PhD Seminar: An FCA-based approach to informative feature extraction for query refinement in IR-chatbots

Мероприятие завершено
Speaker:  Elizaveta Goncharova, second-year PhD student, Faculty of Computer Science.
Where: Faculty of Computer Science, Pokrovskii bulvar 11, room R306.
When: Feburary 10, 18:10–19:30

Information retrieval (IR) chatbot is a special class of virtual assistants, which are widely used nowadays in customer support services. Our aim is to be able to identify item features that may interest the user while he is looking for a specific product using an IR-chatbot. During the first part of the presentation, we explain the proposed information-retrieval concept-based system, which operates as an IR-chatbot. The experimental results obtained using the real data will be presented and analyzed. In the second part, we consider the usage of Formal Concept Analysis (FCA) technique and Pattern Structures as its extension for processing unstructured data (objects with a text description), which has recently become a common way of presenting various items.