Web users usually suffer from the hitches of finding desirable and accurate information on the Web due to two problems of low precision and low recall caused by significant and rapid growth in the amount of information and the number of users. Thus there is a requirement to develop more proficient and effective techniques to satisfy the increasing demands of Web users, such as retrieving the desirable and related information, creating good quality Web communities, extracting the informative knowledge from the available information, capturing the underlying usage pattern from the Web observation data, recommending user customized information to offer better Internet service, and furthermore mining valuable business information from the common or individual customer navigational behaviour as well. we study different tools and techniques of Text Mining for IR. Main aim of this research work is to develop methods which will overcome the limitations of different Text Mining techniquesstudied and discussed above, for achieving accurate and efficient Information Retrieval process results.