You are here
Context-aware Recommender Systems in the context of Cognitive Computing
The fundamental question that recommender system aims to answer is “What does the end-user really want?”. If the end-user already knows what s/he specifically wants, s/he would benefit from using a search engine. If the end-user does not know the answer to this question, s/he would benefit from using a recommender system. Taking into account context in recommender system is emerging as a new frontier that significantly improves the level of engagement between the end-users and the recommender system. The majority of traditional recommender systems do not focus on contextual information. Instead, traditional recommender systems tend to focus on entities, users and items. This approach works as long as the circumstances do not impact the system. This limits the use and the performance of traditional recommender system. When circumstances do affect the outcome, it becomes necessary – if not critical – to go beyond a traditional recommender system and to incorporate contextual information. In this talk I will go over different approaches to build context-aware recommender systems.
About the Speaker:
Dr. Ayse Basar Bener is a professor and the director of Data Science Laboratory (DSL) in the Department of Mechanical and Industrial Engineering, and director of Big Data in the Office of Provost and Vice President Academic at Ryerson University. She is a faculty research fellow of IBM Toronto Labs Centre for Advance Studies, and affiliate research scientist in St. Michael’s Hospital in Toronto. Her current research focus is big data applications to tackle the problem of decision-making under uncertainty by using machine learning methods and graph theory to analyze complex structures in big data to build recommender systems and predictive models in health care, software engineering, smart energy grid, and green software. She is a member of AAAI, INFORMS, AIS, and a senior member of IEEE.