Personalized recommendation based on item dependency map

In data mining, one wants to find hidden knowledge, unexpected patterns and new rules from massive data. In this paper, the authors intend to find user´s item purchasing pattern and recommend good that he/she wants. So they suggest an item dependency map which express the relation between purchased items. Using an algorithm that they suggest, one can recommend an item which a user has not bought yet, but is likely to interested in. The item dependency map is used as a pattern for association in a Hopfield network so one can extract a users´ global purchasing item pattern only using a users´ partial information. The Hopfield network is an iterative auto-associative network consisting of a single layer of fully connected processing elements which can function as an associative memory. A Hopfield network can extract global information from sub-information. Therefore, this algorithm obtains an advantage of Hopfield networks. The authors´ algorithm can be applied to real web sites and help web masters to know users´ taste