Optimal rough fuzzy clustering for user profile ontology based web page recommendation analysis

Personalized information recommendation in view of social labeling is a hot issue in the scholarly community and this web page data collected from the Internet of Things (IoT). To accomplish personalized web pages, the current investigation proposes a recommendation framework with two methodologies on user access behavior using Rough-Fuzzy Clustering (RFC) technique. In this paper, Fuzzy-based Web Page Recommendation (WPR) framework is provided with the user profile and ontology design. At first, the weblog documents were gathered from IoT to clean the data and undergo learning process. In the profile ontology module, the learner profile was spared as the ontology with an obvious structure and data. For identification of the similar data, innovative similarity measure was considered and for effective WPR process, the generated rules in RFC were optimized with the help of Chicken Swarm Optimization (CSO) technique. Finally, these optimal rules-based output recommends e-commence shopping websites with better performances. A group of randomly-selected users was isolated and on the basis of the obtained data, their clustering was performed by cluster analysis. Based on the current proposed model, the results were analyzed with performance measures and a number of top recommended pages were provided to users compared to existing clustering tech-niques.