سارا حاجی آقاجانی

Question-Answering Systems are powerful tools in the field of artificial intelligence that can automatically provide clearer answers to users' natural language questions. This research is concerned with the development of an advanced question-answering system for the Persian language that can provide accurate and comprehensive answers to a wide range of questions and topics. This system is composed of two main passage retrieval and reader components. In the passage retrieval part, a new dense passage retrieval method is used which is a pioneering approach in the field of Persian language. This method, which is implemented based on the BERT language model, outperforms traditional methods such as BM۲۵ in identifying relevant passages containing the answer to the user's question. The dense passage retrieval method effectively uses semantic and textual information to determine the most relevant passages, using BERT to ensure that the system retrieves very relevant passages for processing. The reader part extracts the answer from the retrieved passages using advanced language models such as BERT and RoBERTa. By fine-tuning these models on Persian language data, a powerful tool has been created that can understand the details of Persian passages and extract the most relevant and meaningful information. This question-answering system, using a combination of dense passage retrieval and advanced language models, is capable of providing accurate and comprehensive answers to a wide range of questions, including open and factual questions. The system's ability to understand the nuances of the Persian language and extract text information ensures that its responses are fully aligned with the user's intent and the provided passages. This innovative system, by combining dense passage retrieval and advanced language models, has the potential to improve the performance of a variety of natural language processing tasks. This system paves the way for a new era of intelligent question-answering in Persian and facilitates human-machine interaction. It also helps users to access the information they need more easily and efficiently.