Paper Details

Nowadays, a huge amount of data is present on web, so to extract useful knowledge and to manage thosehuge files will become mandatory to obtain fruitful business analysis results. To extract useful knowledge from WorldWide Web ( WWW) is known as web mining. Web usage mining has emerging trends on network traffic control and flowanalysis, website management, personalization, etc. Neural network have capability of self organization and is alsomatched with ant colony behavior and adaptive learning. Such concept is used for information retrieval from huge webdata. It is also used for complex classification, optimization and distributed control problems [1].With the help of NeuralNetwork algorithms for classification of web log data; it produces the best result of classification. So, in this paper wehave introduced solutions for self-organizing and growing network which helps in information retrieval from huge webdata and also discussed various neural network algorithms i.e. GNG (Growing Natural Gas), ART(Adaptive ResonanceTheory) model, LVQ(Learning Vector Quantization) and its series. Input for neural algorithms is web log files andexpected outcome would be optimal representation of network that is further used for Information extraction in webusage mining. Such trained network is used for classification which gives effective classification of data.

Disha Patel, Shraddha Joshi, “NEURAL NETWORK CLASSIFICATION ALGORITHMS FOR WEB USAGE MINING AND PROPOSED SOLUTIONS FOR HUGE WEB DATA CLASSIFICATION”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 1, Issue 11, pp. 294-300, November 2014.