24-07-2012, 02:46 PM
Mining Web Graphs for Recommendations
Web Graphs .pptx (Size: 252.23 KB / Downloads: 152)
Introduction
There is a huge explosion of various contents generated on the Web in the present run and recommendation techniques have become increasingly indispensable to select the best one.
Innumerable different kinds of recommendations are made on the Web every day, including images recommendations, query suggestions, etc. which can be modeled in the form of graphs
Existing System
In present mining system, we have different recommendation algorithms for different recommendation tasks.
But actually, most of these recommendation problems have some common features, where a general framework is needed to unify the recommendation tasks on the Web.
Moreover, most of the existing methods are complicated and require tuning a large number of parameters.
Disadvantages
It is becoming increasingly harder to find the relevant content and also which is recommended by the users.
Designing different recommendation algorithms for different recommendation tasks is tedious and inefficient due to their similar implementation.
Proposed System
Recommender Systems, a technique that automatically predicts the interest of an active user by collecting rating information from other similar users or items is used.
The proposed method consists of two stages: generating candidate queries and determining “generalization/specialization” relations between these queries in a hierarchy.
Advantages
It is a general method, which can be utilized to many recommendation tasks on the Web.
It can provide latent semantically relevant results to the original information needed.
The designed recommendation algorithm is scalable to very large datasets