High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Prediction Model on Web Mining

Author(s):

Pratik J. Bhise , tulsiramji gaikwad patil college of engg & technology; Mithun Funde, tulsiramji gaikwad patil college of engg & technology; Prajakta Ramteke, tulsiramji gaikwad patil college of engg & technology

Keywords:

Web Usage Mining, Markov models, Bayesian theorem, Prediction

Abstract

Because of the rapid growth of the word-wide-web, the problem of predicting a user’s browsing behavior on a web-site has gained important and the need is personalize and influence a user’s browsing experience. For solving this problem Markov models and its variations have been found. Personalization, building proper websites, promotion, getting marketing information, and forecasting market trends etc, the prediction result can be used. By using Markov model the users’ browsing behaviors can be predicted at category level. Probability of present and infer users’ browsing behaviors at webpage level by applying the Bayesian theorem. The system can effectively filter the possible category of the websites by Markov models and predict websites accuracy by the Bayesian theorem. In this paper we present different techniques for prediction of users’ browsing behaviors and intelligently selecting parts of different order Markov model. The resulting model has improved prediction accuracy and reduced state complexity.

Other Details

Paper ID: IJSRDV4I10837
Published in: Volume : 4, Issue : 1
Publication Date: 01/04/2016
Page(s): 739-743

Article Preview

Download Article