摘要
随着电子商务和大数据技术的发展,在信息超载的情况下,如何提升营销的效率,更加全面的结合用户的喜好对产品进行推荐,成为关注的重点。针对上述的需求,结合用户的兴趣特征,提出一种基于隐式特征的用户兴趣协同过滤算法。对此,文章首先对隐语义模型进行分析,对用户特征和物品特征的提取;其次,通过协同过滤算法,计算两者之间的相似度;最后,通过预测的准确率来对上述构建算法的正确性和可行性的进行判定,验证了算法可提高预测的准确行,具有一定的工程借鉴价值。
with the development of e- commerce and big data technology, in information overload, how to improve marketing efficiency, more comprehensive combination of user preferences for product recommendation, has become the focus of attention. In view of the above requirements, combined with the characteristics of the user＇s interest, this paper proposes a collaborative filtering algorithm based on implicit features. This article first carries on the analysis to the latent semantic model; secondly, through the collaborative filtering algorithm to extract the user characteristics and features of objects, calculating the similarity between the two; finally, the prediction accuracy of the correctness and feasibility of the algorithm for constructing decision, verify the algorithm can improve the prediction accuracy. The project has a certain reference value.
出处
《自动化与仪器仪表》
2017年第9期201-202,205,共3页
Automation & Instrumentation
关键词
大数据
用户兴趣
协同过滤算法
相似度
准确率
big data
user interest
collaborative filtering algorithm
similarity
accuracy
