We at AIM Lab are working on applied and core artificial intelligence and machine learning problems such as text mining and natural language processing to understand human emotions and interaction among other things embedded in text. This work includes developing novel sentiment and aspect analysis techniques as well as developing lexicons for emotion detection. We are also engaged in developing smart software solutions for energy constrained devices by generating tools to help developers in creating energy efficient applications.
Artificial Intelligence and Machine Learning Lab (AIM) is conducting research and development in linguistic and computational aspects of languages, specifically of Pakistan and developing Asia. Urdu is a South Asian language spoken by more than 100 million people mostly in Pakistan and India. Urdu has been ranked second among world’s 2,301 languages followed by English with 527 million speakers around the globe, said a report by Washington Post. We are working on Urdu language. Having two startup grant projects for development of tools for Urdu language.
AIM lab aims to create opportunities for local populations to access information and communicate in their local languages, to enable them to use Information and communication technology for their socio-economic benefit.Muhammad Umair Arshad, Farrukh Bashir, Adil Majeed, Mirza Omer Beg, Waseem Shahzad. Corpus for Emotion Detection on Roman Urdu. In Proceedings of the IEEE 22nd International Multitopic Conference 2019 (INMIC), pages 1-6, Islamabad, Pakistan, November 2019
Artificial Intelligence and Machine Learning Lab (AIM) is also conducting research to improve our understanding of social knowledge embedded in text. Identificaiton of relationships from conversations, classification of fake news and information extraction are few of the many topics that interest us.
Saira Qamar, Hasan Mujtaba, Hammad Majeed, Mirza O. Beg. Relationship Identification between Conversational Agents using Emotion Analysis. Cognitive Computation. 2020.
Mubashar Nazr Awan, Mirza O. Beg. TOP-Rank: A TopicalPostionRank for Extraction and Classification of Keyphrases in Text. Computer Speech and Language. Volume 65. 2020.
Bilal Naeem, Aymen Khan, Mirza O. Beg, Hasan Mujtaba. A Deep Learning Framework for Clickbait Detection on Social Area Network using Natural Language Cues. Journal of Computational Social Science, Volume 3, Issue 1, Pages 231-243. 2020.
Current research challenges in natural language processing and understanding include extraction of implied semantic and affect information from text. This includes emotions, aspects, hidden sentiments, references etc. Solving these challenges requires complex formalizations of how the human mind processes language constructs. We use the state of the art Deep Learning techniques to solve some of the unique challenges underlying emotion transfer through text and human conversations.
Hussain Khawaja, Saira Qamar, Mirza Beg. Domain Specific Emotion Lexicon Expansion. In Proceedings of the IEEE International Conference on Emerging Technologies. 2018 (ICET~'18), pages 1-6, Islamabad, Pakistan, December 2018
Noman Dilawar, Hammad Majeed, Mirza O. Beg. Naveed Ejaz, Khan Muhammad, Irfan Mehmood, Yunyoung Nam. Understanding Citizen Issues through Reviews: A Step towards Data Informed Planning in Smart Cities. Applied Sciences, Volume 8, Issue 9, 1589. 2018
Green Computation attempts to connect optimization techniques with programming languages, compilers and computer architecture. It attempts to develop tools and techniques to help developers improve the energy consumption of software on power-constrained mobile devices.
Ahmed Uzair, Mirza O. Beg, Hasan Mujtaba, Hammad Majeed . WEEC: Web Energy Efficient Computing: A Machine Learning Approach. Sustainable Computing: Informatics and Systems. Volume 22. pp 230-243 . Elsevier. 2019
Hareem Sahar, Abdul A. Bangash, Mirza O. Beg. Towards Energy Aware Object-Oriented Development of Android Applications. Sustainable Computing: Informatics and Systems, Elsevier. Volume 21, pp. 28-46, 2019
Abdul Ali Bangash, Hareem Sahr, Mirza Beg. A Methodology for Relating Software Structure with Energy Consumption. In Proceedings of the 17th IEEE International Conference on Source Code Analysis and Manipulation. 2017 (SCAM '17), pages 111-120, Shanghai, China, September 2017
Hamza Alvi, Hareem Sahr, Abdul Ali Bangash, Mirza Beg. EnSights: A Tool for Energy Aware Software Development. In Proceedings of the IEEE International Conference on Emerging Technologies. 2017 (ICET~'17), pages 1-6, Islamabad, Pakistan, December 2017