Abstract
This paper presents the development and implementation of an intelligent digital consultant (chatbot) designed to enhance the student experience at Al-Farabi Kazakh National University (KazNU). Universities face the challenge of providing accessible and comprehensive information to a diverse student body. Traditional methods can be inefficient and time consuming. To address this challenge, we developed a modular chatbot leveraging natural language processing (NLP) techniques with spaCy and Transformers, a locally deployed large language model (LLM) using Ollama for data security, and a PostgreSQL knowledge base with vector embeddings for semantic search. The paper details the system architecture, implementation workflow, key components, and optimization strategies. Performance is evaluated using metrics such as accuracy, response time, and user satisfaction. Throughout the paper, data analysis and visual representations, including tables and figures, are used to support our findings and illustrate key performance aspects. Results demonstrate the potential for improved student engagement, faster access to information, and enhanced overall satisfaction, showcasing the chatbot’s value as a valuable tool for the entire university community.