Invited talk at PyDATA Educational Conference 2020
Research InterestsI am intriqued by the challenging problems in artificial intelligence, data mining, natural language processing, and knowledge graphs. In specific, my research is about a class of Neuro-Symbolic AI in which explicit knowledge plays a central role. My dissertation thesis, titled Knowledge-infused Mining and Learning advances the state of the art in five research thrust areas: (1) Recommender Systems, (2) Learning to Rank, (3) Summarization, (4) Conversational AI, and (5) Computational Social Data Science. An important corollary of my research is that it addresses one of the most important hurdles in the wider acceptance of AI: 91% of the companies surveyed indicated the need to have explainable AI, which forms a pertinent component in KiML. By using KiML, I contribute towards this timely need for Interpretable and Explainable Machine Learning. I have demonstrated its benefits in various multidisciplinary research mental healthcare, crisis informatics, conversational information seeking, virtul health assistants, and digital security.
In the following article, I position my understanding of personalized recommendations considering diverse, ambiguous, and sparse environmental variables at p...
2nd International Workshop on Knowledge-infused Learning
The Conversation: Investigating Social Media for COVID19 impact on Mental Health