WHAT: TWed Talk: "TMS-KG: Wrapping Whyis in a task-specific API"
WHO: Danielle Villa and Gia Santos
WHEN: 4p, Weds, 25 Feb 2026
WHERE: Winslow 1140
WEBEX: https://rensselaer.webex.com/meet/erickj4
EVENT PAGE: https://bit.ly/3ML5qjI
Please join us Weds, 25 Feb (4p) as Danielle Villa and Gia Santos lead us in a discussion of TMS-KG, the transactive memory system (knowledge graph) supporting the EFFECT project. Pizza et.al. arrives approx. 3:30p, the talk begins at 4p.
DESCRIPTION: Intelligence analysts use complicated systems to work on a variety of tasks such as reading documents, writing reports, and conversing with other analysts or AI agents. Tracking this information allows for post-hoc analysis of the user's path through the information, but is difficult to do with a traditional database. TMS-KG builds around Whyis to allow users unfamiliar with knowledge graphs to add information to the system. In this talk we will discuss our Transactive Memory System (Knowledge Graph), how it was designed to allow for easy tracking of user and AI actions, its current capabilities, and future plans.
Danielle Villa is a 4th year PhD student under Dr. Deborah McGuinness studying how semantic technologies can be used to improve LLMs, with a focus on improving explanation evaluation metrics. Her current work is on an ontology for describing and organizing measure for evaluating the faithfulness of AI explanations.
Gia Santos is a first-year PhD student at Rensselaer Polytechnic Institute, advised by Dr. Deborah McGuinness. Her research focuses on personal knowledge graphs, particularly in the healthcare domain, with an emphasis on ensuring data interoperability and privacy across heterogeneous information systems. In addition to contributing to the EFFECT project, she is exploring how large language models (LLMs) can support ontologists in ontology engineering. She also continues her work on the IoT Knowledge Graph project initiated during her time at the University of Maryland, Baltimore County (UMBC).
LINKS: