A Study of Big Data Mining: A New Approach and Handling Techniques
Today web is producing ever growing and huge amounts of data and information. This extremely huge amount of data called big data are in terms of bulk quantity, complexity, denotation, circulation, and processing costs in computer science & information technology, web-based processing, cloud computing, and computational intelligence. To provide the ability to make sense and maximum utilization of such vast amounts of web data for knowledge discovery and decision-making is crucial to scientific advancement; we need new tools for such a big web data mining. The web data are in the form of structured and unstructured type which is directly or indirectly influencing society, peoples or researchers. Design and implementation of a web mining research support system has become a challenge for people with interest in utilizing information from the big web data for their research. This paper presents a new prototype tool for extracting information from big data across different web sites. Our prototype tool (advance model) uses a new approach for pattern finding from new web pages across different sites. It does so by focusing on the different link present in the seed Web sites and exploring and saving the links to find new pattern.
Keywords - Seed URL, HAM, Web Mining, Web content Mining, Mining task.