Proposing a classification algorithm for user identification according to user web log analysis

S. Peyvandi, H.M. Shirazi, A. Faraahi

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Request classification is one of important strategies of Web. User behavior analysis can make Web service more intelligent and secure. Recently, tree structures have become a popular way for storing and manipulating huge amount of data. The classification of these data can facilitate storage, retrieval, indexing, query answering and different processing operations. In this paper, we propose User- Classifier algorithm for rule based classification of tree structured data that to classify web log user. This algorithm is based on extracting special tree pattern from training dataset. Our experiments show that User- Classifier reduces running time. In the case of complete classification, User- Classifier shows the best classification quality.

Original languageEnglish
Pages (from-to)645-652
Number of pages8
JournalAustralian Journal of Basic and Applied Sciences
Volume5
Issue number9
Publication statusPublished - Sep 2011
Externally publishedYes

Keywords

  • Induce closed tree pattern
  • Rule based classification
  • Structural classification
  • Tree-structured data

Fingerprint Dive into the research topics of 'Proposing a classification algorithm for user identification according to user web log analysis'. Together they form a unique fingerprint.

Cite this