PUAA: Personal University AI Assistant using Retrieval Augmented Generation

A significant improvement in knowledge acquisition has been made by introducing Large Language Models (LLMs) like Chat-GPT and Bard, which provide a more efficient way to acquire information than the labor-intensive procedure of going through various online checkpoints. This new tendency in LLMs makes the widely used rule-based chatbots that colleges use look antiquated and inadequate. To better serve the needs of students looking for information about their universities, this research project suggests incorporating LLM technology into university websites. Our research introduces Personal University AI Assistant (PUAA) with methodology leveraging the RetrievalAugmented Generation (RAG) framework, utilizing the power of the LangChain in combination with cutting-edge LLMs like Mistral-7B, which is made available by the Hugging Face as an open-source option, and Chat GPT 3.5 as the most optimal solution for the problem of fast knowledge acquisition. PUAA improves the students’ experience of accessing university knowledge by providing instant, accurate information while reducing the workload of administrative staff and allowing them to focus on more complex inquiries and tasks. PUAA aims to promote the adoption of artificial intelligence (AI) in educational institutions, demonstrating the viability and benefits of advanced AI tools in enhancing the academic experience and information accessibility for students and staff.