Image Similarities and Duplication in Web based Searching
Search engines are used worldwide to get information related to any topic at any time. Whenever user is searching for images then there is possibility of getting ambiguous results. Ambiguity is nothing but fact which is having more than one possible meaning and therefore possibly causing confusion. Many researchers are working on this type of ambiguities and provided best re-rank results. But many times it has been observed that search engine provides same image as output multiple times. Which unnecessarily occupy system space and also it will be tedious task for the user to pick any one from these duplicate results. This paper majorly focuses on the difference between similarity and duplication, and how this duplication can be avoided to improve system performance.
Keywords - Intent, Similarity, Duplication, Re-ranking, Semantic signature, Relevance feedback