A Survey On Side Information for Text Mining

Text Mining

Any text mining application may contain side information. This side information may be any links in the document, web logs which contain user access behavior, provenance information, the links for any document or any other non-textual attributes which are embedded into the text document. All these attributes may contain a huge amount of information for clustering purposes. But it is difficult to count the concerned importance of this side information especially when some of the data is noisy. In that matter, it is dangerous to merge side-information into the mining process because it can upgrade the quality of the representation for the mining process or can add noise in this system. Thus, there should be a right way to do this mining process so that it will make use of side information to maximize their advantages. Therefore, it is suggested to design an efficient algorithm which makes combination of classical portioning algorithm with probabilistic models in order to create an effective clustering approach. Afterwards, extension to the classification problem is also shown.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.