微型电脑应用

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 advantage is 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 improved algorithm-based on a combination of average differential recommendation algorithm, the combination algorithm to some extent alleviate the problem of the original algorithm, to improve the quality of the recommended system.