An Approach to Enhance the Quality of Recommendation Using Collaborative Tagging

Collaborative labeling portrays the process by which numerous users put in metadata in the form of keywords to shared data. Nowadays, collaborative labeling has grown in reputation on the web, on sites that permit users to label bookmarks, photographs and other details. It has been recently become useful and well known as one effective way of classifying items for future search, sharing information, and filtering. So, as to predict the future search of users, we propose a novel collaborative tagging-based page recommendation algorithm using fuzzy classifier. The method consists of three phases: Grouping, Rule Generation Phase and Page Recommendation Phase. In the proposed method, we calculate the resemblance of users in selecting tags and thereby, calculate the nearest neighbors of each user and cluster them. Then, the priority of tags and items for each user is calculated for constructing a Nominal Label Matrix and Nominal Page Matrix. Finally, the fuzzy rules are generated for page recommendation. The experimentation is carried out on delicious datasets and the experimental results ensured that the proposed algorithm has achieved the maximum hit ratio of 6.6% for neighborhood size of 20, which is higher than the existing technique which obtained only 5.5%.