Question-answering (Q&A) systems are essential for quickly and effectively sifting through massive volumes of data to locate pertinent information and respond succinctly to user queries. Rapid advancements in machine learning techniques have given rise to a variety of strategies for improving Q&A system performance. This paper provides a thorough comparative review of the various machine learning techniques applied to responsive systems. The study aims to assess and compare the effectiveness, advantages, disadvantages, and practical applications of different tactics to offer guidance to researchers and industry professionals. It offers researchers and practitioners direction so they can select and implement suitable strategies based on needs and constraints. To enhance the functionality and accuracy of the responses in Q&A systems for both structured and unstructured data, the paper emphasizes some recent advances.