Создание интеллектуального ассистента-помощника на основе RAG-системы

Abstract

This paper presents the development of an intelligent assistance system aimed at users, utilizing the Retrieval-Augmented Generation (RAG) model. The proposed project aims to provide personalized assistance in a specific domain, based on a knowledge base. By integrating RAG, the model improves the quality of responses by incorporating relevant information from external sources. The software implementation of the project has been carried out using the DeepSeek AI API. Key words: Retrieval Augmented Generation, Large Language Model, neural networks.