Faculty Achievement View More Faculty Achievements
Published Works
IBA faculty co-authored a paper, titled ''Taxing tobacco as a strategy to reduce consumption and increase public health benefits in Pakistan", published in the Eastern Mediterranean Health Journal, (EMHJ ).
Dr. Tariq Mahmood, Professor of Computer Science, School of Mathematics & Computer Science co-authored a paper, titled ''Taxing tobacco as a strategy to reduce consumption and increase public health benefits in Pakistan", published in the Eastern Mediterranean Health Journal, (EMHJ ).
Abstract: Background: Tobacco consumption poses a significant challenge to global health and contributes to the increase in noncommunicable diseases and premature deaths. Aim: To investigate the potential impact of a 70% tobacco tax on consumption and government revenue in Pakistan.
Methods: We analysed secondary data from 2011 to 2022 (after imposition of a 70% excise tax) from the Pakistan Bureau of Statistics, Pakistan Social and Living Standard Survey, financial yearbooks and Federal Board of Revenue reports for tobacco consumption and government revenue. Variables included tobacco price inflation, per capita income, cigarette price, federal excise duty, and government revenue.
Results: The higher taxes reduced tobacco production by PKR 3.72 billion (≈US$ 13.4 million). Price elasticity analysis indicated an inelastic demand for cigarettes, mostly among the rural populations. Imposition of excise duty of 70% of the retail price caused a decrease in government revenue by PKR 390 million (≈US$ 1.4 million).
Conclusion: Implementing 70% taxation on tobacco products is beneficial, however, to fully realize its benefit, there is a need for strict regulation on brand shifting and illegal trade. Keyword: tobacco taxation, tobacco tax, tobacco use, noncommunicable disease, revenue, excise tax, Pakista
The paper can be accessed: here.
Published Works
IBA faculty co-authors a paper on evolutionary computing to solve product inconsistencies in software product lines
Dr. Tariq Mahmood, Professor, Department of Computer Science, School of Mathematics and Computer Science (SMCS), co-authored a research paper titled 'Evolutionary Computing to solve product inconsistencies in Software Product Lines', published in the Journal of Science of Computer Programming.
Abstract: In Software Product Lines (SPLs), multiple design teams work collectively to configure products. Often, having multiple sub-designs leads to inconsistencies which prevent and delay the configuration of the final product at the end. Solutions are available to identify and resolve these inconsistencies for limited scale SPLs. For large scale and industrial SPLs, resolution of inconsistencies is still a potential research topic. In this paper, we propose, implement and validate an application of evolutionary computation, particularly genetic algorithms, to resolve inconsistencies in large scale SPL products. We label our approach ga-SPLIS (genetic algorithm-Software Product Line Solution). Using a concrete experimental methodology with different feature set sizes and multiple standard performance metrics, we generate results which validate the applicability of ga-SPLIS for the SPL community.
The article can be accessed here.
Published Works
IBA faculty co-authors an article analyzing product configurations in software product lines
Dr. Tariq Mahmood, Professor, Department of Computer Science, School of Mathematics and Computer Science, has co-authored an article titled, "Predictive Analytics for Product Configurations in Software Product Lines", in the International Journal of Computational Intelligence Systems.
The paper is the result of a PhD work and concerns the development of a predictive analytics tool for solving product configuration issues in software product lines. The tool has been tested in real-life software product lines and is the first formal application of machine learning in this domain.
The published article can be accessed here.
Abstract
The paper examines Software Product Line (SPL), that configures software products, often leading to inconsistencies in the resulting product configurations. This paper aims to solve the problem by learning, or mathematically modeling, all previous patterns of feature selection by SPL developers, and then use these patterns to predict inconsistent configuration patterns at runtime.
Recognition
IBA faculty examines the applications of AI at the InsurTech Summit - 2021
Dr. Tariq Mahmood, Professor, Department of Computer Science, School of Mathematics and Computer Science, delivered a talk on 'Linking AI with Healthtech and Insurtech' at the InsurTech Summit – 2021, organized by the CxO Global Forum.
Dr. Mahmood elaborated on applications of Artificial Intelligence to healthcare and insurance processes. Under the healthcare theme, attendees explored the topics including telemedicine, IoT-based body area networks, BI real-time alerts, improved clinical staffing and intelligent healthcare processes. Specific topics regarding insurance included valuation and damage prediction, automated claims processing and automated risk assessment. Recommendations for applying these technologies effectively in business processes were also discussed.
The link to the talk: https://www.youtube.com/watch?v=EVakrFyb3RI
Published Works
IBA faculty co-authors paper on intrusion detection systems
Dr. Tariq Mahmood, Associate Professor, Department of Computer Science, has coauthored a paper titled "A Novel Deep Learning Framework for Intrusion Detection System" in IEEE's International Conference on Advances in the Emerging Computing Technologies (AECT, 2019), which was held in Al-Madinah Al-Munawwarah (KSA).
The paper is based on work of an intensive graduate course project at the IBA Karachi and uses auto-encoder technology to detect network intrusions through data analysis, and Recurrent network technology to predict the type of network intrusion.
The publication can be accessed at: https://ieeexplore.ieee.org/document/9194224
Recognition
IBA faculty delivers keynote address for online deep learning and AI forum
Dr. Tariq Mahmood, Associate Professor, Department of Computer Science, delivered a keynote speech titled "Concept Drift in Machine Learning of Streaming Data: A Systematic Literature Review" at the online NeurIPS Meetup 2020, which was organized by NeurIPS and Aladin Solutions (URL: https://sites.google.com/a/nu.edu.pk/noman-islam/deep-learning-meetup
The talk was based on work done by a research-survey student at the IBA Karachi, Tatheer Fatima, and comprised a systematic literature review to determine different approaches in research for concept drift or change in probability distribution of predicted business KPIs. The talk was well-accepted by industry and academic professionals.
Research Publication
IBA faculty uses data science to predict Pakistan Cricket Team's performance
Dr. Tariq Mahmood has published his paper entitled "Is the performance of a cricket team really unpredictable? A case study on Pakistan team using machine learning" in the Indian Journal of Science and Technology, an HEC-recognized journal. This work targets the application of a comprehensive data science methodology to determine the extent to which the performance of Pakistan's ODI cricket team can be predicted in advance. The main motivation of this work is the "unpredictable" tag which is associated with Pakistan's team. This work shows that, in fact, using intensive data science models, it is possible to extract useful patterns or regularities in Pakistan's performance and to predict this performance with a high accuracy of 82% (before the match starts).
This work can be presented to cricketing authorities of Pakistan or of any cricketing nation to help them understand the reasons for a win or a loss and to predict this performance right at the start of the match.
The publication can be accessed at: https://indjst.org/articles/is-the-performance-of-a-cricket-team-really-unpredictable-a-case-study-on-pakistan-team-using-machine-learning
Research Publication
IBA faculty discusses big data applications for healthcare systems
Dr. Tariq Mahmood has published his paper entitled "Big Data Analytics in Healthcare - A Systematic Literature Review and Roadmap for Practical Implementation" in IEEE's journal of Automatica Sinica (Impact Factor: 5.142). This was done as part of a PhD work and is the most comprehensive systematic literature review on big data and its analytical applications to patient care and related healthcare domains. The article can be accessed at: https://ieeexplore.ieee.org/document/9205683(early access)
