Typicality-Based Collaborative Filtering Recommendation
Download as DOC, PDF1 like392 views
This document discusses a novel collaborative filtering recommendation method named Tyco, which incorporates object typicality from cognitive psychology to improve recommendation accuracy and reduce prediction errors. The proposed system outperforms traditional collaborative filtering methods, especially in data-sparse scenarios, by utilizing user typicality degrees to identify neighbors for recommendations. The Tyco approach demonstrates marked improvements over existing methods in both recommendation accuracy and efficiency.
Typicality-Based Collaborative Filtering Recommendation
- 1. TYPICALITY-BASED COLLABORATIVE FILTERING RECOMMENDATION ABSTRACT: Collaborative filtering (CF) is an important and popular technology for recommender systems. However, current CF methods suffer from such problems as data sparsity, recommendation inaccuracy and big-error in predictions. In this paper, we borrow ideas of object typicality from cognitive psychology and propose a novel typicality-based collaborative filtering recommendation method named TyCo. A distinct feature of typicality-based CF is that it finds ‘neighbours’ of users based on user typicality degrees in user groups (instead of the co-rated items of users, or common users of items, as in traditional CF). To the best of our knowledge, there has been no prior work on investigating CF recommendation by combining object typicality. TyCo outperforms many CF recommendation methods on recommendation accuracy (in terms of MAE) with
- 2. an improvement of at least 6.35% in Movielens Data set, especially with sparse training data (9.89% improvement on MAE) and has lower time cost than other CF methods. Further, it can obtain more accurate predictions with less number of big-error predictions. EXISTING SYSTEM: The basic idea of user-based CF approach is to find out a set of users who have similar favour patterns to a given user (i.e., ‘neighbours’ of the user) and recommend to the user those items that other users in the same set like, while the item-based CF approach aims to provide a user with the recommendation on an item based on the other items with high correlations (i.e., ‘neighbours’ of the item). In all collaborative filtering methods, it is a significant step to find users’ (or items’) neighbours, that is, a set of similar users (or items). Currently, almost all CF methods measure users’ similarity (or items’ similarity) based on co-rated items of users (or common users of items). Although
- 3. these recommendation methods are widely used in E-Commerce, a number of inadequacies have been identified. DISADVANTAGES OF EXISTING SYSTEM: It is difficult to find out correlations between users and items. Major issue that limits the quality of CF recommendations. Some predictions provided by current systems may be very different from the actual preferences or ratings given by users. These inaccurate predictions, especially the bigerror. PROPOSED SYSTEM: in this paper, we borrow the idea of object typicality from cognitive psychology and propose a typicalitybased CF recommendation approach named TyCo. The mechanism of typicality-based CF recommendation is as follows. First, we cluster all items into several item groups. For example, we can cluster all movies into ‘war movies,’ ‘romance movies,’ and so on. Second, we form a user group corresponding to each item
- 4. group (i.e., a set of users who like items of a particular item group), with all users having different typicality degrees in each of the user groups. Third, we build a user-typicality matrix and measure users’ similarities based on users’ typicality degrees in all user groups so as to select a set of ‘neighbours’ of each user. Then we predict the unknown rating of a user on an item based on the ratings of the ‘neighbours’ of at user on the item. ADVANTAGES OF PROPOSED SYSTEM: It generally improves the accuracy of predictions when compared with previous recommendation methods. It works well even with sparse training data sets, especially in data sets with sparse ratings for each item. It can reduce the number of big-error predictions.
- 5. SYSTEM CONFIGURATION:- HARDWARE REQUIREMENTS:- Processor - Pentium –IV Speed - 1.1 Ghz RAM - 512 MB(min) Hard Disk - 40 GB Key Board - Standard Windows Keyboard Mouse - Two or Three Button Mouse Monitor - LCD/LED SOFTWARE REQUIREMENTS: Operating system : Windows XP. Coding Language : .Net Data Base : SQL Server 2005 Tool : VISUAL STUDIO 2008.
- 6. REFERENCE: Yi Cai_, Ho-fung Leung†, Qing Li_, Huaqing Min_, Jie Tang_ and Juanzi Li_, “Typicality-based Collaborative Filtering Recommendation” IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 28, NO. 3, March 2014
- 7. REFERENCE: Yi Cai_, Ho-fung Leung†, Qing Li_, Huaqing Min_, Jie Tang_ and Juanzi Li_, “Typicality-based Collaborative Filtering Recommendation” IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 28, NO. 3, March 2014