Alternative approach to tree-structured web log representation and mining
Access Status
Authors
Date
2011Type
Metadata
Show full item recordCitation
Source Title
Source Conference
School
Remarks
© 2011 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Collection
Abstract
More recent approaches to web log data representation aim to capture the user navigational patterns with respect to the overall structure of the web site. One such representation is tree-structured log files which is the focus of this work. Most existing methods for analyzing such data are based on the use of frequent subtree mining techniques to extract frequent user activity and navigational paths. In this paper we evaluate the use of other standard data mining techniques enabled by a recently proposed structure preserving flat data representation for tree-structured data. The initially proposed framework was adjusted to better suit the web log mining task. Experimental evaluation is performed on two real world web log datasets and comparisons are made with an existing state-of-the art classifier for tree-structured data. The results show the great potential of the method in enabling the application of a wider range of data mining/analysis techniques to tree-structured web log data.
Related items
Showing items related by title, author, creator and subject.
-
Chang, Elizabeth; Tan, H.; Dillon, Tharam S.; Feng, L.; Hadzic, Fedja (2005)An XML enabled framework for representation of association rules in databases was first presented in [Feng03]. In Frequent Structure Mining (FSM), there are techniques proposed to mine frequent patterns from complex trees ...
-
Hadzic, Fedja; Dillon, Tharam S.; Chang, Elizabeth (2008)Tree-structured knowledge representations are increasingly being used since the relationships between data objects can be represented in a more meaningful way. A number of tree mining algorithms were developed for mining ...
-
Mohd Shaharanee, Izwan Nizal (2012)Deriving useful and interesting rules from a data mining system are essential and important tasks. Problems such as the discovery of random and coincidental patterns or patterns with no significant values, and the generation ...