Towards an Automatic Characterization of Criteria

摘要: The number of documents is growing exponentially with the rapid expansion Web. new challenge for Internet users now to rapidly find appropriate data their requests. Thus information retrieval, automatic classification and detection opinions appear as major issues in our society. Many efficient tools have already been proposed ease search over web support them choices. Nowadays, would like genuine decision that efficiently when focusing on relevant according specific criteria area interest. In this paper, we propose a approach characterization such criteria. We bring out able automatically build lexicon each criterion. then show how can be useful or segmentation tasks. Experiments carried real datasets efficiency proposal.