Online Image Retrieval Using Query Clustering Algorithm

An online image retrieval system (similar to Google image search) where users search for images by submitting queries that are made of keywords. The queries formed by the users of a search engine are semantically refined, the keywords representing concise semantics. The aim is to improve user satisfaction by returning images that have a higher probability to be accepted (downloaded) by the user. The assumption is that the users search for images by issuing queries, each query being an ordered set of keywords. The system responds with a list of images. The user can download or ignore the returned images and issue a new query instead. As the user’s issues queries and pick images the system annotates the images in an automatic manner and at the same time establishes relevance relations between the keywords. The new method is shown to posses the Query clustering algorithm using agglomerative method for discovering similar queries on a search engine to retrieve the similar images. Experimental result signifies the query clustering to show accuracy. This algorithm gives the precision and recall values, which are helpful in determine the efficiency of search engine queries.