Improved Log Miner for Frequent Items Generation

Rajdeep Marathe, Mrs. Dhanashree Phalke

Abstract Web log mining is the newest technology of data mining. There are various web related activities that are taken into consideration. Such data are mostly structured in nature as they are collected from various web pages and other web logs that are maintained in the server. Web Mining is divided into three types web content mining, web usage mining and web structure mining. In case of Web usage minings, the main aim and area is to focus on Web users and to learn the way they interacts with various Web sites available. As web log data are mostly noisy and extremely ambiguous, still there is a way where we can discover useful information and structure in the way the users interacts with a web site. The main objective of using mining is to quickly and automatically identify users from the vast log data. We can identify information such as frequent access paths, frequent access page groups and then cluster the users. With the help of web usage mining algorithms, the web application server logs, registration information, the user interest and other data such as user access patterns can be mined which will be helpful in laying foundation for decision making of organizations. Keywords: Web Mining, candidate sets, framing, Improved Apriori algorithm, AprioriAll algorithm, E-Web Miner algorithm, Web Log Analysis

Title:	Improved Log Miner for Frequent Items Generation
Author Name:	Rajdeep Marathe, Mrs. Dhanashree Phalke
Abstract:	Abstract Web log mining is the newest technology of data mining. There are various web related activities that are taken into consideration. Such data are mostly structured in nature as they are collected from various web pages and other web logs that are maintained in the server. Web Mining is divided into three types web content mining, web usage mining and web structure mining. In case of Web usage minings, the main aim and area is to focus on Web users and to learn the way they interacts with various Web sites available. As web log data are mostly noisy and extremely ambiguous, still there is a way where we can discover useful information and structure in the way the users interacts with a web site. The main objective of using mining is to quickly and automatically identify users from the vast log data. We can identify information such as frequent access paths, frequent access page groups and then cluster the users. With the help of web usage mining algorithms, the web application server logs, registration information, the user interest and other data such as user access patterns can be mined which will be helpful in laying foundation for decision making of organizations. Keywords: Web Mining, candidate sets, framing, Improved Apriori algorithm, AprioriAll algorithm, E-Web Miner algorithm, Web Log Analysis
Cite this article:	Rajdeep Marathe, Mrs. Dhanashree Phalke , " Improved Log Miner for Frequent Items Generation " , International Journal of Emerging Trends & Technology in Computer Science (IJETTCS), Volume 4, Issue 3, May - June 2015 , pp. 230-233 , ISSN 2278-6856.
Full Text [PDF] Home