modeling and simulation tools to conduct systematic inquiry for design and optimization of the transportation system. Free and open-source software (FOSS) community has developed many vehicular traffic simulation software that are cost-effective and extensible, however are difficult to use. Usually, domain experts do not have hands-on software skills such as XML-based interfaces, software source versioning, and build management. To bridge this gap, in our work, an integrated development environment (IDE), namely, integrated vehicular traffic simulation environment (ITE), has been developed. ITE is developed on top of an existing FOSS multimodal open-source vehicular-traffic simulator (MovSim). ITE includes a scenario building tool for domain experts. It could be used to model road infrastructure and vehicular traffic in an integrated manner. The tool could be used to build and experiment novel traffic infrastructures and policies. Our work helps in reducing learning curve of domain experts by providing a free and open source, easy and readily usable vehicular traffic simulation tool.
-
Title: "Information Mining from Criminal Judgments of Lahore High Court."
Authors: Iftikhar, A., Jaffry, S. W., Malik, M. K.
Abstract
Since the last few years, computers have become a prominent part of the court of law. Courts generate an enormous amount of unstructured text on a daily basis. Extraction of the desired information from this unstructured legal text is one of the major issues. So, there is a need to develop an intelligent system that can automatically find useful and critical information from the available text. Such a system will help judges and lawyers in their judgments and case preparations, common people in understanding law, and finding appropriate lawyer for their legal issues. Therefore, in this research, Punjab University Legal Mining System (PULMS) is developed using three different supervised machine learning algorithms; conditional random field (CRF), maximum entropy (MaxEnt), and trigrams N tag (TNT). To train the system, 304 criminal miscellaneous judgments of the Lahore High Court (LHC) of Pakistan are manually tagged for nine named entities (NE). After training, among three machine learning algorithms, the system achieved significant precision, recall, and f-measure using CRF which are 0.97, 0.87, and 0.89, respectively.
-
Title: "Computational Complexity of Cycle Discrepancy."
Authors: Aslam, L., Sarwar, S., Yousaf, M. M., Jaffry, S. W.
Abstract
The irregularities or the deviations from the state of absolute uniformity are the core focus of the area of discrepancy theory. Cycle discrepancy is about the quantification of the least possible deviation of bi-labeling of graph vertices from an ideal bi-labeling which divides each cycle of a graph in two equal parts. This paper shows that it is NP-hard to compute the cycle discrepancy of a given graph. A polynomial time reduction of Hamiltonian problem to the problem of computing cycle discrepancy is used to establish this result.
-
Title: "Information Extraction from Scientific Articles: A Survey."
Authors: Nasar, Z., Jaffry, S. W., Malik, M. K.
Abstract
In last few decades, with the advent of World Wide Web (WWW), world is being overloaded with huge data. This huge data carries potential information that once extracted, can be used for betterment of humanity. Information from this data can be extracted using manual and automatic analysis. Manual analysis is not scalable and efficient, whereas, the automatic analysis involves computing mechanisms that aid in automatic information extraction over huge amount of data. WWW has also affected overall growth in scientific literature that makes the process of literature review quite laborious, time consuming and cumbersome job for researchers. Hence a dire need is felt to automatically extract potential information out of immense set of scientific articles to automate the process of literature review. Therefore, in this study, aim is to present the overall progress concerning automatic information extraction from scientific articles. The information insights extracted from scientific articles are classified in two broad categories i.e. metadata and key-insights. As available benchmark datasets carry a significant role in overall development in this research domain, existing datasets against both categories are extensively reviewed. Later, research studies in literature that have applied various computational approaches applied on these datasets are consolidated. Major computational approaches in this regard include Rule-based approaches, Hidden Markov Models, Conditional Random Fields, Support Vector Machines, Na?¨ve-Bayes classification and Deep Learning approaches. Currently, there are multiple projects going on that are focused towards the dataset construction tailored to specific information needs from scientific articles. Hence, in this study, state-ofthe-art regarding information extraction from scientific articles is covered. This study also consolidates evolving datasets as well as various toolkits and code-bases that can be used for information extraction from scientific articles.
-
Title: "Optimized Scheduling for Parallel Computing Environment."
Authors: Ahmad, A., Yousaf, M. M., Sarwar, S. Jaffry, S. W., Aslam, L., Khalid, M.
Abstract
Multiple priority packets are aggregated together to form a composite data-burst in an optical burst-switched network. The tail of the composite data-burst contains low-priority packets. It assumes that contention between data-bursts can be resolved by clipping and dropping the tail of data-burst which contains low-priority packets that result in quality-of-service (QoS) differentiation. This mechanism requires implementation of complex preemption-based scheduling scheme which employs increased signaling on the control channel. Further, it fails to handle several contention and scheduling scenarios especially when a data-burst contends at head- and tail-ends, simultaneously. Furthermore, in order to provide maximal class isolation, it has been identified that the best approach is to safeguard the high-priority packets by dropping low-priority packets. Thus, we propose a novel scheme for assembling a composite data-burst in which the packets having high-priority are placed at the tail-end. It has been learnt from the results of simulation that our proposed scheme guarantees maximum class isolation between traffic classes having low and high priority. Furthermore, class isolation is investigated for several percentage of class-0, highest priority, packets in a composite data-burst.
-
Title: "Trust-based Situation Awareness: Comparative Analysis between Agent-Based and Population-Based Modeling."
Authors: Nasar, Z., Jaffry, S. W.
Abstract
The data model of the classical data warehouse (formally, dimensional model) does not offer comprehensive support for temporal data management. The underlying reason is that it requires consideration of several temporal aspects, which involve various time stamps. Also, transactional systems, which serves as a data source for data warehouse, have the tendency to change themselves due to changing business requirements. The classical dimensional model is deficient in handling changes to transaction sources. This has led to the development of various schemes, including evolution of data and evolution of data model and versioning of dimensional model. These models have their own strengths and limitations, but none fully satisfies the above-stated broad range of aspects, making it difficult to compare the proposed schemes with one another. This paper analyses the schemes that satisfy such challenging aspects faced by a data warehouse and proposes taxonomy for characterizing the existing models to temporal data management in data warehouse. The paper also discusses some open challenges.
-
Title: "Cycle Discrepancy of Cubic Toeplitz Graphs."
Authors: Aslam, L., Sarwar, S., Yousaf, M. M., Jaffry, S. W.
Abstract
A Toeplitz graph is one whose adjacency matrix is a Toeplitz matrix. A Toeplitz matrix is also known as a constant diagonal matrix. This paper defines cubic Toeplitz graphs and establishes that the cycle discrepancy of a cubic Toeplitz graph is at most 1. That is cycdisc(G) ? 1, where G is a cubic Toeplitz graph. Further this bound is shown to be tight.
-
Title: "A Novel Parallel Algorithm for Edit Distance Computation."
Authors: Yousaf, M. M., Sadiq, U., Aslam, L., Jaffry, S. W., Sarwar, S.
Abstract
The edit distance between two sequences is the minimum number of weighted transformation-operations that are required to transform one string into the other. The weighted transformation-operations are insert, remove, and substitute. Dynamic programming solution to find edit distance exists but it becomes computationally intensive when the lengths of strings become very large. This work presents a novel parallel algorithm to solve edit distance problem of string matching. The algorithm is based on resolving dependencies in the dynamic programming solution of the problem and it is able to compute each row of edit distance table in parallel. In this way, it becomes possible to compute the complete table in min(m,n) iterations for strings of size m and n whereas state-of-the-art parallel algorithm solves the problem in max(m,n) iterations. The proposed algorithm also increases the amount of parallelism in each of its iteration. The algorithm is also capable of exploiting spatial locality while its implementation. Additionally, the algorithm works in a load balanced way that further improves its performance. The algorithm is implemented for multicore systems having shared memory. Implementation of the algorithm in OpenMP shows linear speedup and better execution time as compared to state-of-the-art parallel approach. Efficiency of the algorithm is also proven better in comparison to its competitor.
-
Title: "Framework for Evaluating Credibility of External Links in Wikipedia."
Authors: Latif, I., Jaffry, S. W., Sarwar, S., Aslam, L., Yousaf, M. M.
Abstract
On the advent of Web 2.0, web users have graduated from mere information-consumers and have become information-producers. Wikipedia is one of the paramount examples of this phenomenon. Open collaborative editing model of Wikipedia allows anyone to contribute information, from anywhere. Hence, general public and particularly researchers are skeptical about the information available at Wikipedia. The whole data set of Wikipedia that ranges from Wikipedia content to editors’ communication is publically available and open to use. Using this information, it seems practical to design models and frameworks to measure authenticity of Wikipedia content. In order to measure authenticity of Wikipedia information, external links play an important role. The presence of external sources or links on a web page to other sites can increase credibility of information as it allows visitors to cross-check information at external sites. However, there should be some mechanism to validate these external sources. In this work, an External Link Verification framework has been proposed and evaluated on External Links of Wikipedia articles. The proposed framework could be used to compute credibility of an external link of any web page.
-
Title: "Novel Composite Burst Assembly for OBS-Networks."
Authors: Sarwar, S., Aslam, L., Jaffry, S. W., Yousaf, M. M.
Abstract
Multiple priority packets are aggregated together to form a composite data-burst in an optical burst-switched network. The tail of the composite data-burst contains low-priority packets. It assumes that contention between data-bursts can be resolved by clipping and dropping the tail of data-burst which contains low-priority packets that result in quality-of-service (QoS) differentiation. This mechanism requires implementation of complex preemption-based scheduling scheme which employs increased signaling on the control channel. Further, it fails to handle several contention and scheduling scenarios especially when a data-burst contends at head- and tail-ends, simultaneously. Furthermore, in order to provide maximal class isolation, it has been identified that the best approach is to safeguard the high-priority packets by dropping low-priority packets. Thus, we propose a novel scheme for assembling a composite data-burst in which the packets having high-priority are placed at the tail-end. It has been learnt from the results of simulation that our proposed scheme guarantees maximum class isolation between traffic classes having low and high priority. Furthermore, class isolation is investigated for several percentage of class-0, highest priority, packets in a composite data-burst.
-
Title: "Temporal and Evolving Data Warehouse Design."
Authors: Faisal, S., Sarwar, M., Shahzad, K., Sarwar, S., Jaffry, S. W. and Yousaf, M. M.
Abstract
The data model of the classical data warehouse (formally, dimensional model) does not offer comprehensive support for temporal data management. The underlying reason is that it requires consideration of several temporal aspects, which involve various time stamps. Also, transactional systems, which serves as a data source for data warehouse, have the tendency to change themselves due to changing business requirements. The classical dimensional model is deficient in handling changes to transaction sources. This has led to the development of various schemes, including evolution of data and evolution of data model and versioning of dimensional model. These models have their own strengths and limitations, but none fully satisfies the above-stated broad range of aspects, making it difficult to compare the proposed schemes with one another. This paper analyses the schemes that satisfy such challenging aspects faced by a data warehouse and proposes taxonomy for characterizing the existing models to temporal data management in data warehouse. The paper also discusses some open challenges.
-
Title: "Analytical Model of Delayed Server System (DSS) for Energy Conservation."
Authors: Sarwar, S., Khalid, H., Aslam, L., Jaffry, S. W., Yousaf, M. M.
Abstract
This paper proposes a delayed server system (DSS) which regulates the number of servers in ON-state, and corresponding OFF-state servers, on the basis of queue occupancy for the sake of energy conservation. Upon arrival of demand, server allocation is delayed if predefined servers are already busy and, thus, customer is going to wait in the queue before it gets service. Hence, DSS allows us to keep some servers at least OFF for fraction of total operational, time resulting in energy conservation. We have developed the Markov chain models for DSS and M/M/n-s systems. Using the Markov chain models, the numerical evaluation has been carried out for the key performance metrics such as blocking probability and average energy conservation. Furthermore, our study has investigated and figured out the optimal configurations of delayed server system in order to maximize the energy conservation.
-
Title: "Sentiment Analysis and Opinion Mining – A Facebook Posts and Comments Analyzer."
Authors: Junaid, S. M., Jaffry, S. W., Yousaf, M. M., Aslam, L., Sarwar, S.
Abstract
Since last few years, the trend of social networking is at its peak. People post their personal feelings and thinking about any topic or product for social liking or for marketing. Such posts often get hundreds or thousands of comments and it becomes difficult for a reader to read all of these to assess public opinion. Sometimes one just wants to know common opinion, behavior, trend or thinking discussed there or to determine whether those opinions are positive or negative. Particularly in case of product marketing, the company would like to judge the success of an ad campaign or new product launch or which products or services are popular and what people like or dislike about particular features of a product. In such situations automatic sentiment analysis and opinion mining can help a lot. Hence, in this paper a novel sentiment analysis and opinion mining framework is proposed. This framework utilizes various techniques of computational linguistics to measure sentiment orientation of user's opinion around different entities. The proposed framework is used to perform sentiment analysis and opinion mining of users' posts and comments on social media through a Facebook App. Furthermore a user study is conducted to gauge performance of the proposed framework. The results of this study have shown that the framework is capable of finding opinions of the users and sentiments around those opinions with more than 85 percent accuracy when compared with actual human judges.
-
Title: "Comparison of Resource Management Frameworks for Processing Big Data."
Authors: Ahmed, S., Yousaf, M. M., Jaffry, S. W., Sarwar, S., Aslam, L.
Abstract
Big data analytic solutions are usually deployed on computer-cluster or cloud computing environment. Efficient resource management is a challenging task in a cluster or cloud environment. Big data processing frameworks hide the details of resource management from the end user. In this paper, we have described the traditional Hadoop and state-of-the-art resource management frameworks used for big data processing. For the evaluation of resource management framework we have identified a feature vector comprising resource assignment, scheduling, security, and physical systemlevel resources. Using the feature vector, resource management frameworks have been compared in order to identify strengths and weaknesses of each. Identified weaknesses indicate the future research directions for further improvement of respective framework. It has been observed that YARN framework qualifies for the most features in the feature vector.
-
Title: "Cycle Discrepancy of d-Colorable Graphs."
Authors: Aslam, L., Sarwar, S., Yousaf, M. M., Jaffry, S. W.
Abstract
We show that cycle discrepancy of a 3-colorable graph, G, on at least five vertices is bounded by 2×??n/3?/2?; that is, cycdisc(G) ?2×??n/3?/2?. We also show that this bound is best possible by constructing 3-colorable graphs, on at least five vertices for which cycle discrepancy is at least 2×??n/3?/2?. Let Gt be the set of 3-colorable graphs on n ? 5 vertices with t vertices in the smallest color class. We show that for a graph, G from Gt, cycdisc(G) ?2×?t/2?. Furthermore a graph G' exists in Gt with large cycle discrepancy, such that cycdisc (G') ?2×?t/2? for t ? 1. We also construct such d-colorable graphs for d > 3 that have maximum possible cycle discrepancy.
-
Title: "Incorporation of the Driver’s Personality Profile in an Agent Model."
Authors: Mubasher, M. M., Jaffry, S.W.
Abstract
Urban traffic flow is a complex system. Behavior of an individual driver can have butterfly effect which can become root cause of an emergent phenomenon such as congestion or accident. Interaction of drivers with each other and the surrounding environment forms the dynamics of traffic flow. Hence global effects of traffic flow depend upon the behavior of each individual driver. Due to several applications of driver models in serious games, urban traffic planning and simulations, study of a realistic driver model is important. Hence cognitive models of a driver agent are required. In order to address this challenge concepts from cognitive science and psychology are employed to design a computational model of driver cognition which is capable of incorporating law abidance and social norms using big five personality profile.
-
Title: "Design and Validation of a Relative Trust Model."
Authors: Hoogendoorn, M., Jaffry, S.W., Maanen, P.P. van, and Treur, J.
Abstract
When considering intelligent agents that interact with humans, having an idea of the trust levels of the human, for example in other agents or services, can be of great importance. Most models of human trust that exist assume trust in one trustee is independent of trust in another trustee. The model introduced here addresses so-called relative trust. The idea of relative trust is that trust in a certain trustee not only depends on the experiences with that trustee, but also on trustees that are perceived competitors of that trustee. Such models for relative trust contain parameters to represent the specific dependency between trust for different trustees. In order to tailor the model towards a specific human, dedicated parameter estimation techniques are used. The validation shows that such a model for relative trust is able to predict human trust based behaviour significantly better compared to a benchmark model.
-
Title: "Agent-Based and Population-Based Modelling of Trust Dynamics."
Authors: Jaffry, S.W., and Treur, J.
Abstract
Trust is usually viewed at an individual level in the sense of an agent having trust in a certain trustee. It can also be considered at a population level, in the sense of how much trust for a certain trustee exists in a given population or group of agents. The dynamics of trust states over time can be modelled per individual in an agent-based manner. These individual trust states can be aggregated to obtain the trust state of the population. However, in an alternative way trust dynamics can be modelled from a population perspective as well. Such a population-level model is much more efficient computationally. In this paper both ways of modelling are investigated and it is analyzed how close they can approximate each other. This is done both by simulation experiments and by mathematical analysis. It is shown that the approximation can be reasonably accurate, and for larger numbers of agents even quite accurate.
-
Title: "Modelling Biased Human Trust Dynamics."
Authors: Hoogendoorn, M., Jaffry, S.W., Maanen, P.P. van, and Treur, J.
Abstract
Within human trust related behaviour, according to the literature from the domains of Psychology and Social Sciences often non-rational behaviour can be observed. Current trust models that have been developed typically do not incorporate non-rational elements in the trust formation dynamics. In order to enable agents that interact with humans to have a good estimation of human trust, and take this into account in their behaviour, trust models that incorporate such human aspects are a necessity. A specific non-rational element in humans is that they are often biased in their behaviour. In this paper, models for human trust dynamics are presented incorporating human biases. In order to show that they more accurately describe human behaviour, they have been evaluated against empirical data, which shows that the models perform significantly better.
-
Title: "Comparative Analysis of Agent-Based and Population-Based Modelling in Epidemics and Economics."
Authors: Bosse, T., Jaffry, S.W., Siddiqui, G.F., and J. Treur
Abstract
This paper addresses comparative evaluation of population-based simulation in comparison to agent-based simulation for different numbers of agents. Population-based simulation, such as for example in the classical approaches to predator-prey modelling and modelling of epidemics, has computational advantages over agent-based modelling with large numbers of agents. Therefore the latter approaches can be considered useful only when the results are expected to deviate from the results of population-based simulation, and are considered more realistic. However, there is sometimes also a silent assumption that for larger numbers of agents, agent-based simulations approximate population-based simulations, which would indicate that agent-based simulation just can be replaced by population-based simulation. The paper evaluates such assumptions by two detailed comparative case studies: one in epidemics, and one in economical context. The former case study addresses the spread of an infectious disease over a population. The latter case study addresses the interplay between individual greed as a psychological concept and global economical concepts. It is shown that under certain conditions agent-based and population-based simulations may show similar results, but not always.
-
Title: "Cognitive and Neural Modeling of Dynamics of Trust in Competitive Trustees."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.
Abstract
Trust dynamics can be modeled in relation to experiences. In this paper two models to represent human trust dynamics are introduced, namely a model on a cognitive level and a neural model. These models each include a number of parameters, providing the possibility to express certain relations between trustees. The behavior of each of the models is further analyzed by means of simulation experiments and formal verification techniques. Thereafter, both models have been compared to see whether they can produce patterns that are comparable. As each of the models has its own specific set of parameters, with values that depend on the type of person modeled, such a comparison is nontrivial. To address this, a special comparison approach is introduced, based on mutual mirroring of the models in each other. More specifically, for a given parameter values set for one model, by an automated parameter estimation procedure the most optimal values for the parameter values of the other model are determined in order to show the same behavior. Roughly spoken the results are that the models can mirror each other up to an accuracy of around 90%.
-
Title: "Design and Validation of a Model for a Human's Functional State and Performance."
Authors: Bosse, T., Both, F., Hoogendoorn, M., Jaffry, S.W., Lambalgen, R. van, Oorburg, R., Sharpanskykh, R., Treur, J., and Vos, M. de.
Abstract
This paper presents a computational model of the dynamics of a human?s functional state in relation to task performance and environment. It can be used in intelligent systems that support humans in demanding circumstances. The model takes task demand and situational aspects as input and determines internal factors such as the experienced pressure, exhaustion and motivation, and how they affect performance. Simulation experiments under different parameter settings pointed out that the model is able to produce realistic behavior of different types of personalities. Moreover, by a mathematical analysis the equilibria of the model have been determined, and by automated checking a number of expected properties of the model have been confirmed. In addition to the internal validation of the model, an experiment has been designed for the purpose of external validation addressing the estimation of the for the application relevant aspects of the human process. Output from the experiment like personality characteristics and performance quality has been used to perform estimation of the parameters of the model. By the parameter estimation a set of parameter values has been identified by which an adequate representation of a person?s functional state when performing a task is achieved.
-
Title: "Agent-Based versus Population-Based Simulation of Displacement of Crime: A Comparative Study."
Authors: Bosse, T., Gerritsen, C., Hoogendoorn, M., Jaffry, S.W., and Treur, J.
Abstract
Central research questions addressed within Criminology are how the geographical displacement of crime can be understood, explained, and predicted. The process of crime displacement is usually explained by referring to the interaction of three types of agents: criminals, passers-by, and guardians. Most existing simulation models of this process take a 'local' perspective, i.e., they are agent-based. However, when the number of agents considered becomes large, more 'global' approaches, such as population-based simulation have computational advantages over agent-based simulation. This article presents both an agent-based and a population-based simulation model of crime displacement, and reports a comparative evaluation of the two models. In addition, an approach is put forward to analyse the behaviour of both models by means of formal techniques. The results suggest that under certain conditions, population-based models approximate agent-based models, at least in the domain under investigation.
-
Title: "Agent based Model for Zakat Distribution."
Authors: Younas, A., Hussain, M., Anam, M., Jaffry, S. W.
Abstract
Equal wealth distribution is a very important economical aspect of a society. If wealth is being circulated throughout society in a well-managed way, it can eliminate many social problems such as poverty, criminal activities, child labor etc. Thus, effective wealth circulation is one of the major building blocks that helps in forming a progressive as well as flourishing society. One of the pertinent research questions in this regard is how wealth circulation can be automated keeping societal dynamics in account. If state of different societies is examined, wealth distribution is getting wicked, as poor are getting poorer over time. As these dynamics affect population at large, hence it is not feasible to experiment social reforms directly in real life. Therefore, in this study an agent-based computational model is proposed. This model could be simulated to conduct different experiments and test various social intervention to understand and solve this problem in detail. In the light of existing literature, there exist several models to address the issue of wealth circulation. Keeping the cultural and religious aspects of a Muslim society in view, in this study, the concepts of inheritance and Zakat are being incorporated in an existing wealth distribution model. In addition to that, proposed model is capable to provide solution for the betterment of wealth circulation in a society. After several simulations it is evident that Zakat and national Zakat treasury centers play a vital role in eradication of improper wealth distribution that is currently prevailing in several societies. In addition to that, awareness among masses regarding the treasury centers can also significantly impact the societal dynamics. Thus, the proposed model can be highly significant for deprived societies where poor class is swelling rapidly.
-
Title: "Modeling the Strategies to Control the Impact of Photochemical Smog on Human Health."
Authors: Chaudary, A., Mubasher, M. M., Jaffry, S. W.
Abstract
Photochemical smog is one of the most dangerous air pollutants which spans over a very large scale around the urban world making it a global crisis. This paper explores how photochemical smog develops and emerges as a major air contamination. Air pollutants such as Nitrogen Oxides (NO, NO 2 ) and Volatile Organic Compounds (VOCs) are notable constituents of photochemical smog. These air pollutants in the presence of sunlight, which acts as a catalyst, undergo a series of reactions to form a layer of ozone (O3) at ground level. This newly formed ozone layer is known to implicate injurious effects on human health. Therefore, in this study, a model is proposed which shows the impact of photochemical smog on health of the human beings. The intention of this paper is to provide solutions for reduction of amount of pollutants in air too. Specifically, it focuses on tree plantation and cloud seeding to reduce the intensity of smog production. The simulation results augment the proposal that effects of these solutions can have positive impact on human health by reducing air pollution.
-
Title: "Extraction of Useful Information from Crude Job Descriptions."
Authors: Tanveer, H., Mubasher, M. M., Jaffry, S. W.,
Abstract
As we know, one of the essential social problems is unemployment. Hence, many job advertising agencies tend to consolidate various job offerings to assist the job seekers. As these job descriptions carry unstructured text, therefore, it makes the process of finding and grouping relevant jobs difficult. Hence, the aim of this paper is to automatically extract information from job descriptions. This involves application of text mining and information extraction approaches. To carry out this study, first various features are extracted and later Gradient Boosting classification algorithm is used to perform information extraction. Fields such as salary, required degree, required experience etc. are being extracted automatically. Results show that various features affect the overall result. This proposed model is currently evaluated with the help of precision, recall, receiver operating characteristics curve, and area under the curve (AUC). Experiments show that boosting classification tends to provide reasonable result with 77% of precision.
-
Title: "Integrating Human Panic Factor in Intelligent Driver Model."
Authors: Tanveer, H., Mubasher, M. M., Jaffry, S. W.,
Abstract
This study aims to explore the effects of human panic factor on drivers’ driving behavior. Most of the car following models focus on idealistic situations aiming for perfection, traffic psychology, however, suggests that emotions do play a significant role in drivers’ behavior which in result effect their driving and decision making. Therefore, it is necessary to incorporate human factors in car following models for better realistic results in driving situations where external task demand increases (for example, poor weather conditions like fog, or making up to a meeting in time). Despite the fact that car following models have sublime appreciation in literature, none of them has focused on incorporating human panic factor in these models. Although some work is being done on understanding panic factor in drivers which helps us to understand their driving behaviors and effect on acceleration under panic situations, but this work is limited to statistical approach. This study is intended to fill this void by reviewing literature and making latest advancements by integrating human panic factor in Intelligent Driver Model (IDM). We attempted to integrate human panic factor in IDM, and simulation-based results verified our assumptions for the enhanced version of IDM. The enhanced version of model namely P-IDM models the acceleration behavior of drivers under panic condition, and reproduces acceleration as intended.
-
Title: "Modeling Task Capability in Full Velocity Differential Model."
Authors: Batool, W., Mubasher, M. M., Jaffry, S. W.,
Abstract
Car following (CF) models formally explain acceleration behavior of drivers. Historically, human factors are not considered in CF models. Attention is a very critical human factor. Drug use, panic, fear, or anger may negatively affect attention and consequently driving behavior. In the recent years, researchers have focused on modeling of CF behavior considering human factors as an outcome of research by traffic psychologists and engineers. These observations make clear that integration of human factors into car following models is necessary to develop a more realistic depiction of CF maneuvers under intricate driving situations. In complex driving situations, it is important to measure the dynamic interaction of driving task demand and ability of driver to handle the task at hand. The basic idea of Task Capability Interface (TCI) model is to incorporate task difficulty and task demand within a framework which gives the detailed account of their influence on one another. Task demand and capability plays a key role in decision making. TCI model has earlier been used to improve two traditional CF models namely Gipps’ model and Intelligent Driver Model (IDM). The enhanced models are referred as TD-Gipps model and TD-IDM. There is another model namely Full Velocity Differential Model (FVDM). Unlike its predecessors, FVDM doesn’t suffer from unrealistic acceleration and deacceleration. But FVDM has not been enhanced using TCI model. In this work, FVDM has been enhanced to incorporate TCI model. The enhanced model namely TD-FVDM has been verified by comparing it with TDGipps using simulation-based experiments. The enhanced proposed model reproduces acceleration behavior as intended.
-
Title: "Trust-based Situation Awareness: Agent-Based versus Population-Based Modeling - A Comparative Study."
Authors: Nasar, Z., Jaffry, S. W.,
Abstract
Trust based situation awareness represents the flow of information regarding a situation within a group of agents based on their mutual trust. Situation Awareness (SA) in general is regarded as basis for efficient group decision making while trust based situation awareness is of utmost importance in safety critical systems as the trust is a key factor that tend to affect the beliefs of agents that are collaborating in an environment. In this study an existing model of situation awareness is extended to incorporate mutual trust of the participating agents. The proposed model is analyzed using two different paradigms of computational modeling techniques namely Agent Based Modeling (ABM) and Population Based Modeling (PBM). Various experiments are conducted using variety of populations. Results show that with incorporation of trust factor, both approaches show similar patterns with large population size. In addition to that, relation between SA and trust is analyzed and the results show that SA gets affected by the inter-agent trust.
-
Title: "Modeling of Neurotic Personality Trait of Pedestrians in Panic Situation."
Authors: Sarwar, M. N., Jaffry, S. W.,
Abstract
Pedestrian crowd simulation is an emerging field of study that simulates large number of pedestrian movement in different environmental conditions. In literature, several macroscopic and microscopic models are proposed to simulate pedestrian movement. However, for realistic pedestrian motion, the most widely used and highly cited microscopic model is the Social Force Model (SFM). This model has capability to simulate pedestrian motion in both normal and panic conditions. In literature, it is observed that the incorporation of personality traits in pedestrian models for crowd simulations is almost nonexistent. Whereas, it is believed that pedestrians’ personality is a very crucial factor hence it should be modeled to simulate pedestrian’s movement realistically. Personality computing is a vast domain, hence in the current study, one of the most relevant personality attribute of the Big Five personality traits known as ‘neuroticism’ is incorporated in SFM. Exhaustive simulations of several experiments are performed and the results are reported. It is found that if personality trait of pedestrian’s neurotic-ness within a crowd is diverse among each other than the evacuation behavior become complex and chaotic. On contrary, if crowd’s neurotic diversity is less and crowd is more homogeneous than smooth patterns in the crowd evacuation time are observed. In literature, best evacuation time for SFM is reported for global panic value 0.4 but on incorporating pedestrian’s neurotic-ness and performing exhaustive experiments with both homogeneous and heterogeneous crowds, best evacuation time is obtained on global panic value 0.3. This value indicates that on adoption of personality attribute in SFM, relatively more individualistic behavior provides better evacuation time than of herding behavior when compared with SFM.
-
Title: "Modeling of Individual Differences in Car-following Behaviour of Drivers."
Authors: Mubasher, M. M., Jaffry, S. W., Jahangir, R.,
Abstract
Car-following models microscopically express acceleration behavior of an individual driver. There are many car-following models each with its own assumptions. Among these car-following models, Intelligent Driver Model (IDM) has been used and cited extensively by research community. All the models including IDM have been developed with engineering perspective i.e. to reproduce perfect acceleration behavior. This study focuses on development of a humanistic car-following model. We have identified humanistic parameters that have been modeled in IDM from mathematical formulation of the model. In its existing form, parameters of IDM could be assigned arbitrary values from a prescribed range to define different driver profiles. This way, theoretically, infinite driver profiles could be created many of which does not exist in real. Literature of traffic psychology suggests that there are few dominant classes of drivers, which exhibit certain behavioral patterns. These dominant classes are characterized with the help of human personality. In our study, we have modeled a relationship between model of human’s personality profile namely Big Five Factors (BFF) and parameters of IDM. The enhanced model let us reproduce individual differences in driving behaviors. The proposed model has been verified using computer simulation to investigate whether proposed humanistic car-following model produce desirable results or not. The proposed car-following model would be able to help in simulating driving behavior of an individual given that personality profile of that individual is known.
-
Title: "Human Acceptance of Autonomous Social Media Agent."
Authors: Fatima, B., Jaffry, S. W., Mubasher, M. M., Sherwani, A.,
Abstract
On the advent of Web 2.0 online social media has emerged as a paramount vehicle for information sharing, business and entertainment. In order to harness this technology in a seamless and effective manner we need intelligent artifacts which can glue various stockholders and fulfill their information need pervasively. Hence, it is the time to design autonomous agents that would be able to take part in active social communication on social media like human beings. To this main question is whether people will accept and trust such systems in their environment or not. So, in this research an autonomous agent is designed that interacts with human beings and perform permissible social actions on the Facebook. Two types of experiments are conducted to compute the agent's expected level of acceptance by human and actual human acceptance of the agent. As the samples are collected from two different populations hence Student's T Test is performed to determine if two sets of data are significantly different from each other. 'Mean Difference' between both populations is 0.034 which shows that if an agent is capable of performing human-like social interaction it increases the human acceptance of the social agent.
-
Title: "Systematic Literature Review of Vehicular Traffic Flow Simulators."
Authors: Mubasher, M. M., Jaffry, S. W.,
Abstract
In modern world means of communication and commutation plays a significant role in collective social growth. It is a well observed phenomenon that due to concentration of educational, career, and business opportunities along with easy access to services, population is increasing in urban centers. Increasing population in urban centers is posing many challenges to urban managers and engineers. Ensuring smooth vehicular traffic flow is very important among them. It includes maintenance of current traffic infrastructure, optimal deployment of traffic regulators and cost effective extension of existing infrastructure which requires experimentation of different what-if scenarios. Urban traffic is among those systems which cannot be interrupted in order to experiment hence simulation is most suitable option for policy makers to perform different what-if experiments. Choice of vehicular traffic flow simulator play a vital role in experimenting these what-if scenarios. This paper presents a systematic review of existing vehicular traffic flow simulators proposed and used in research literature. This review includes type of license i.e. the simulator is closed or open source and freeware or proprietary, usage of the simulator i.e. what type of modeling and simulation tasks can be performed using the simulator, developer of the simulator, and citations of the simulator in research literature.
-
Title: "A Survey of Augmented Reality."
Authors: Shoaib, H., Jaffry, S. W.
Abstract
This paper is a brief survey in the field of Augmented Reality. Augmented Reality is a representation of the world in which real and virtual objects are combined, or the surrounding environment is augmented with virtual objects in real-time to enhance user experience. The survey describes the goals and characteristics of an Augmented Reality System through a novel process providing up-to-date research results in the field. We describe the breadth of applications in the area, real-time solutions to the registration problem, and potential future work for improvement of user experience towards Augmented Reality Systems. This survey provides the only overview of the complete area since the first comprehensive survey was done in 1993 – despite later incomplete attempts.
-
Title: "An Object Based Conceptual Framework for Location Based Social Networking."
Authors: Haris, M., Jaffry S. W.
Abstract
In the current technological era the value of information sharing has emerged enormously while the contemporary phenomenon of Social Networking (SN) has provided an avenue for sharing information. The ubiquitous nature of SN services has focused mainly on “Who”, “What” and “When”, while the “Where” dimension has mainly been neglected. Only recently after realizing that “Where” dimension of information is present in almost 80% of raw data, the SN platforms have started utilizing the location based information. This has led to the emergence of a new field, namely Location Based Social Networking (LBSN). A comprehensive literature review of LBSN reveals several shortcomings in both, the research and industrial implementation. One of the primary weaknesses is that the location in LBSN is being assumed and treated just as an auxiliary part of information (post, pictures, videos etc.) and not as a core element. This treatment undermines the true significance of location based information in LBSN. To overcome this limitation, current paper proposes an object based conceptual framework in which location reforms itself from a mere non-compulsory attribute of information to a completely new form i.e. an object. The location as an object will have its own attributes and associated behaviors. When this new location based information object is integrated into a LBSN platform, the interactions between location and human objects instigates, which resultantly exhibits new aspects of social and spatial communication not witnessed previously in LBSN.
-
Title: "Information Mining from Islamic Scriptures."
Authors: Saeed, A. R., Jaffry S. W.
Abstract
There are billions of believers of various religions in the world and Islam is the second largest religion having 1.6 billion followers. The primary written sources of religious beliefs and practices of Muslims are the Quran and the Hadith (saying and practices of their prophet Muhammad Peace Be Upon Him). Written text of the Quran and the Hadith books is of manageable size and hence state of the art text mining techniques can easily be applied on it. In this paper first a comprehensive review of existing applications offering various type of the Quran and the Hadith information retrieval is presented then a framework based on text mining techniques is presented. Finally an application is developed to demonstrate the Quran and the Hadith information retrieval framework. This application is evaluated with the help of an end user assessment questionnaire. It is recorded that the end users have observed salient advantages of the designed application.
-
Title: "A Survey on Trust Evaluation Mechanisms for Wikipedia."
Authors: Latif, I., Jaffry S. W.
Abstract
Wikipedia is the wellnigh successful and most popular free encyclopedia developed by many editors in collaborative manner. It provides multitude of opportunities for online large scale knowledge sharing between virtual communities by letting the viewer to create and edit articles directly in the web browser. Information on Wikipedia is expanding largely, but the increase in quantity is not proportional to quality of the content. The cursory observer of Wikipedia may not be able to differentiate between the good and the bad quality of the content. Despite the success of Wikipedia, trust on Wikipedia content is still questioned because of its open editing model. In this paper primarily the challenges for trust evaluation mechanisms, caused by the significant characteristics of Wikipedia’s knowledge base are discussed. Existing Wikipedia trust evaluation models are comprehensively sur-veyed and key issues related to these are highlighted. Finally based on this study new dimensions for effective trust evaluation me-chanisms are proposed, which are aimed to setup clear goals for future research in this area.
-
Title: "Empirical Validation of Model for Human Decision Making."
Authors: Mansoor, F., Rohail, M., Jaffry S. W.
Abstract
In current postmodern socio-technical world when machines are everywhere a harmonious relationship between man and machine is essential. The harmony of this relation and survival of this socio-technical world can only be guaranteed if machines can understand human state of mind and can act accordingly. To this, several computational models of human cognition has been presented in literature while very few efforts have been made to validate them. In current paper a model of trust based human decision making in dynamic environment is taken from the literature and validated against human decision traces generated through computer based experiments. The results of this experiment shows that model under study can be trusted as a computational representative of human decision making process with a satisfactory level.
-
Title: "Modeling and Validation of Biased Human Trust."
Authors: Hoogendoorn, M., Jaffry, S.W., Maanen, P.P. van, and Treur, J.,
Abstract
When considering intelligent agents that interact with humans, having an idea of the trust levels of the human, for example in other agents or services, can be of great importance. Most models of human trust that exist, are based on some rationality assumption, and biased behavior is not represented, whereas a vast literature in Cognitive and Social Sciences indicates that humans often exhibit non-rational, biased behavior with respect to trust. This paper reports how some variations of biased human trust models have been designed, analyzed and validated against empirical data. The results show that such biased trust models are able to predict human trust significantly better.
-
Title: "Exploration and Exploitation in Adaptive Trust-Based Decision Making in Dynamic Environments."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.,
Abstract
Trust is generally considered an important aspect in the decision making of agents. In the literature, a variety of computational models for trust have been proposed that also express how an agent can make a decision by exploiting the trust levels it has for the different options. Within such a decision making mechanism the focus is usually on a single most trusted option, which as a side effect may lead to a lack of information on the other options over time. Therefore it may sometimes be worthwhile to be more explorative in the decision making, especially to adapt to dynamic environments. In this paper an adaptive trust-based decision making model is proposed that varies the extent of exploration and exploitation in the agent's decisions. The model is evaluated and compared to other existing models by means of simulations.
-
Title: "An Adaptive Agent Model Estimating Human Trust in Information Sources."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.,
Abstract
For an information agent to support a human in a personalized way, having a model of the trust the human has in information sources may be essential. As humans differ a lot in their characteristics with respect to trust, a trust model crucially depends on specific personalized values for a number of parameters. This paper contributes an adaptive agent model for trust with parameters that are automatically tuned over time to a specific individual. To obtain the adaptation, four different techniques have been developed. In order to evaluate these techniques, simulations have been performed. The results of these were formally verified.
-
Title: "The Influence of Personalities upon the Dynamics of Trust and Reputation."
Authors: Hoogendoorn, M., Jaffry. S.W.,
Abstract
When an agent resides in a community, the opinion of other community members concerning whether a particular individual is trustworthy or not influences the trust level of this agent. Hereby, the precise influence depends on the personality of the agent (e.g. whether he lets his opinion be influenced by others a lot). In this paper, a computational trust model which has dedicated parameters for agent personalities is applied to such a social context. A variety of different communities (containing agents with different personalities) have hereby been simulated. The resulting patterns hereof are shown in this paper. Furthermore, the simulation results are formally analyzed to show that certain patterns do occur in all different communities.
-
Title: "Comparison of Agent-Based and Population-Based Simulations of Displacement of Crime."
Authors: Bosse, T., Gerritsen, C., Hoogendoorn, M., Jaffry, S.W., and Treur, J.
Abstract
Within Criminology, the process of crime displacement is usually explained by referring to the interaction of three types of agents: criminals, passers-by, and guardians. Most existing simulation models of this process are agent-based. However, when the number of agents considered becomes large, population-based simulation has computational advantages over agent-based simulation. This paper presents both an agent-based and a population-based simulation model of crime displacement, and reports a comparative evaluation of the two models. In addition, an approach is put forward to analyse the behaviour of both models by means of formal techniques.
-
Title: "Agent-Based and Population-Based Simulation: A Comparative Case Study for Epidemics."
Authors: Jaffry, S.W., Treur, J.
Abstract
This paper reports a comparative evaluation of population-based simulation in comparison to agentbased simulation for different numbers of agents. Population-based simulation, such as for example in the classical approaches to predator-prey modelling and modelling of epidemics, has computational advantages over agent-based modelling with large numbers of agents. Therefore the latter approaches can be considered useful only when the results are expected to deviate from the results of population-based simulation, and are considered more realistic. However, there is sometimes also a silent assumption that for larger numbers of agents, agent-based simulations approximate population-based simulations, which would indicate that agent-based simulation just can be replaced by population-based simulation. The paper evaluates this assumption by a detailed comparative case study in epidemics.
-
Title: "FOSS Localization: A Solution for The ICT Dilemma of Developing Countries"
Authors: Jaffry, S. W., Kayyani, U. R.
Abstract
Information and communication technology (ICT) has tremendously expanded over the last three decades making the access to right information at the right time feasible ensuring the success of an individual, organization or culture. In order to make the most out of this exciting revolution one must be in a position to afford and completely comprehend what is offered by this technology. Unfortunately most of the software are controlled by proprietary that are economically unaffordable for developing countries and are based on language that is not comprehendible by their masses. Software localization of Free and Open Source Software (FOSS) is an effort that addresses this twofold dilemma. FOSS made software affordable while localization bridges the language barrier that helps people to fully comprehend and utilize the benefits of ICT. In this research we have explored various aspects of the software localization of free and open source operating system (FOSOS) and developed a working prototype. Paper explains concept and all the technical steps of FOSS localization of Ubuntu Linux that is a FOSOS with a foreseeable future work.
-
Title: "Pedagogical Pattern Language for in Time Student Confidence in Studied Material"
Authors: Jaffry, S. W.
Abstract
Over the past few years, importing design patterns from software engineering to the computer science education (CSE) is followed by defining patterns and pattern languages suitable for pedagogical needs of Computer Science courses. The main goal of patterns incorporation in CSE was to enhance quality of content delivery and comprehension. However it is observed that soon after learning a concept most of the students are not certain about the degree of command over the studied material. Lack of questioning power makes it more devastating. The existing pattern-based materials seem to be insufficient for settling these problems. Hence there is a requirement of pedagogical pattern language that have twofold power of resolving these issues. This paper presents a pedagogical pattern language based on three newly discovered pedagogical patterns to address in time students confidence in their level of comprehension about learned skills.
-
Title: "Enhancements in AODV Routing Using Mobility Aware Agents"
Authors: Idrees, M., Yousaf, M. M., Jaffry, S. W., Pasha, M. A., Hussain, S. A.
Abstract
In infrastructure-less mobile and ad-hoc networks the routes have to be refreshed oftenly due to the mobility of the nodes acting as routers. If a node is aware of the mobility of neighboring nodes, the highly mobile node can be avoided to be the part of routes and ultimately reduces the re-route discoveries. This paper introduces mobility aware agents in ad-hoc network nodes and modifies HELLO packets of the AODV protocol to enhance mobility awareness. Mobility aware agent can update its awareness through inquiry and reply to inquiries about neighbors on quasi-periodic bases. On receiving the HELLO packet with GPS coordinates of the originator, agent compares them with previous ones and hence has awareness about the mobility of the originator with reference to itself. The enhancements in the throughput of the network are studied through simulation in OPNET that has shown optimistic results.
-
Title: "Validation of an Agent Model for Human Work Pressure."
Authors: Both, F., Hoogendoorn, M., Jaffry. S.W., Lambalgen, R. van, Oorburg, R., Sharpanskykh, A., Treur, J., and Vos, M. de,
Abstract
Human performance can seriously degrade under demanding tasks. To improve performance, agents can reason about the current state of the human, and give the most appropriate and effective support. To enable this, the agent needs a work pressure model, which should be valid, as the agent might otherwise give inappropriate advice and even worsen performance. This paper concerns the validation of an existing work pressure model. First, human experiments have been designed and conducted, whereby measurements related to the model have been performed. Next, this data has been used to obtain appropriate parameter settings for the work pressure model, describing the specific subject. Finally, the work pressure model, with the tailored parameter settings, has been used to predict human behavior to investigate predictive capabilities of the model. The results have been analyzed using formal verification.
-
Title: "Parameter Estimation for Human Trust in Information Sources."
Authors: Hoogendoorn, M., Jaffry, S.W., and Treur, J.,
Abstract
In this paper, an approach has been presented to learn parameters of a given trust model based upon observed experiences of a human. This approach has been introduced to enable a personal assistant agent to take such human trust into account when giving advice. Hereby, an existing trust model has been taken as a basis. Several methods have been used to enable learning of these parameters, including exhaustive search, Simulated Annealing, bisection, and an extended form of bisection. The process is adaptive in the sense that new experiences can come in, and are taking into consideration by finding the most appropriate parameter setting. The algorithms have been rigorously tested for various cases, and the results thereof have been analyzed using formal verification techniques. The results show that the computation time of the exhaustive search scales up worst, whereas the Simulated Annealing approach scales up best. When looking at the accuracy however, the inverse is true: exhaustive search finds the most accurate point, whereas Simulated Annealing sometimes only comes up with poor solutions. The bisection, and the more advanced extended bisection approach are right in the middle: They do have a higher accuracy and are computationally less expensive. The choice of which method to choose ultimately depends on the domain. For particular domains a higher computation time might be acceptable as long as the results are good, whereas in other more time critical domain speed could be a necessity. In this respect, the bisection approaches are a good combination of both worlds.
-
Title: "Agent-Based and Population-Based Simulation of Displacement of Crime."
Authors: Bosse, T., Gerritsen, C., Hoogendoorn, M., Jaffry, S.W., and Treur, J.
Abstract
Within Criminology, the process of crime displacement is usually explained by referring to the interaction of three types of agents: criminals, passers-by, and guardians. Most existing simulation models of this process are agent-based. However, when the number of agents considered becomes large, population-based simulation has computational advantages over agent-based simulation. This paper presents both an agent-based and a population-based simulation model of crime displacement, and reports a comparative evaluation of the two models. In addition, an approach is put forward to analyse the behaviour of both models by means of formal techniques.