Extraction of relevant information on a specific query from rapidly growing data is a concern for quiet time in order to scan and analyze data from all the related documents. Therefore, text summarization is paramount research area these days. It is about to find most relevant information from single or multi-documents. A reasonable amount of work is done in this area to overcome extensive searching and to reduce the time required. The knowledge-based and machine learning are the two methods for query-based text summarization where Machine learning approaches are mostly used for calculating probabilistic feature using Natural Language Processing (NLP) tools and techniques for both supervised and unsupervised learning. In the first part of this research work include to identify and analyze machine learning approaches for query-based text summarization for finding a useful summary for the users as specified by their need. In the second part, a comprehensive discussion is done to present the internal working mechanism of machine learning approaches for query-based text summarization.