International Association of Computer Science & Information Technology (IACSIT)
Application of data mining technique to the World Wide Web refers to as Web mining. Web based origination collects large volume of data for their operation. Analysis of such data can help the organization for better working (marketing strategy, services, evaluation of effectiveness, promotional campaigns etc.). This type of analysis require discovery of meaningful relationships from the large collection of primarily unstructured data stored in Web server access logs. The authors propose a new approach for automatically learning (context-free) grammar rules form server access log text (positive set) samples, based on the alignments between the sentences.