Large Language Model Powered In-House Question Answering Assistant

Large language models are widely used in many natural language processing applications today. Automatic question answering is one of the areas where language models are frequently used. In this study, a question answering system was developed that allows Adesso Türkiye employees to access internal company information quickly and accurately. A Retrieval Augmented Generation (RAG)-based question answering structure was created by giving content prepared by experts in the field of human resources and information security as input to the large language model. As a result of the experiments, high accuracy results were obtained on datasets. In future studies, it is aimed to increase performance by using different language models and to make the developed system available to employees by integrating it into MS Teams.