Author : Sonali Narendra Birajdar

As  the  exponential  explosion  of  various  contents  generated  on the Web, Recommendation techniques have become increasingly indispensable.  Innumerable  different  kinds  of  recommendations are made on the web every day, including movies, music, books, images,    books    recommendations,    query    suggestions,    tags recommendations, etc. No matter what types of data sources are used for the recommendations, essentially these data sources can be modelled in the form of various types of graphs. In this paper, aiming at providing a general framework on mining Web graphs for recommendations, we first propose a novel diffusion method which   propagates   similarities   between   different   nodes   and generates recommendations; then we illustrate how to generalize different  recommendation   problems  into  our  graph  diffusion framework.  The  proposed  framework  can  be  utilized  in  many recommendation tasks on the World Wide Web, including query suggestions,    tag    recommendations,    expert    finding,    image recommendations, image annotations, etc.

Keywords:  Recommendation,      Diffusion,      Query      Suggestion,      Image Recommendation