Exploiting two-faceted web of trust for enhanced-quality recommendations

Traditional collaborative filtering (CF) based recommender systems on the basis of user similarity often suffer from low accuracy because of the difficulty in finding similar users. Incorporating trust network into CF-based recommender system is an attractive approach to resolve the neighbor selection problem. Most existing trust-based CF methods assume that underlying relationships (whether inferred or pre-existing) can be described and reasoned in a web of trust. However, in online sharing communities or e-commerce sites, a web of trust is not always available and is typically sparse. The limited and sparse web of trust strongly affects the quality of recommendation. In this paper, we propose a novel method that establishes and exploits a two-faceted web of trust on the basis of users’ personal activities and relationship networks in online sharing communities or e-commerce sites, to provide enhanced-quality recommendations. The developed web of trust consists of interest similarity graphs and directed trust graphs and mitigates the sparsity of web of trust. Moreover, the proposed method captures the temporal nature of trust and interest by dynamically updating the two-faceted web of trust. Furthermore, this method adapts to the differences in user rating scales by using a modified Resnick’s prediction formula. As enabled by the Pareto principle and graph theory, new users highly benefit from the aggregated global interest similarity (popularity) in interest similarity graph and the global trust (reputation) in the directed trust graph. The experiments on two datasets with different sparsity levels (i.e., Jester and MovieLens datasets) show that the proposed approach can significantly improve the predictive accuracy and decision-support accuracy of the trust-based CF recommender system.     

基于用户信任和兴趣的概率矩阵分解推荐方法

传统协同过滤推荐算法存在数据稀疏性、冷启动、新用户等问题.随着社交网络和电子商务的迅猛发展,利用用户间的信任关系和用户兴趣提供个性化推荐成为研究的热点.本文提出一种结合用户信任和兴趣的概率矩阵分解（STUIPMF）推荐方法.该方法首先从用户评分角度挖掘用户间的隐性信任关系和潜在兴趣标签,然后利用概率矩阵分解模型对用户评分信息、用户信任关系、用户兴趣标签信息进行矩阵分解,进一步挖掘用户潜在特征,缓解数据稀疏性.在Epinions数据集上进行实验验证,结果表明,该方法能够在一定程度上提高推荐精度,缓解冷启动和新用户问题,同时具有较好的可扩展性.     

基于兴趣感知和时间因子的个性化菜品推荐 *

目前推荐系统已广泛应用在各种电子商务网站上,但针对菜品的个性化推荐很少。针对菜品推荐中存在别名多、用户菜品矩阵稀疏以及新用户冷启动等难题,对基于用户的协同过滤算法进行改进,设计一种融合专家选择和在线推荐的菜品推荐系统。专家选择通过对菜品进行种类层次划分为用户兴趣建模做准备,在线推荐通过兴趣感知选择算法选择餐厅中的专家用户和候选菜品,从而实现对用户菜品的推荐。最后通过在候选菜品选择时引入时间敏感因子和协同过滤中引入时间遗忘因子,改进兴趣感知算法和菜品偏好预测效果。实验结果表明,所设计算法较传统算法在准确性和推荐效率有明显改进,并得出了针对菜品推荐时引入时间因子有利提高推荐准确性的结论。     

Web page recommendation via twofold clustering: considering user behavior and topic relation

Web page recommendations have attracted increasing attention in recent years. Web page recommendation has different characteristics compared to the classical recommenders. For example, the recommender cannot simply use the user-item utility prediction method as e-commerce recommendation, which would face the repeated item cold-start problem. Recent researches generally classify the web page articles before recommending. But classification often requires manual labors, and the size of each category may be too large. Some studies propose to utilize clustering method to preprocess the web page corpus and achieve good results. But there are many differences between different clustering methods. For instance, some clustering methods are of high time complexity; in addition, some clustering methods rely on initial parameters by iterative computing whose results probably aren’t stable. In order to solve the above issues, we propose a web page recommendation based on twofold clustering by considering both effectiveness and efficiency, and take the popularity and freshness factors into account. In our proposed clustering, we combined the strong points of density-based clustering and the k-means clustering. The core idea is that we used the density-based clustering in sample data to get the number of clusters and the initial center of each cluster. The experimental results show that our method performs better diversity and accuracy compared to the state-of-the-art approaches.

     

Improving recommendations through an assumption-based multiagent approach: An application in the tourism domain

Recommender systems are popular tools dealing with the information overload problem in e-commerce web sites. The more they know about the users, the better recommendations they can provide. However, sometimes, in real situations, it is necessary to make guesses about the value of missing but useful data in order to generate a recommendation immediately, rather than waiting the data becomes available. This paper presents an assumption-based multiagent recommender system capable of making these types of assumptions about the preferences of the users. The approach was validate in the tourism domain (recommendation of travel packages). Experiments were conducted to illustrate the impact of various assumption making strategies on the quality of the recommendations as well as the impact of trust assignment.     

一种基于用户播放行为序列的个性化视频推荐策略

本文针对在线视频服务网站的个性化推荐问题,提出了一种基于用户播放行为序列的个性化推荐策略.该策略通过深度神经网络词向量模型分析用户播放视频行为数据,将视频映射成等维度的特征向量,提取视频的语义特征.聚类用户播放历史视频的特征向量,建模用户兴趣分布矩阵.结合用户兴趣偏好和用户观看历史序列生成推荐列表.在大规模的视频服务系统中进行了离线实验,相比随机算法、基于物品的协同过滤和基于用户的协同过滤传统推荐策略,本方法在用户观看视频的Top-N推荐精确率方面平均分别获得22.3%、30.7%和934%的相对提升,在召回率指标上分别获得52.8%、41%和1065%的相对提升.进一步地与矩阵分解算法SVD++、基于双向LSTM模型和注意力机制的Bi-LSTM+Attention算法和基于用户行为序列的深度兴趣网络DIN比较,Top-N推荐精确率和召回率也得到了明显提升.该推荐策略不仅获得了较高的精确率和召回率,还尝试解决传统推荐面临大规模工业数据集时的数据要求严苛、数据稀疏和数据噪声等问题.     

基于联合概率矩阵分解的上下文广告推荐算法

上下文广告与用户兴趣及网页内容相匹配,可增强用户体验并提高广告点击率.而广告收益与广告点击率直接相关,准确预测广告点击率是提高上下文广告收益的关键.目前,上下文广告推荐面临如下问题:(1) 网页数量及用户数量规模很大;(2) 历史广告点击数据十分稀疏,导致点击率预测准确率低.针对上述问题,提出一种基于联合概率矩阵分解的因子模型AdRec,它结合用户、广告和网页三者信息进行广告推荐,以解决数据稀疏时点击率预测准确率低的问题.算法复杂度随着观测数据数量的增加呈线性增长,因此可应用于大规模数据.     

Leveraging clustering approaches to solve the gray-sheep users problem in recommender systems

Recommender systems apply data mining and machine learning techniques for filtering unseen information and can predict whether a user would like a given item. This paper focuses on gray-sheep users problem responsible for the increased error rate in collaborative filtering based recommender systems. This paper makes the following contributions: we show that (1) the presence of gray-sheep users can affect the performance – accuracy and coverage – of the collaborative filtering based algorithms, depending on the data sparsity and distribution; (2) gray-sheep users can be identified using clustering algorithms in offline fashion, where the similarity threshold to isolate these users from the rest of community can be found empirically. We propose various improved centroid selection approaches and distance measures for the K-means clustering algorithm; (3) content-based profile of gray-sheep users can be used for making accurate recommendations. We offer a hybrid recommendation algorithm to make reliable recommendations for gray-sheep users. To the best of our knowledge, this is the first attempt to propose a formal solution for gray-sheep users problem. By extensive experimental results on two different datasets (MovieLens and community of movie fans in the FilmTrust website), we showed that the proposed approach reduces the recommendation error rate for the gray-sheep users while maintaining reasonable computational performance.     

嵌入LDA主题模型的协同过滤推荐算法

协同过滤推荐算法由于其推荐的准确性和高效性已经成为推荐领域最流行的推荐算法之一。该算法通过分析用户的历史评分记录来构建用户兴趣模型,进而为用户产生一组推荐。然而,推荐系统中用户的评分记录是极为有限的,导致传统协同过滤算法面临严重的数据稀疏性问题。针对此问题,提出了一种改进的嵌入LDA主题模型的协同过滤推荐算法(ULR-CF算法)。该算法利用LDA主题建模方法在用户项目标签集上挖掘潜在的主题信息,进而结合文档-主题概率分布矩阵和评分矩阵来共同度量用户和项目相似度。实验结果表明,提出的ULR-CF算法可以有效缓解数据稀疏性问题,并能显著提高推荐系统的准确性。     

基于Web日志挖掘和相关性度量的电子商务推荐系统

基于Web日志挖掘的个性化推荐技术已在电子商务网站中广泛应用,针对现有推荐系统的准确性不高等问题,提出一种基于Web日志挖掘和相关性度量的个性化推荐系统. 首先,提取用户的访问日志,并对其进行预处理,以获得精简的结构化数据. 然后,对日志进行分析,提取出特征序列. 再后,根据特征的出现频率和页面停留时间,计算出页面与交易文本文档的相关性. 最终,利用夹角余弦公式计算出用户与页面的相关性,并以此形成推荐列表. 实验结果表明,该方案能够根据用户偏好精确的给出个性化推荐.     

基于码本聚类和因子分解机的多指标推荐算法

传统的协同过滤推荐算法普遍存在数据稀疏问题,且仅利用单一综合评分来计算用户相似度,无法找到在多个指标上偏好相似的用户,因而影响推荐的准确度。多指标评分推荐算法力图寻找在多个指标上偏好相似的用户,但是其评价成本高,导致数据稀疏性问题更加严重。为了找到与目标用户在多个指标上偏好相似的用户,提出基于码本聚类的思想来获取用户在各指标上的评分风格信息,然后基于评分风格信息将用户和项目在各指标上进行双向聚类,最后利用因子分解机模型(Factorization Machines,FMs)基于同一簇内的用户、项目、多指标评分信息、评分风格信息进行推荐。实验结果表明,与传统的协同过滤算法和其他多指标推荐方法相比,基于多指标评分信息的因子分解机推荐算法能够在一定程度上缓解数据稀疏问题,提高推荐的准确度。     

基于改进聚类和矩阵分解的协同过滤推荐算法

大数据背景下,对于传统的协同过滤推荐算法在电子商务系统中的数据稀疏性、准确性不高、实时性不足等问题,提出一种改进的协同过滤推荐算法。该算法首先通过矩阵分解实现对原始数据的降维及其数据填充,并引入了时间衰减函数预处理用户评分,用项目的属性向量来表征项目,用用户的兴趣向量来表征用户,通过k-means聚类算法对用户和项目分别进行聚类;然后使用改进相似性度量方法在簇中查找用户的最近邻和项目推荐候选集,产生推荐。实验结果表明,该算法不仅可以有效解决数据稀疏和新项目带来的冷启动问题,而且还可以在多维度下反映用户的兴趣变化,推荐算法的准确度明显提升。     

结合使用挖掘和内容挖掘的web推荐服务

随着Internet的基础结构不断扩大和其所含信息的持续增长,Internet用户越来越感觉容易在WWW服务中“资源迷向”。提高用户访问效率的方法有页面预取技术,站点动态重构技术和web个性化推荐技术等。现有的大多数web个性化推荐技术主要是基于用户使用记录的数据挖掘方法,没有或很少考虑结合页面内容—这才是用户真正感兴趣的。该文提出一种结合用户使用挖掘和内容挖掘的web推荐服务,该推荐服务根据频繁最大前向访问路径,提出含有导航页和内容页的频繁访问路径图概念,根据滑动窗口内的最近用户访问页面内容和候选推荐集中页面内容相关性,来向用户提供个性化推荐服务。经推荐质量分析,这种方法具有较好的推荐优化能力。     

An optimized approach for massive web page classification using entity similarity based on semantic network

With the development of mobile technology, the users browsing habits are gradually shifted from only information retrieval to active recommendation. The classification mapping algorithm between users interests and web contents has been become more and more difficult with the volume and variety of web pages. Some big news portal sites and social media companies hire more editors to label these new concepts and words, and use the computing servers with larger memory to deal with the massive document classification, based on traditional supervised or semi-supervised machine learning methods. This paper provides an optimized classification algorithm for massive web page classification using semantic networks, such as Wikipedia, WordNet. In this paper, we used Wikipedia data set and initialized a few category entity words as class words. A weight estimation algorithm based on the depth and breadth of Wikipedia network is used to calculate the class weight of all Wikipedia Entity Words. A kinship-relation association based on content similarity of entity was therefore suggested optimizing the unbalance problem when a category node inherited the probability from multiple fathers. The keywords in the web page are extracted from the title and the main text using N-gram with Wikipedia Entity Words, and Bayesian classifier is used to estimate the page class probability. Experimental results showed that the proposed method obtained good scalability, robustness and reliability for massive web pages.     

Research on personalized recommendation hybrid algorithm for interactive experience equipment

Interactive calligraphy experience equipment has the characteristics of a large amount of data, various types, and strong homogeneity, which makes it difficult for users to find interesting resources. In this article, a hybrid personalized recommendation algorithm is proposed, which uses collaborative filtering and content-based recommendation methods in turn to make recommendations. In the initial recommendation, Latent Dirichlet Allocation (LDA) topic model is used to reduce the dimension of high-dimensional user behavior data and establish a user-writing theme matrix to reduce inaccurate recommendation caused by high sparsity data in collaborative filtering algorithm. The user interest list is obtained by calculating the similarity between users. Then, on the basis of the preliminary recommendation results, VGG16 model is used to extract the feature vector of the calligraphy image and calculate the similarity between the user's calligraphy words and the primary recommended calligraphy words, thus obtaining the final recommendation results. The experimental results verify the effectiveness and accuracy of the recommendation algorithm, which are better than other recommendation algorithms on the whole, and have important engineering guiding significance.     

融合类别信息和用户兴趣度的协同过滤推荐算法

协同过滤直接根据用户的行为记录去预测其可能感兴趣的项目,是现今最成功、应用最广泛的推荐技术。推荐的准确度受相似性度量方法效果的影响。传统的相似性度量方法主要关注用户共同评分项之间的相似度,忽视了评分项目中的类别信息,在面对数据稀疏性问题时存在一定的不足。针对上述问题,提出基于分类信息 的评分矩阵填充方法,结合用户兴趣相似度计算方法并充分考虑到评分项目的类别信息,使得兴趣度的度量更加符合推荐系统应用的实际情况。实验结果表明,该算法可以弥补传统相似性度量方法的不足,缓解评分数据稀疏对协同过滤算法的影响,能够提高推荐的准确性、多样性和新颖性。     

基于多示例学习技术的Web目录页面链接推荐

在Web目录页面中,向用户推荐其感兴趣的链接有助于用户高效地访问网络资源.然而,用户往往不愿花费很多时间来标记训练样本,其提供的数据可能只能说明某个目录网页是否包含其感兴趣的内容,而不能明确标示出其感兴趣的具体链接.由于训练数据中缺乏对链接的标记,但预测时却需要找出用户感兴趣的链接,这就使得Web目录页面链接推荐问题相当困难.CkNN-ROI算法被提出用于解决该问题.实验表明,CkNN-ROI算法在解决这一困难的链接推荐问题上比其他一些算法更为有效.     

基于多上下文信息的协同过滤推荐算法

随着电子商务和互联网的发展,数据信息呈爆炸式增长,协同过滤算法作为一种简单而高效的推荐算法,能在一定程度上有效地解决信息爆炸问题.但是传统协同过滤算法仅通过单一评分来挖掘相似用户,推荐效果并不占优势.为了提高个性化推荐的质量,如何充分利用用户(物品)的文本、图片、标签等上下文信息以使数据价值最大化是当前推荐系统亟待解决...     

An Adaptive Recommendation System without Explicit Acquisition of User Relevance Feedback

Recommendation systems are widely adopted in e-commerce businesses for helping customers locate products they would like to purchase. In an earlier work, we introduced a recommendation system, termed Yoda, which employs a hybrid approach that combines collaborative filtering (CF) and content-based querying to achieve higher accuracy for large-scale Web-based applications. To reduce the complexity of the hybrid approach, Yoda is structured as a tunable model that is trained off-line and employed for real-time recommendation on-line. The on-line process benefits from an optimized aggregation function with low complexity that allows the real-time aggregation based on confidence values of an active user to pre-defined sets of recommendations. In this paper, we extend Yoda to include more recommendation sets. The recommendation sets can be obtained from different sources, such as human experts, web navigation patterns, and clusters of user evaluations.More over, the extended Yoda can learn the confidence values automatically by utilizing implicit users' relevance feedback through web navigations using genetic algorithms (GA). Our end-to-end experiments show while Yoda's complexity is low and remains constant as the number of users and/or items grow, its accuracy surpasses that of the basic nearest-neighbor method by a wide margin (in most cases more than 100%). The experimental results also indicate that the retrieval accuracy is significantly increased by using the GA-based learning mechanism.     

基于信任传播的协同过滤算法

电子商务环境下为用户提供高效的推荐是一个非常有意义的课题,然而稀疏性问题严重影响了推荐系统的推荐质量。为了有效解决这个问题,提出了一种基于信任传播的TSRCF协同过滤算法,在信任传播的基础上,提出了信任度,相似度,关系度的混合权重TSR,取代了传统的协同过滤算法的相似度,作为寻找邻居用户的标准。TSRCF算法在一定程度上缓解了稀疏性问题,帮助用户在信息过载的情境下得到高质量的推荐。在Epinions数据集和FilmTrust数据集上的仿真实验也验证了TSRCF算法比传统CF算法有更高的推荐精确度。