Wrappers for web access logs feature selection
Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics
| Hi-index | 0.00 |
Web log files store data related to the use of a website. Analyzing these data in detail is therefore crucial for improving the user browsing experience. However, usually Web log data are stored in flat files in different formats which hinders their analysis, thus obliging to use specific Web log analysis tools. In this context, approaches for structuring Web log data to better analyze them are highly required. To this aim, in this paper we propose a metamodel for Web log data in order to unify features of every available format at the same time that unnecessary technical details are omitted. This metamodel supports the design of Web log models regardless the format of the Web log files. A set of guidelines have been also proposed to define a multidimensional schema of a data warehouse from our Web log model to be used in advanced analysis tools such as OLAP (OnLine Analytical Processing) or data mining tools in order to enhance the analysis of Web usage by means of log data