Year: 2025
Type: dissertation
Abstract: The paper considers the problem of automation of technical support in a bank using the Retrieval-Augmented Generation (RAG) architecture. The novelty of the study lies in the development of a system that combines a language model and a vector knowledge base for accurate generation of responses to queries. The purpose of the work is to increase the efficiency of support by implementing the RAG approach. To achieve this goal, methods of natural language processing, vector search, and integration of components into a single architecture were used.
Authors: Левченко, А. С., Levchenko, A. S.
Language: Russian
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7115038977