共查询到19条相似文献,搜索用时 46 毫秒
1.
2.
3.
石京京;于敬;赛娜 《电子技术与软件工程》2022,(14):183-187
本文提出了一种融合时序和信任值矩阵的协同过滤推荐算法。通过使用信任值矩阵替代了用户-用户相似度矩阵,一定程度上缓解了数据稀疏对推荐效果的负面影响。其次该算法融合了用户和物品交互行为的时序信息,构建用户及物品的网络关系图。最后利用矩阵分解技术预测推荐评分,生成最终的推荐列表。通过在公开数据集MovieLens-1M上进行实验,结果表明,本文方法在RMSE指标上具有显著的推荐效果提升。 相似文献
4.
5.
6.
学生成绩预测及告警是协助高等院校职能部门人员管理学生学习情况和监测教师教学质量的有效方法.通过提前预测学生考试分数,对预测成绩偏低的学生加强管理,提前重点关注该类学生的学习情况,可以降低学生考试不及格的风险.文章利用矩阵分解算法提取学生历史分数中的关键特征值,并将其用于学生成绩预测.实验结果表明,使用改进的奇异值分解方... 相似文献
7.
8.
推荐系统已成为电子商务企业吸引客户、实现盈利的有效技术支持,它能够根据用户的网络点击数据预测其偏好,做出个性化推荐。研究了一个基于动态矩阵分解模型的NETFLIX电影推荐系统。该系统通过训练一个来自NETFLIX平台、包含9 000部电影历史评分的数据集进行预测评分。核心算法包括运用矩阵分解(Matrix Factorization, MF)建立有效的数据处理模型,以及使用随机梯度下降(Stochastic Gradient Descent, SGD)训练该模型。数据集采用稀疏矩阵存储,以节省空间。在训练过程中,对预测评分增加了特定的偏向值。该系统与市场同类产品相比拥有更高的预测准确度,并向电影观众推荐符合他们喜好的电影,能极大地提高电影观看票房值。 相似文献
9.
传统矩阵分解算法和基于用户画像的算法存在数据稀疏性和冷启动等问题,且多数情况下只注重于用户项目交互数据,而对用户本身的属性信息缺少借鉴,从而导致推荐准确性不高.将K-means与矩阵分解相结合,提出了一种基于K-means的矩阵分解推荐算法(Matrix Decomposition Based on K-means,K... 相似文献
10.
11.
12.
本文针对评分预测中用户评分主观性及评分数据稀疏带来的预测不准确问题,围绕社交推荐的特点,设计实现一种社交网络评分预测方法.首先,针对评分主观性问题,引入并优化相关云模型理论,提出采用综合云模型生成评分标准并转化用户评分的方法.其次,针对预测不准确问题,通过引入隶属度达到数据降维和目标用户定位的作用,同时考虑到社交网络用户关系对评分结果的影响,分别利用社交关系及相似群体建立两个评分预测模型,并基于高斯变换融合两部分预测结果生成预测评分.实验表明,该方案不仅克服了用户评分主观性,同时有效改善了用户评分数据稀疏情况下传统预测方法准确度偏差的问题. 相似文献
13.
14.
15.
Social tagging is one of the most important characteristics of Web 2.0 services, and social tagging systems (STS) are becoming more and more popular for users to annotate, organize and share items on the Web. Moreover, online social network has been incorporated into social tagging systems. As more and more users tend to interact with real friends on the Web, personalized user recommendation service provided in social tagging systems is very appealing. In this paper, we propose a personalized user recommendation method, and our method handles not only the users’ interest networks, but also the social network information. We empirically show that our method outperforms a state-of-the-art method on real dataset from Last.fm dataset and Douban. 相似文献
16.
融合显/隐式信任关系的社会化协同过滤算法TrustSVD在推荐系统中有广泛的应用,但该算法存在用户隐私泄漏的风险.基于背景知识的用户个人隐私信息推断是当前Internet用户隐私信息泄漏的巨大隐患之一,差分隐私作为一种能为保护对象提供严格的理论保证的隐私保护机制而备受关注.本文把差分隐私保护技术引入TrustSVD中,提出了具有隐私保护能力的新模型DPTrustSVD.理论分析和实验结果显示,DPTrustSVD不仅为用户的隐私信息提供了严格的理论保证,而且仍然保持了较高的预测准确率. 相似文献
17.
In view of the problem of trust relationship in traditional trust-based service recommendation algorithm,and the inaccuracy of service recommendation list obtained by sorting the predicted QoS,a trust expansion and listwise learning-to-rank based service recommendation method (TELSR) was proposed.The probabilistic user similarity computation method was proposed after analyzing the importance of service sorting information,in order to further improve the accuracy of similarity computation.The trust expansion model was presented to solve the sparseness of trust relationship,and then the trusted neighbor set construction algorithm was proposed by combining with the user similarity.Based on the trusted neighbor set,the listwise learning-to-rank algorithm was proposed to train an optimal ranking model.Simulation experiments show that TELSR not only has high recommendation accuracy,but also can resist attacks from malicious users. 相似文献
18.
Combined with contexts and SVM,a service recommendation method based on context-embedded support vector machine (SRM-CESVM) was proposed.Firstly,according to the different contexts,the user rating matrix was modified to make it with embedded contexts.Secondly,the rating vectors with embedded contexts were used as service feature vectors to construct training set,meanwhile the dimension of service feature vector were not increased by the introduction of contexts.Thirdly,a separation hyperplane for active user was acquired based on training set using SVM,and then the SVM prediction model was built.Finally,the distances between the feature vector points representing the active users' unused services and the hyperplane were calculated.Considering the distances and the recommendation of similar users,the service list was recommended.The experimental results further demonstrate that the proposed method has better recommendation accuracy under different rating matrix densities and can reduce recommendation time. 相似文献
19.
传统的推荐算法受限于单领域中用户和项目的稀疏关系,也存在用户冷启动等问题.跨领域推荐能够通过学习辅助领域的知识去丰富目标领域的稀疏数据以提高推荐准确率.本文提出了一种知识聚合和迁移相结合的跨领域推荐算法ATCF.与已有算法不同,在对共性知识和个性知识的表示学习中,ATCF均充分融合了辅助域和目标域的知识,通过基于矩阵分解的两级矩阵拼接和两次矩阵填充,得到在群集矩阵及评分矩阵上的共性知识表示;通过知识迁移,构建了重叠用户和非重叠用户的个性知识表示,有效避免了负迁移.在两个跨领域数据集上开展的实验表明,ATCF算法与已有单领域和跨领域推荐算法相比RMSE降低了3%~7%,准确率召回率增加了8%~15%. 相似文献