Research relevant novel categorization of domain specific scientific articles using data mining techniques

Domain specific technical scientific articles are published in most of the international journals, conferences or workshops. Since these multiple articles are domain specific and application centric therefore have similarity in contents and the relevant information is also repeated. Thus reading these all related articles completely one by one to get the latest research developments in the interested domain is time-consuming, unnecessary, irrelevant, cumbersome and impossible. These problems are solved by developing an innovative solution called Optimized Summary System (OSS) which optimizes and summarizes these articles using various Data Mining strategies such as extraction and clustering which identifies `Research Relevant Novel´ (RRN) terms and then categorizes article contents into various RRN categories. `Maximal Marginal Relevance´ (MMR) criteria is used for optimization and reducing redundancies. OSS presents short, condensed, accurate and most relevant summarized contents from multiple relevant research articles for research scholars who are the regular users of innovative authentic information. The RRN categories also informs earlier and latest research developments, progress, and challenges and future scope in particular field of study providing starting material for further innovation.