A Visual Search Engine for Searching Tamil Web Pages Using Web Community Mining and Natural Language Processing

With the growing Tamil interest and Internet, the amount of Tamil data doubles every 12-14 months and will increase even more dramatically in the coming year. With an enormous amount of Tamil data stored in web pages, it is increasingly important to develop powerful tools for analysis of such Tamil data and mining interesting patterns from it. There is a strong interest in employing methods of data mining to generate models of Tamil related web pages forming web communities. Web community refers collection of web pages of similar interest implicitly or explicitly. This paper proposes a new initiative for forming Tamil web communities with concise introduction about web community mining. The main intention of this paper is to employ web community mining technique for providing better results in search engines and to visualize the search engine results as Tamil web communities using a suitable visualization tool. This paper exploits visualization, web community mining and natural language processing (NLP) techniques. Visualization is the graphical presentation of information, with the goal of providing the viewer with a qualitative understanding of the information contents. This paper focuses on selecting the appropriate visualization tool best suited for displaying search engine results using visualization techniques. Various visualization techniques are also described in this paper. This community mining will yield benefits to all Tamil lovers,who want to be well-versed in a Tamil domain of his own interest. Tamil research publications and literatures in Tamil are grouped using bibliometric analysis. By forming people communities (i.e., people belonging to similar interest) using social network analysis,the domain knowledge in Tamil can be shared. Hence, web community mining may play an important role in forming Tamil Web Communities for gathering Tamil resources and documents of similar interest from the ocean of web very easily.