Adaptive Semantic Search: Re-Ranking of Search Results Based on Webpage Feature Extraction and Implicitly Learned Knowledge of User Interests

The quest for information in the contemporary world ends at search engines that crawl millions of web pages on the World Wide Web and it is clearly essential that the results should be ranked in an order that would best fit the user interests. This paper proposes a method of re-ranking the search results that have been primarily ranked using either conventional algorithms that use link structure and user clicks or semantic algorithms, using a combination of general webpage features and user interests. The features of web pages like images, videos etc., are extracted by crawling them and the user´s general interest in those features are learnt from past queries made and clicks on particular results. Using the degree to which each feature is present and the corresponding interest of the user, the user´s interest in a particular search result is predicted and consequently the results are re-ranked in such a way that it augments the efficiency and effectiveness of conventional intent / meaning driven semantic search concept.