As the internet services are increasing day-byday
various websites are working to find new techniques for
webpage recommendations. Researchers have developed
new methods for increasing the accuracy of predicted web
pages. This paper has utilized web content feature of the
website pages for developing the Term network where terms
from each page help in specifying the relations between all
pages. Web log is another feature used in this work where
Markov model of third order helps in improving the
prediction accuracy. Experiments are performed on
different dataset sizes with these feature combinations. It is
observed that proposed model is better as compared to
previous works. Results show that the use of Markov Model
with web content feature is better for prediction.
Published In:IJCAT Journal Volume 3, Issue 11
Date of Publication : November 2016
Pages : 510-514
Figures :02
Tables : 04
Neetu Sahu : Atal Bihari Vajpayee Hindi Vishwavidyalaya
Bhopal, M.P., India
Pragyesh Kumar Agrawal : Institute for Excellence in Higher Education
Bhopal, M.P., India
Information Extraction, Text Analysis, Feature
Extraction, Text Categorization, Clustering
Internet has become the need of the modern world by
providing lots of services and tools. So to increase its
efficiency is primary requirement of the researchers. This
paper has contributed the page prediction work by
utilizing the weblog and web content features. Here web
content is used for developing the relation between the
terms in form of term network. In similar fashion weblog
is used to find FWAP. It can be concluded from the tables
presented in the result section that the proposed model
provides better results as compared to previous models on
different evaluation parameters. This work will be carried
forward to increase the efficiency by using other features.