Efficient Personalized Web Mining: Utilizing The Most Utilized Data

TL;DR

This paper proposes a personalized web mining approach that ranks links based on user interaction data, such as clicks and time spent, to improve information retrieval accuracy.

Contribution

It introduces a method to incorporate both interested and uninterested links with weighted rankings based on user behavior for personalized web mining.

Findings

Enhanced link ranking accuracy

Improved user satisfaction in search results

Effective utilization of user interaction data

Abstract

Looking into the growth of information in the web it is a very tedious process of getting the exact information the user is looking for. Many search engines generate user profile related data listing. This paper involves one such process where the rating is given to the link that the user is clicking on. Rather than avoiding the uninterested links both interested links and the uninterested links are listed. But sorted according to the weightings given to each link by the number of visit made by the particular user and the amount of time spent on the particular link.