Keyphrase Extraction using supervise learning

Abstract: Text mining is knowledge intensive process in which a user communicates with a collection of documents. Text categorization is a kind of “supervised” learning where the categories are known beforehand and determined in advance for each training document. Manually extraction of keywords is slow, expensive and tedious. Therefore automatic keyword extraction is necessary. Keyphrases help the users to get idea about the content of document. Kea- means clustering used for extracting test document from large quantity of text data. In Kea-means algorithm, documents are clustered into several groups like K-means, but the number of clusters is determined automatically by using the extracted keyphrases. Set of training documents and machine learning is used to determine phrases are keyphrase or not.

Keywords: Text mining, Keyphrases, clustering, supervise learning.