A Study on LVQ Based Switching Hybrid Movie Recommendation

• Main Authors: Chung-Yu Lin, 林重佑
• Other Authors: Chuen-Min Huang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/11873132375795364614

碩士 === 國立雲林科技大學 === 資訊管理系碩士班 === 100 === The great development of Internet technology brings more and more people to use computers to extract abundant content from this platform by their high speed computing ability. To keep the most valued consumers, most corporations have launched to the electronic environment in order to provide personalized services to their consumers. Content-based filtering and collaborative filtering are widely used techniques in recommendation system. The former method analyzes used records from users to make recommendation. The latter one takes the advantage of user preferences to recommend suitable products. Although they can offer proper recommendations, some shortcomings are existed individually. Thus, the hybrid recommendation technique combines the above advantages to recommend content corresponded with users’ requirements. Recently, hybrid recommendation technique is affected by neural network’s learning ability. A lot of supervised neural networks are combined with hybrid recommendation. Previous studies adopted three layers or multiple layers to construct recommendation. Their drawbacks are slow convergence and hard to design. In this paper, we present a novel switching hybrid recommendation framework based on Learning Vector Quantization (LVQ) and collaborative filtering to provide personalized recommendation. Our approach applies the two-layer architecture in LVQ and collaborative filtering to build switching hybrid recommendation. MovieLens data set is used to test our framework. Results show that switching hybrid strategy provides promising personalized recommendation. Our experiment gains 79% of precision, and the recall rate also reaches 82%.