Enabling personalization recommendation with WeightedFP for text information retrieval based on user-focus

Personalization recommendation is a valid method for lightening the user´s burden on information retrieval. In order to implement personalization recommendation for text information retrieval (TR), user-focus is defined and algorithms for the construction of user-focus are given in this paper. The construction of user-focus for a user depends on his entire query requests at a period of time. Each query request of the user is treated as a transaction between TR system and him. Items in a transaction are non-noise words in the query request corresponding with the transaction. Each item is weighted with a value to measure the importance of the item for the user. Weighted frequent itemset is used for the user-focus´s construction. In order to mine weighted frequented itemset fast, an algorithm named as WeightedFP is presented. The experimental result shows that the implementation of personalization recommendation based on user-focus can lighten the user´s burden caused by the work of filtering valid information from vast information in some degree while time requirement of TR is satisfied well.