Personalized Tag based Image based Search Engines using clustering and similarity Indexing

Tejwant Telang
2018 International Journal for Research in Applied Science and Engineering Technology  
Nowadays, search engine plays a vital role in common human being. In our proposed system, we propose image based personalized search engine in which the images will be searched according to specified query as well as user's profile and preferences. The searched images will be categorized into different communities. The communities will be detected with the help of Affinity propagation clustering algorithm. Images within communities will be re-ranked with the help of user's profile/ preferences/
more » ... google distance. The communities will be re-ranked on the basis of their cosine similarity coefficient. Our system will calculate cosine similarity coefficient of community text and query specified by the user. To demonstrate our work we will use google online dataset.
doi:10.22214/ijraset.2018.3380 fatcat:qa7ppangrbfbjptyrxplecmody