Illustrate the Mining Techniques

Text mining is a cross between IR and AI. IR has gone through several generations of development. In the 1970s, computational techniques based on inverted indexes and vector spaces were developed and tested on computer systems. Also, probabilistic retrieval methods based on Bayesian statistics were developed. Although more than 30 years old, this still forms the basis of modern IR systems. In the 1980's, coinciding with the developments of new AI techniques, knowledge-based and expert systems that aim to emulate domain specialists were developed. User modeling and natural language processing (NLP) techniques were developed to assist in representing users and documents. These were applied to improve online searching.