School of Electrical Engineering & Computer Science

Generative AI

The Generative-AI Research at NUST

At the Generative AI Research Group, we are dedicated to advancing the frontiers of artificial intelligence through cutting-edge research and innovation. Our team is committed to exploring the vast potential of generative AI, pushing the boundaries of what’s possible, and making significant contributions to the field.
  • Exploring new frontiers in Generative AI
  • Projects in Natural Language Processing (NLP), Computer Vision, and Reinforcement Learning.

 

Our Team

Meet Our Research Team

Dr Seemab Latif
Research Project Lead

Dr Mehwish Fatima
Research Group Co-Lead

Dr Fahad Ahmed Satti
Research Group Co-Lead

Dr Muhammad Moazam Fraz
Research Group Co-Lead

Huma Ameer
Data Analysis and Deep Learning

Saher Arshad
Natural Language Processing

Sahar Sohail
Finance and Machine Learning

Sahar Sohail
Finance and Machine Learning

Faiza Qamar
Natural Language Processing

Sana Sajid
Finance and Machine Learning

Iram Tariq
Machine Learning

Dr. Qurat ul Ain
Machine Learning

Hira Tanveer
Machine Learning

Empowering Students through Mentorship

Multitask learning for legal judgement prediction” has won appreciation award at COMPPEC 2024 3MT award, 2024.
Project “BinWatch: Real-Time AI Detection of Littering Actions Near Trash Cans for a Greener Future” has won first position at SEECS Open House, 2024.
Project “BinWatch: Real-Time AI Detection of Littering Actions Near Trash Cans for a Greener Future” won “AI InnoFest 2024” organized by Center of Excellence- Artificial Intelligence Bahria University, Islamabad under the theme of “Smart Cities”, 2024.
FYP “Language learning for Autistic Children” awarded first prize in Best Adjudged Industry Project during Open House 2019.

Generative AI Research

  1. Mehmood, S. Latif, R. Latif, A. Malik, 2024, “DRIFTNET-EnVACK: Adaptive Drift Detection in Cloud Data Streams with Ensemble Variational Auto-encoder Featuring Contextual Network”, IEEE Access, IF 3.4.
  2. Mehmood, S. Latif. N.S.M. Jamail, A. Malik, R. Latif, 2024, “LSTMDD: An optimized LSTM based drift detector for concept drift in dynamic cloud computing”, PeerJ Computer Science IF 3.8.
  3. R. Khattak, A. Salman, S. Ghafoor and S. Latif, 2024, “Multi-modal LSTM network for anomaly prediction in piston engine aircraft”, Heliyon, 10(3), https://doi.org/10.1016/j.heliyon.2024.e25120, IF 4.0.
  4. Qamar, S. Latif, A.A. Shah, 2023, “Techniques, Datasets, Evaluation Metrics and Future Directions of a Question Answering System”, Knowledge and Information System, IF 2.7
  5. Zoya, S. Latif, R. Latif, H. Majeed, and N.S.M. Jamail. 2023. “Assessing Urdu Language Processing Tools via Statistical and Outlier Detection Methods on Urdu Tweets”. ACM Trans. Asian Low-Resource. Lang. Inf. Process. 22, 10, Article 234, https://doi.org/10.1145/3622939 IF 2.0
  6. Altaf, A., Iqbal, F., Latif, R., Yakubu, B. M., Latif, S., & Samiullah, H. (2023). A Survey of Blockchain Technology: Architecture, Applied Domains, Platforms, and Security Threats. Social Science Computer Review, 41(5), 1941-1962. https://doi.org/10.1177/08944393221110148
  7. Zoya, S. Latif, F. Shafait, R. Latif, 2021, “Analyzing LDA and NMF Topic Models for Urdu Tweets via Automatic Labeling”, IEEE Access, vol. 9, pp 127531-127547, IF 3.367, 10.1109/ACCESS.2021.3112620
  8. Latif, M. U. Ahmed, S. Tahir, S. Latif, W. Iqbal, A. Ahmad, 2021, “A Novel Trust Management Model for Edge Computing”, Complex & Intelligent Systems, IF 4.927, https://doi.org/10.1007/s40747-021-00518-3
  9. Waleed, R. Latif, B. M. Yakubu, M. I. Khan, S. Latif, 2021, “T-smart: Trust model blockchain based smart market-place”, Journal of Theoretical and Applied Electronic Commerce Research, vol. 2021 no. 16, pp 2405–2423, IF 3.049, https://doi.org/10.3390/jtaer16060132
  10. Iqbal, S. Latif, Y. Yan, C. Yu, Y. Sh, 2021, “Reducing Arm Fatigue in Virtual Reality by Introducing 3D-Spatial Offset”, IEEE Access, vol. 9, pp. 64085-64104, IF 3.745, doi: 10.1109/ACCESS.2021.3075769
  11. Latif, S. H. Afzaal, S. Latif, 2021, “A Novel Cloud Management Framework for Trust Establishment and Evaluation in a Federated Cloud Environment”, Journal of Supercomputing, IF 2.469, https://doi.org/10.1007/s11227-021-03775-8.
  12. Javed, S. O. Gilani, S. Latif, A. Waris, M. Jamil and A. Waqas, 2021, “Predicting Risk of Antenatal Depression and Anxiety Using Multi-Layer Perceptron and Support Vector Machines”, Journal of Personalized Medicine, Vol. 199, No. 11, IF: 4.433.
  13. Khan, R. Latif, S. Latif, S. Tahir, G. Batool, T. Saba, 2020, “Malicious Insider Attack Detection in IoTs Using Data Analytics”, IEEE Access, vol. 8, pp. 11743 – 11753, Jan 2020. IF: 4.098.
  14. Ahmed, R. Latif, S. Latif, H. Abbas, 2018, “Malicious Insiders Attack in IoT based Cloud e-Healthcare environment: A Systematic Literature Review”, Multimedia Tools and Applications, vol. 77, pp. 21947–21965, Impact Factor 1.53.
  15. Durrani, S. Latif, R. Latif, H. Abbas, 2018, “Detection of Denial of Service (DoS) Attack in Vehicular Ad Hoc Networks: A Systematic Literature Review”, Ad-Hoc and Sensor Wireless Networks, vol. 42, no. 1-2, pp. 35-61, Impact Factor 1.034.
  16. Yaqoob, S. Latif, R. Latif, H. Abbas, A. Yaseen, 2017, “An Adaptive Rule-Based Approach to Resolving Real-Time VoIP Wholesale Billing Disputes”, Journal of Information Science and Engineering, vol. 33, no. 6, pp. 1433- 1446, Impact Factor 0.414.
  17. Tariq, S. Latif, 2016, “A Mobile Application to Improve Learning Performance of Dyslexic Children with Writing Difficulties”, Educational Technology and Society, vol. 9, no. 4, pp. 151-166, Impact Factor 1.376 .
  18. Abbas, R. Latif, S. Latif, A. Masood, 2016, “Performance Evaluation of Enhanced Very Fast Decision Tree (EVFDT) Mechanism for Distributed Denial of Service Attack Detection in Healthcare Systems”, Annals of Telecommunications, vol. 71, no. 9-10, pp. 477–487, Impact Factor 0.48.
  19. Latif, H. Abbas, S. Latif, A. Masood, 2016 “DDOS Attack Source Detection Using Efficient Traceback Technique (ETT) in Cloud-Assisted Healthcare Environment”, In Special Issue on Advances in Big-Data based mHealth Theories and Applications, Journal of Medical Systems, 40(7), pp. 161, Impact factor 2.24.
  20. Latif, H. Abbas, S. Latif, A. Masood, 2015, “EVFDT: An Enhanced Very Fast Decision Tree Algorithm for Detecting Distributed Denial of Service Attack in Cloud-Assisted Wireless Body Area Network,” Mobile Information Systems, vol. 2015, Impact Factor 0.94
  21. Latif, H. Abbas and S. Latif, 2015, “Distributed Denial of Service DDoS attack detection using data mining approach in cloud Assisted Wireless Body Area Networks”, International Journal of Ad hoc and Ubiquitous Computing (IJAHUC), vol. 23, no. 1-2, pp. 24-35, Impact factor: 0.55.
  22. L. Bhatti and S Latif, 2014 “Knowledge Sharing Intentions in Doctors of Private and Government Hospitals” In Journal of Issues in Business Management and Economics, Global Impact Factor 1.2668.
  23. L. Bhatti and S Latif, 2014 “The Impact of Visual Merchandising on Consumer Impulse Buying Behavior” In Eurasian Journal of Business and Management, 2(1), pp. 24-35.
  • Latif, R. Tariq, S. Tariq, R. Latif, 2015, “Designing an Assistive Learning Aid for Writing Acquisition: A Challenge for Children with Dyslexia”, Studies in Health Technology and Informatics, Volume 217: Assistive Technology, pp180-188, DOI: 10.3233/978-1-61499-566-1-180.
  • Latif, H. Abbass, S. Assar and S. Latif, 2014, “Analyzing feasibility for deploying Very fast Decision Trees for detecting DDoS attack in Cloud assisted WBAN”, In Intelligent Computing Methodologies, Lecture Notes in Computer Science Springer Scopus-Indexed.
  • Ahmed, S. A. Rauf and S. Latif, “Leveraging Large Language Models and Prompt Settings for Context-Aware Financial Sentiment Analysis,” 5th IEEE International Conference on Advancements in Computational Sciences (ICACS), 2024, pp. 1-9, DOI: 10.1109/ICACS60934.2024.10473283.
  • Mehmood, S. Latif, “Dynamic Big Data Drift Visualization of CPU and Memory Resource Usage in Cloud Computing”. Artificial Intelligence Applications and Innovations. AIAI 2022. IFIP Advances in Information and Communication Technology, pp. 27-36, vol. 646. Springer, Cham. https://doi.org/10.1007/978-3-031-08333-4_3.
  • Sarwar, S. Latif, R. Irfan, A. Hasan, F. Shafait, “Text Summarization from Judicial Records Using Deep Neural Machines”, International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME), 2022, pp 1132-1137.
  • Ahmed, S. Latif, R. Irfan, A. Hasan, F. Shafait, “Comparison of Transformer Models for Information Extraction from Court Room Records in Pakistan”, International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME), 2022, pp. 1190-1195.
  • Azhar and S. Latif, “Roman Urdu Sentiment Analysis Using Pre-trained DistilBERT and XLNet,” 2022 Fifth International Conference of Women in Data Science at Prince Sultan University (WiDS PSU), 2022, pp. 75-78, doi: 10.1109/WiDS-PSU54548.2022.00027.
  • Kulsoom, S. Latif, T. Saba and R. Latif, “Students Personality Assessment using Deep Learning from University Admission Statement of Purpose,” 2022 7th International Conference on Data Science and Machine Learning Applications (CDMA), 2022, pp. 224-229, doi: 10.1109/CDMA54072.2022.00042.
  • Arif, S. Latif and R. Latif, “Question Classification Using Universal Sentence Encoder and Deep Contextualized Transformer,” 2021 14th International Conference on Developments in eSystems Engineering (DeSE), 2021, pp. 206-211, doi: 10.1109/DeSE54285.2021.9719473.
  • Iqra, S. Latif, R. Latif, “Impact of Data Representation Techniques on Document Compliance Checking”, Proceedings of 17th International Conference on Foundations of Computer Science (FCS’21), July 2021.
  • M. Raza, Z. S. Butt, S. Latif, A. Wahid, “Sentiment Analysis on COVID Tweets: An Experimental Analysis on Impact of Count Vectorizer and TF-IDF on Sentiment Predictions Using Deep Learning Models”, Proceedings of IEEE 2021 International Conference on Digital Futures and transformative Technologies (ICoDT2), May 2021.
  • Khurshid, S. Latif, R. Latif, “Transfer Learning Grammar for Multilingual Surface Realisation”, Proceedings of IEEE 2021 International Conference on Digital Futures and transformative Technologies (ICoDT2), May 2021.
  • M. Rani, S. Latif, A. Tahir, R. Mumtaz, “A Survey of Sentiment Analysis of Internet Textual Data, and Application to Pakistani YouTube User Comments”, Proceedings of IEEE 2021 International Conference on Digital Futures and transformative Technologies (ICoDT2), May 2021.
  • Riaz, S. Latif, R. Latif, “From Transformers to Reformers”, Proceedings of IEEE 2021 International Conference on Digital Futures and transformative Technologies (ICoDT2), May 2021.
  • F. Fatima, S. Latif, R. Latif, “Fine Tuning BERT for Unethical Behavior Classification”, Proceedings of IEEE 2021 International Conference on Digital Futures and transformative Technologies (ICoDT2), May 2021.
  • Gerard, R. Latif, S. Latif, W. Iqbal, T. Saba and N. Gerard, ” MAD-Malicious Activity Detection Framework in Federated Cloud Computing”, Proceedings of the DeSE 2020, 13th International Conference on Developments in e-Systems Engineering, Dec 2020, United Kingdom, pp 273 – 278.
  • Amir, R. Latif, N. Shafqat. and S. Latif, ” Crowdsourcing Cybercrimes through Online Resources”, Proceedings of the DeSE 2020, 13th International Conference on Developments in e-Systems Engineering, Dec 2020, United Kingdom, pp 158 – 163.
  • Latif , S. Bashir, M. M. A. Agha and R. Latif , “Backward-Forward Sequence Generative Network for Multiple Lexical Constraints”, Proceedings of the AIAI 2020 16th International Conference on Artificial Intelligence Applications and Innovations, Jun 2020, Porto Carras Grand Resort, Halkidiki, Greece.
  • Latif, F. S. Faizan and A. M. Zaidi, “i* (iStar) Security Hierarchy for Cloud Computing”, Proceedings of the ICSEA 2019, The Fourteenth International Conference on Software Engineering Advances, Nov 2019, Valencia, Spain
  • Mehmood, S. Latif and S. Malik, “Prediction of Cloud Computing Resource Utilization”, Proceedings of the 2018 15th International Conference on Smart Cities: Improving Quality of Life Using ICT & IoT (HONET-ICT), Nov 2018, Islamabad, Pakistan.
  • Rahman, S. and S. Latif, “Lightweight Detection of Malicious Nodes in Mobile Ad Hoc Networks”, Proceedings of the International Conference on Communication Technologies (ComTech-2017), Jan 2017, Islamabad , Pakistan.
  • Raziq and S. Latif, 2016, “Pakistan Sign Language Recognition and Translation System using Leap Motion Device” In proceedings of Advances on P2P, Parallel, Grid, Cloud and Internet Computing, Lecture Notes on Data Engineering and Communications Technologies, pp. 895-902.
  • Tariq and S. Latif, 2015, “Designing a Learning Aid to Assist the Dyslexic Children with Writing Difficulties” In Proceedings of INTED 2015 ISI Indexed Conference Proceedings.
  • Jahangir, N. Yaqoob and S. Latif, 2015, “Identification and Improvement of Design Issues of a Sales Management System for VOIP Wholesaler”, In proceedings of 2015 National Software Engineering Conference (NSEC).
  • Ahmed, M. Adil and S. Latif, 2015, “Web Application Prototype: State-of-art Survey Evaluation”, In proceedings of 2015 National Software Engineering Conference (NSEC).
  • Ashraf and S. Latif, 2014, “Handling Intrusion and DDoS Attacks in Software Defined Networks Using Machine Learning Techniques”, In proceedings of IEEE National Software Engineering Conference.
  • Awan and S. Latif, 2014, “Synthesis of an Adaptive CPR Filter for Identification of Vehicle Make & Type”, In proceedings of IEEE National Software Engineering Conference.
  • L. Bhatti and S. Latif, 2013, “The Impact of Visual Merchandising on Consumer Impulse Buying Behavior”, In 4th Asia-Pacific Business Research Conference, Singapore.
  • Latif, M.M Wood and G. Nenadic, 2012 “Improving HCCA using Automatic Summarization”. In Proceedings of International Conference on Machine Learning and Computing (ICMLC) IPCSIT vol. 25, Hong Kong, pp. 39-43.
  • Saleem and S. Latif, 2012 “Information Extraction from Research Papers by Data Integration and Data Validation from multiple Header Extraction Sources”. In World Congress on Engineering and Computer Science (WCECS), San Francisco, USA
  • Rasheed and S.Latif, 2012 “Dictionary based Urdu Word Segmentation using Maximum Matching Algorithm”. In 24th International Conference on Asian Language Processing (IALP), Hanoi, Vietnam
  • Rasheed and S. Latif, 2012 “Dictionary based Urdu Word Segmentation using Dynamic Matching Algorithm”. In Conference on Language and Technology (CLT), Lahore, Pakistan.
  • Latif, M.M Wood and G. Nenadic, 2009 “Correlation between Human Assessment of Essays and ROUGE Evaluation of Essays’ Summaries”. In Proceeding of the 8th International Symposium on NLP (SNLP 2009 Bangkok, Thailand) pp 122-127.
  • Latif, 2009 “Automatic Summarization as a Pre-processing Technique for Document Clustering”. In Proceedings of the 12th Annual CLUKI Research Colloquium, The UK and Ireland special interest group for computational linguistics, University College Dublin (CLUKI 2009 Ireland).
  • Latif and M.M Wood, 2009 “A Novel Technique for Automated Linguistic Quality Assessment of Students’ Essays Using Automatic Summarizers”. In Proceedings of World Congress on Computer Science and Information Engineering, IEEE Computer Society (CSIE 2009 Los Angeles, USA) pp. 144-149.
  • Latif and M. Wood, 2008 “Text Pre-processing for Document Clustering”.  In Proceedings of the Natural Language and Information Systems, 13th International Conference on Applications of Natural Language to Information Systems (NLDB 2008 London, UK) pp. 358–359.
  • Latif, 2008 “Text Pre-processing for Document Clustering”.  In Proceedings of the 11th Annual CLUK Research Colloquium, The UK special interest group for computational linguistics, Oxford University Computing Laboratory (CLUK 2008 Oxford, UK).
  • Latif and S. Khan, 2006 “Correlations and Associations Analysis for Identification and Prediction of Relationships and Their Activeness”. In Proceedings of 2nd IEEE International Conference on Information and Communication Technologies: from Theory to Applications (ICTTA 2006 Damascus, Syria) pp. 1855- 1859.
  • Latif and A. Khalid, 2004 “Distributed Intelligent Network for Recording, Retrieval, and Prediction of Criminal Activities”. In Proceedings of IEEE International Conference on Software Engineering Applications Islamabad Session (ICSEA 2004 Islamabad, Pakistan).
  • Machvis Lab
  • NUST-Careem Next Gen Lab in collaboration with Careem Dubai, 2024.
  • China Pakistan Intelligent Systems Lab (CPInS) in collaboration with Guangzhau Institute of Software Application Technology (GZIS) and Cogniser, China, 2022.

Question-Answering System for Quran and Hadith

Our project, “Question Answering System for Quran and Hadith,” focuses on advancing natural language processing in religious texts. We’ve published a comprehensive survey highlighting techniques, datasets, and evaluation metrics for QA systems, laying out future research pathways. Leveraging Large Language Models (LLMs), including a RAG-based approach, we’ve developed a robust system for addressing long-form queries about the Quran, Tafseer, and Ahadith. Through curated datasets and LLM fine-tuning, we’ve established a strong foundation for accurate, insightful responses. This work aims to enhance understanding and accessibility of religious texts using advanced AI technologies.

Event-Driven Investment Recommendations using Synthesized Financial News for Stock Market

Our research delves into automating data-driven trading decisions by analyzing the correlation between news events and stock market trends, specifically tailored for the Pakistan Stock Exchange. We have identified a significant 1:2 ratio between positive and negative news events, highlighting the pronounced impact of negative news on market volatility. This study aims to enhance investment strategies by synthesizing financial news to provide informed and timely recommendations aligned with market dynamics.

Fine-tuning Urdu NER Models Using Context-Aware Embeddings

Our project involves fine-tuning Urduhack’s roberta-urdu-small model on the MK-PUCIT corpus to enhance Named Entity Recognition (NER). Specifically, we focus on extracting “Person” entities for PEP analysis and “Organization” entities for stock trend analysis from Urdu news articles. Achieving a 94.2% F1-score, our model has identified mislabeling errors within the corpus, prompting ongoing efforts to correct these errors, expand the dataset, and introduce a new “Designation” tag for people’s titles. This work aims to improve the accuracy and applicability of NER models in Urdu language processing.

Aspect based sentiment analysis of urdu text
Our team has led advancements in Urdu language processing through three pioneering research papers leveraging deep learning techniques:

We explored deep learning-based topic modeling techniques like LDA, NMF, and others on a dataset of 0.8 million Urdu tweets collected via Twitter API with diverse hashtags for enhanced topic diversity. Through rigorous pre-processing and feature extraction, including n-grams, we evaluated model effectiveness using qualitative and quantitative measures. Additionally, we introduced innovative deep learning methodologies for outlier detection and normalization algorithms, significantly improving Urdu language processing tools’ tokenization accuracy and robustness across 50 million tweets. Our implementation of a deep learning-based weakly supervised ABSA model using BiLSTM classifiers for Urdu tweets demonstrated superior sentiment and aspect classification performance, underscoring our commitment to advancing natural language processing in social media data.