Pattern based topic model for data mining
Balaji Subhash Jadhav, D. S. Bhosale, Dipali Jadhav · 2016
Recently previous years many term based & pattern based approaches used for exacting user needed information from collection of documents. Topic modeling such as Latent Dirichlet Allocation (LDA) was proposed to generate multiple topics from collection of documents. Topic modeling has been used in the area of text mining & machine learning. The traditional word-based & term-based topic representations May not be able semantically represented document but pattern are always better than single terms for representing documents. In this paper proposed the topic modeling with pattern mining techniques to generate pattern based topic models. This Pattern Based Topic Model (PBTM) utilized for more enhanced representation of word based topic model. Utilizing PBTM user interest can be represent with multiple topics & each topic represented with semantically patterns.