用户需求网络信息的优先协同过滤推荐仿真  

User-Required Network Information Prioritized Collaborative Filtering Recommendation Simulation

作　　者：耿晓利[1] 邓添文 GENG Xiao-li;DENG Tian-wen(South China Institute of Software Engineering,Guangzhou University,GuangzhouGuangdong 510990,China)

机构地区：[1]广州大学华软软件学院

出　　处：《计算机仿真》2019年第11期352-355,共4页Computer Simulation

基　　金：广东省2017重点平台及科研项目(2017GXJK257);2017年广州大学华软软件学院教学研究科学研究项目(ky201724)

摘　　要：针对传统用户需求网络信息的优先协同过滤推荐方法存在推荐准确度较低、推荐范围小等问题,提出基于相似性度量的用户需求网络信息优先协同过滤推荐方法。根据目标网络用户对网络信息需求的感兴趣程度对用户进行分组,引用贝叶斯算法计算用户具有不同特征时对网络信息需求的喜好程度。在项目近似度公式中引入用户对项目的喜好程度概率值计算项目间的相似度,得到最高相似度邻居项目的评分。再利用云模型计算用户的相似度得到邻近用户,得到最终的预测评分,将预测评分值较高的前几个项目优先推荐给用户,完成推荐。仿真结果表明,相比传统的协同过滤推荐方法,所提方法可在一定程度上提高推荐准确度和推荐覆盖率,能够更准确的为用户提供所需要的网络信息。Traditional collaborative filtering recommendation methods have some problems such as low recommendation accuracy and small recommendation range. Therefore, a preferential collaborative filtering recommendation method for user demand network information based on similarity measure was proposed. According to the interest degree of target network user to the network information demand, user was put into groups. Then, Bayesian algorithm was used to calculate the preference for the network information demand when user had different features. In the item approximation formula, the probability value of preference for item was introduced to calculate the similarity between items, and then score of neighbor item with highest similarity was obtained. Moreover, the cloud model was used to calculate the similarity of user and thus to obtain the neighboring user. After that, the final predicted score was obtained. Finally, the first few items with higher predicted values were preferentially recommended to users. Thus, the recommendation was completed. Simulation results show that, compared with the traditional collaborative filtering recommendation method, the proposed method can improve the recommendation accuracy and recommended coverage rate. Meanwhile, this method can provide users with the required network information more accurately.

关 键 词：用户需求 网络信息 优先协同过滤推荐 

分 类 号：TM715[电气工程—电力系统及自动化]