基于平均差分的协同过滤推荐算法分析与研究

摘要

目前,协同过滤技术是个性化推荐系统中广泛使用的一种技术,该技术最大的优点是对推荐对象没有特殊的要求,能够处理非结构化的复杂对象,然而算法中普遍存在的数据稀疏性、可扩展性问题影响了算法的推荐效果.本文在分析了原有算法的基础上,提出了一个改进了的算法——基于平均差分的组合推荐算法,这个组合算法在一定程度上缓解了原有算法的问题,提高了推荐系统的质量.%The collaborative filtering technology is the personalized recommendation system is a widely used technology nowadays, the biggest advantageis the recommended object without special requirements, can handle unstructured complex objects, however, prevalent algorithm the data sparseness ,scalability issues affecting the algorithm recommended results. Based on the analysis of the original algorithm is proposed based on an improvedalgorithm -based on a combination of average differential recommendation algorithm, the combination algorithm to some extent alleviate the problem ofthe original algorithm, to improve the quality of the recommended system.