The term legal research generally refers to the process of identifying and retrieving appropriate information necessary to support legal decision making from past case records. At present, the process is mostly manual, but some traditional technologies such as keyword searching are commonly used to speed the process up. But a keyword search is not a comprehensive search to cater to the requirements of legal research as the search result includes too many false hits in terms of irrelevant case records. Hence the present generic tools cannot be used to automate legal research. This paper presents a framework which was developed by combining several Text Mining techniques to automate the process overcoming the difficulties in the existing methods. Further, the research also identifies the possible enhancements that could be done to enhance the effectiveness of the framework.