A Multi-Source Information Retrieval System for Question Answering with Sentiment Analysis and LLM Interpretation

In the era of ubiquitous digital information, question-answering systems have become indispensable tools for accessing and extracting knowledge from vast datasets. This research explores the design, implementation, and evaluation of a question-answering system capable of processing user queries, retrieving information from diverse sources, and generating accurate responses. The study focuses on assessing system performance under three operational modes: offline alone, online alone, and combined (offline and online). A meticulously curated dataset comprising diverse user queries and corresponding ground truth answers forms the basis of the experimental evaluation. Performance metrics including response time, user satisfaction, and coverage are meticulously evaluated to gain insights into system functionality and efficacy across different operational paradigms. The findings highlight the significance of leveraging both offline and online data sources to optimize system performance, underscoring the importance of a balanced approach in designing intelligent question-answering systems. This research contributes to advancing the state-of-the-art in question-answering technology and lays the foundation for future developments aimed at enhancing information retrieval and knowledge extraction capabilities in various domains.