Dr. Kalyan Gupta
- Information Systems
- Alumni
Contact Information
| Email: | contact@knexusresearch.com |
I conduct intelligent systems research to develop advanced software prototypes and transition them into commercially viable applications. I lead multi-disciplinary teams through proposal development, project planning, R&D, transition, and business case development. I am interested in artificial intelligence (AI), and information systems, in general.
Thesis Title: A Framework For The Design and Development of Diagnostic Case-Based Reasoning Systems.
Supervisor: Dr. Ali Montazemi
Artificial Intelligence Techniques
- Spatio-temporal Cognition: I develop computational models to understand how we perceive, conceptualize, and act in a 3D space.
- Cognitive Semantics: This is a topic that is closely related to spatio-temporal cognition where I am interested in developing cognitively inspired knowledge representation for understanding and generating spatial language.
- Planning and acting : It turns out that it is hard to make sense of anything without the ability to sense, plan and act. Therefore, I am looking into methods that tie planning and acting with spatial cognition and semantics.
- Information Extraction : This is an area where a natural language processing (NLP) techniques combined with pattern learning from AI are used to autonomatically extract information (e.g., addresses or menus) from plain text or webpages. I have worked on a variety of techniques including one for acronym extraction.
- Standard and Collective Classification : Classification is a task of mapping an information item with a set of descriptive features to a category; for example, identifying a blob in an image as a car or a bird, or categorizing a document. My recent interest has been in collective and relational classification, where we exploit the information surrounding the objects to improve classification accuracy. It turns out that classifiying many co-related items simultaneously can improve the overall classification performance.
- Taxonomic Case-Based Reasoning: Case based reasoning (CBR) is an approach to solving new problems by recalling and reusing past experiences with solving similar problems. I have worked on various approaches for similarity assessment while generating a dialog to collect more information (i.e., a conversation), especially when experiences are recorded at varying levels of detail (or abstraction).
Artificial Intelligence Applications
I have transitioned my research on AI techniques to a variety of intelligent applications such a communicative agents, automatic schema mapping and brokering among websites, maritime activity and plan recognition, acronym extraction, and conversational diagnosis.
Enterprise Information Systems
I am generally interested in enterprise architectures, business process modelling, and process model learning.