Supervised Learning Techniques In Web Usage Mining: Comparison And Analysis

@article{1700059,
      author = {Parth Suthar},
      title = {Supervised Learning Techniques In Web Usage Mining: Comparison And Analysis},
      journal = {Iconic Research And Engineering Journals},
      year = {2017},
      volume = {1},
      number = {3},
      pages = {12-18},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1700059.pdf},
      abstract = {The World Wide Web contains enormous amount of information about almost every imaginable subject. Web mining is a special area of data mining, which deals with identifying interesting patterns and useful information from the web. This knowledge is useful in improving the quality of services provided by web. The information about user access is stored in the form of web access logs at web servers and proxies. Web usage mining is a discipline that deals with extracting users? information regarding user interests and behaviour profiles by processing those web access logs. This knowledge is useful in improving the areas like web personalization, recommendation systems, business intelligence, market segmentation etc. In this paper, we review, analyse and compare various existing supervised learning techniques utilized for web usage mining. We also present methods to compare and test the performance of those techniques},
      keywords = {supervised learning, web usage mining, web access logs, classification, KNN, SVM, Na?ve Bayes classifier, decision tree classifier, rule based classifier, k-fold cross validation, bootstrapping, confusion matrix},
      month = {September},
  }