Improving personalized image retrieval from the photo sharing social networks

Data mining, or Knowledge discovery, is the process of digging through and analyzing enormous sets of data and then extracting the meaning of the data. Photo sharing websites helps in publishing or transferring user's digital photos online, thus enabling the user to share them with others publicly or privately. This function is provided through both websites and applications that facilitate the upload and display of images. Nowadays, there has been an increase in the photo sharing websites which allows users not only to create, share, annotate and comment multimedia contents, but also provide useful information to improve media retrieval and management. But, most of the photo sharing websites used today, results in poor user experience (i.e.) when given a query, the searchers did not find any relevant results. To overcome this problem, a framework is proposed that contains the following three components 1) a ranking-based multicorrelation tensor factorization model (RMTF), 2) user-specific topic modeling,3)Latent Semantic Indexing. In this paper, a proposed framework is used to handle the complex multiple words-based queries and simultaneously considering the user and query relevance for personalized image search By using the above method, the image retrieved results in what the user's had expected.