Text Mining Using Metadata for Generation of Side Information

Abstract

Text Mining is knowledge discovery process from large database to find out unknown patterns. In many metadata based text mining applications, side information also known as metadata which is associated with the text document. There are different types side information containing large amount of data i.e. metadata, weblogs and non-textual data (image, video, etc.). The side information is difficult to estimate when it contains noisy data. To achieve this, there is scope of improvement in generating side information i.e. selecting efficient classification and clustering algorithms, providing security for clustered side information, document organization, exploring filtering approaches. In future, there is a scope to design an extended approach for clustering using classical partitioning and probabilistic model.