Exploiting rich user information for one-class collaborative filtering

One-class collaborative filtering (OCCF) is an emerging setup in collaborative filtering in which only positive examples or implicit feedback can be observed. Compared with the traditional collaborative filtering setting where the data have ratings, OCCF is more realistic in many scenarios when no ratings are available. In this paper, we propose to improve OCCF accuracy by exploiting the rich user information that is often naturally available in community-based interactive information systems, including a user’s search query history, and purchasing and browsing activities. We propose two major strategies to incorporate such user information into the OCCF models: One is to linearly combine scores from different sources, and the other is to embed user information into collaborative filtering. Furthermore, we employ the MapReduce framework for similarity computation over millions of users and items. Experimental results on two large-scale retail datasets from a major e-commerce company show that the proposed methods are effective and can improve the performance of the OCCF over baseline methods through leveraging rich user information.     

Enriching one-class collaborative filtering with content information from social media

In recent years, recommender systems have become popular to handle the information overload problem of social media websites. The most widely used Collaborative Filtering methods make recommendations by mining users’ rating history. However, users’ behaviors in social media are usually implicit, where no ratings are available. This is a One-Class Collaborative Filtering (OCCF) problem with only positive examples. How to distinguish the negative examples from missing data is important for OCCF. Existing OCCF methods tackle this by the statistical properties of users’ historical behavior; however, they ignored the rich content information in social media websites, which provide additional evidence for profiling users and items. In this paper, we propose to improve OCCF accuracy by exploiting the social media content information to find the potential negative examples from the missing user-item pairs. Specifically, we get a content topic feature for each user and item by probabilistic topic modeling and embed them into the Matrix Factorization model. Extensive experiments show that our algorithm can achieve better performance than the state-of-art methods.     

耦合用户公共特征的单类协同过滤推荐算法

将显式特征与隐式反馈相结合是提高单类协同过滤(OCCF)推荐准确性的常用方法.但目前的研究一般是直接将原始显式特征或交叉特征集成到OCCF模型中,因其难以判断哪些显式特征是真正重要的,故很难获得显著的性能改进.基于此,提出了一种耦合用户公共特征的单类协同过滤推荐算法(UCC-OCCF).首先,建立基于邻居的共同偏好表示...     

Conversational Collaborative Recommendation – An Experimental Analysis

Traditionally, collaborative recommender systems have been based on a single-shot model of recommendation where a single set of recommendations is generated based on a user’s (past) stored preferences. However, content-based recommender system research has begun to look towards more conversational models of recommendation, where the user is actively engaged in directing search at recommendation time. Such interactions can range from high-level dialogues with the user, possibly in natural language, to more simple interactions where the user is, for example, asked to indicate a preference for one of k suggested items. Importantly, the feedback attained from these interactions can help to differentiate between the user’s long-term stored preferences, and her current (short-term) requirements, which may be quite different. We argue that such interactions can also be beneficial to collaborative recommendation and provide experimental evidence to support this claim.     

Hybrid Recommender Systems: Survey and Experiments

Recommender systems represent user preferences for the purpose of suggesting items to purchase or examine. They have become fundamental applications in electronic commerce and information access, providing suggestions that effectively prune large information spaces so that users are directed toward those items that best meet their needs and preferences. A variety of techniques have been proposed for performing recommendation, including content-based, collaborative, knowledge-based and other techniques. To improve performance, these methods have sometimes been combined in hybrid recommenders. This paper surveys the landscape of actual and possible hybrid recommenders, and introduces a novel hybrid, EntreeC, a system that combines knowledge-based recommendation and collaborative filtering to recommend restaurants. Further, we show that semantic ratings obtained from the knowledge-based part of the system enhance the effectiveness of collaborative filtering.     

融合显式反馈与隐式反馈的协同过滤推荐算法

显式反馈与隐式反馈相结合,可以有效提升推荐性能.但是现有的融合显式反馈与隐式反馈的推荐系统存在未能发挥隐式反馈数据缺失值反映用户隐藏偏好的能力,或者未能保留显式反馈数据反映用户偏好程度的能力的局限性.为了解决这个问题,提出了一种融合显式反馈与隐式反馈的协同过滤推荐算法.该算法分为两个阶段:第1阶段利用加权低秩近似处理隐式反馈数据,训练出隐式用户/物品向量;第2阶段引入了基线评估,同时将隐式用户/物品向量作为补充,通过显隐式用户/物品向量结合,训练得出用户对物品的预测偏好程度.该算法与多个典型算法在标准数据集上进行了实验比较,其可行性和有效性得到验证.     

基于用户部分特征的协同过滤算法

协同过滤算法作为推荐系统中应用最广泛的算法之一,在大数据环境下面临严重的数据稀疏问题,使得近邻选择的效果不佳,直接影响了算法的推荐性能.为了解决这一问题,本文提出了一种基于用户部分特征的协同过滤算法（UPCF）,该算法首先基于评分偏差和项目流行度进行矩阵缺失值填充,随后利用初始聚类中心优化的K-means算法对该填充矩阵进行项目聚类,并利用用户在项目分类下的局部特征进行近邻集合构建,最终采用基于用户的协同过滤算法获得推荐.我们采用流行的MAE指标对算法在MovieLens数据集上进行评测.实验表明,与目前流行的协同过滤算法相比,提出的UPCF算法在没有增加算法复杂性的前提下,性能有近10%的提升.     

A new approach for combining content-based and collaborative filters

With the development of e-commerce and the proliferation of easily accessible information, recommender systems have become a popular technique to prune large information spaces so that users are directed toward those items that best meet their needs and preferences. A variety of techniques have been proposed for performing recommendations, including content-based and collaborative techniques. Content-based filtering selects information based on semantic content, whereas collaborative filtering combines the opinions of other users to make a prediction for a target user. In this paper, we describe a new filtering approach that combines the content-based filter and collaborative filter to capitalize on their respective strengths, and thereby achieves a good performance. We present a series of recommendations on the selection of the appropriate factors and also look into different techniques for calculating user-user similarities based on the integrated information extracted from user profiles and user ratings. Finally, we experimentally evaluate our approach and compare it with classic filters, the result of which demonstrate the effectiveness of our approach.     

结合项目分类和云模型的协同过滤推荐算法

为了解决用户评分数据稀疏性问题和传统相似性计算方法因严格匹配对象属性而产生的弊端,结合项目分类和云模型提出了一种改进的协同过滤推荐算法。首先,按项目分类得到类别矩阵;然后利用云模型计算类内项目间的相似度并获取具有最高相似度的邻居项目的评分,为类内未评分项目进行预测填充;再利用云模型计算类内用户间的相似度得到用户邻居,最后给出最终的预测评分并产生推荐。实验结果表明,该算法不仅有效地解决了数据稀疏性及传统相似性方法存在的弊端,还提高了用户兴趣及最近邻寻找的准确性;同时,该算法只需计算新增用户或项目所在的类别即可,大大增强了系统的可扩展性。     

优化稀疏数据集提高协同过滤推荐系统质量的方法

协同过滤是目前个性化推荐系统中效果较好的一种推荐技术。由于用户和项目数量的急剧增加,使得反映用户喜好信息的评分矩阵非常稀疏,严重影响了协同过滤技术的推荐质量。针对这一问题提出了综合均值优化填充方法,该方法相比较于缺省值法和众数法,考虑到了用户评分尺度问题,同时也不存在众数法中的“多众数”和“无众数”问题。在同一数据集上,通过使用传统的基于用户的协同过滤算法进行验证,表明此方法可以有效提高推荐系统的推荐质量。     

基于内容的推荐与协同过滤融合的新闻推荐方法

针对基于内容的新闻推荐方法中用户兴趣多样性的缺乏问题和混合推荐方法存在的冷启动问题,提出一种基于内容与协同过滤融合的方法进行新闻推荐。首先利用基于内容的方法发现用户既有兴趣;再用内容与行为的混合相似度模式,寻找目标用户的相似用户群,预测用户对特征词的兴趣度,发现用户潜在兴趣;然后将用户既有兴趣与潜在兴趣融合,得到兼具个性化和多样性的用户兴趣模型;最后将候选新闻与融合模型进行相似度计算,形成推荐列表。实验结果显示,与基于内容的推荐方法相比,所提方法的F-measure和整体多样性Diversity均有明显提高;与混合推荐方法相比,性能相当,但候选新闻无需耗时积累足够的用户点击量,不存在冷启动问题。     

协同过滤推荐系统综述

随着互联网和信息计算的飞速发展,衍生了海量数据,我们已经进入信息爆炸的时代。网络中各种信息量的指数型增长导致用户想要从大量信息中找到自己需要的信息变得越来越困难,信息过载问题日益突出。推荐系统在缓解信息过载问题中起着非常重要的作用,该方法通过研究用户的兴趣偏好进行个性化计算,由系统发现用户兴趣进而引导用户发现自己的信息需求。目前,推荐系统已经成为产业界和学术界关注、研究的热点问题,应用领域十分广泛。在电子商务、会话推荐、文章推荐、智慧医疗等多个领域都有所应用。传统的推荐算法主要包括基于内容的推荐、协同过滤推荐以及混合推荐。其中,协同过滤推荐是推荐系统中应用最广泛最成功的技术之一。该方法利用用户或物品间的相似度以及历史行为数据对目标用户进行推荐,因此存在用户冷启动和项目冷启动问题。此外,随着信息量的急剧增长,传统协同过滤推荐系统面对数据的快速增长会遇到严重的数据稀疏性问题以及可扩展性问题。为了缓解甚至解决这些问题,推荐系统研究人员进行了大量的工作。近年来,为了提高推荐效果、提升用户满意度,学者们开始关注推荐系统的多样性问题以及可解释性等问题。由于深度学习方法可以通过发现数据中用户和项目之间的非线性关系从而学习一个有效的特征表示,因此越来越受到推荐系统研究人员的关注。目前的工作主要是利用评分数据、社交网络信息以及其他领域信息等辅助信息,结合深度学习、数据挖掘等技术提高推荐效果、提升用户满意度。对此,本文首先对推荐系统以及传统推荐算法进行概述,然后重点介绍协同过滤推荐算法的相关工作。包括协同过滤推荐算法的任务、评价指标、常用数据集以及学者们在解决协同过滤算法存在的问题时所做的工作以及努力。最后提出未来的几个可研究方向。     

融合显/隐式反馈的社会化协同排序推荐算法

传统的基于评分预测的社会化协同过滤推荐算法存在预测值与真实排序不匹配的固有缺陷,而基于排序预测的社会化协同排序推荐算法更符合真实的应用场景。然而,现有的大多数基于排序预测的社会化协同排序推荐算法要么仅仅关注显式反馈数据,要么仅仅关注隐式反馈数据,没有充分挖掘这些数据的价值。为充分挖掘用户的社交网络和推荐对象的显/隐式评分信息,同时克服基于评分预测的社会化协同过滤推荐算法存在的固有缺陷,在xCLiMF模型和TrustSVD模型基础上,提出一种新的融合显/隐式反馈的社会化协同排序推荐算法SPR_SVD++。该算法同时挖掘用户评分矩阵和社交网络矩阵中的显/隐式信息,并优化排序学习的评价指标预期倒数排名（ERR）。在真实数据集上的实验结果表明,采用归一化折损累计增益（NDCG）和ERR作为评价指标,SPR_SVD++算法均优于最新的TrustSVD、MERR_SVD++和SVD++算法。可见SPR_SVD++算法性能好、可扩展性强,在互联网信息推荐领域有很好的应用前景。     

融合双重正则化机制的低秩矩阵分解推荐模型

基于矩阵分解技术的社会化推荐通过加入用户信任关系来加强学习准确性,但忽略了物品之间的关联信息在模型分解过程中对用户兴趣的影响。对此首先提出在物品相似度计算方法中加入用户参与度进行改进,并构建了融合物品关联正则项和信任用户正则项双重约束的矩阵分解推荐模型,在优化隐式特征矩阵过程中体现了物品之间的关联信息对推荐的重要影响。最后通过对两个不同稀疏级别的数据集的实验证明,相比主流的矩阵分解模型,提出的双重正则项的矩阵分解模型能够提高稀疏数据集上预测评分的准确性,并能明显缓解用户冷启动问题。     

面向完全冷启动的深度混合协同过滤推荐算法

基于矩阵分解的协同过滤算法近年来获得了巨大的成功,但是依然存在冷启动,忽视用户及物品特征等问题,从而导致推荐质量不佳,用户体验度下降。论文提出了一种基于深度学习的混合协同过滤推荐算法,尝试引入堆栈降噪自编码器学习物品的隐含特征,同时结合半监督S4VM和隐含因子模型,综合考虑物品的内容特征及时间因素,以预测未评分的数据,解决冷启动问题。在标准数据集Movielens上进行的测试表明:该算法能有效预测冷启动物品的评分,性能提升显著,较传统推荐性能提升约为12%。     

一种基于用户反馈的时间感知推荐方法

将社交网络的动态性和用户反馈信息融入到推荐方法中,提出一种基于用户反馈的时间感知推荐方法。该方法利用时间衰减因子对带有时间加权的动态社交网络进行兴趣衰减分析,使时间间隔较近用户的选择行为对资源对象的推荐作用获得较高的贡献度,体现用户兴趣的时间效应特性。扩展相似度计算方法,将用户反馈表示为正反馈信息和负反馈信息,考虑用户反馈信息对推荐方法的影响。通过在社交网络真实推荐数据集上的对比实验,结果表明该方法优于基于协同过滤的推荐方法。     

A content-collaborative recommender that exploits WordNet-based user profiles for neighborhood formation

Collaborative and content-based filtering are the recommendation techniques most widely adopted to date. Traditional collaborative approaches compute a similarity value between the current user and each other user by taking into account their rating style, that is the set of ratings given on the same items. Based on the ratings of the most similar users, commonly referred to as neighbors, collaborative algorithms compute recommendations for the current user. The problem with this approach is that the similarity value is only computable if users have common rated items. The main contribution of this work is a possible solution to overcome this limitation. We propose a new content-collaborative hybrid recommender which computes similarities between users relying on their content-based profiles, in which user preferences are stored, instead of comparing their rating styles. In more detail, user profiles are clustered to discover current user neighbors. Content-based user profiles play a key role in the proposed hybrid recommender. Traditional keyword-based approaches to user profiling are unable to capture the semantics of user interests. A distinctive feature of our work is the integration of linguistic knowledge in the process of learning semantic user profiles representing user interests in a more effective way, compared to classical keyword-based profiles, due to a sense-based indexing. Semantic profiles are obtained by integrating machine learning algorithms for text categorization, namely a naïve Bayes approach and a relevance feedback method, with a word sense disambiguation strategy based exclusively on the lexical knowledge stored in the WordNet lexical database. Experiments carried out on a content-based extension of the EachMovie dataset show an improvement of the accuracy of sense-based profiles with respect to keyword-based ones, when coping with the task of classifying movies as interesting (or not) for the current user. An experimental session has been also performed in order to evaluate the proposed hybrid recommender system. The results highlight the improvement in the predictive accuracy of collaborative recommendations obtained by selecting like-minded users according to user profiles.     

基于聚类模式的推荐算法研究

经典的协作式过滤算法基于记忆的非参数局部模型,该模型应用最近邻算法（K-nearest neighbors,KNN）技术,把目标用户近邻对于目标推荐项的喜好,作为向该用户进行有效推荐的标准。该方法在预测时需要较长的运算时间,并且在特定参数的限制下,不能保证对所有的用户进行有效预测。为了解决以上问题,介绍1种基于聚类模式的新的推荐方法。该算法首先假设目标用户和推荐项均能以一定的概率划归于不同的用户模式和推荐项模式中;通过计算各个用户模式对于各个推荐项模式的评分,以及用户属于不同用户模式的概率,推荐项属于不同项目模式的概率;从而产生目标用户对于具体推荐项的预测评分。通过与经典的协作式过滤推荐算法结果的对比,该方案可以在较短的时间预测所有用户对于所有推荐项的评分,并且其推荐效果与其他方法对比有了很好的改进。     

适应用户兴趣变化的协同过滤推荐算法

协同过滤算法是至今为止最成功的个性化推荐技术之一,被应用到很多领域中 .但传统协同过滤算法不能及时反映用户的兴趣变化 .针对这个问题,提出两种改进度量:基于时间的数据权重和基于资源相似度的数据权重,在此基础上将它们有机结合,并将这两种权重引入基于资源的协同过滤算法的生成推荐过程中 .实验表明,改进后的算法比传统协同过滤算法在推荐准确度上有明显提高 .     

一种面向协同过滤的快速最近邻居搜索方法

针对协同过滤模型中寻找邻居集耗时,且部分邻居信息未能有效用于预测计算的问题,提出了一种快速搜寻最近邻居的方法。该方法改变了评分矩阵中数据组织方式,通过构建项目的用户评分列表和用户的项目评分列表,以此来筛选出对预测评分值产生影响的用户或项目,进而得到目标用户或项目的邻居集。该方法排除了不必要的相似性计算,提高了运算效率;而且还有效保证了预测计算中的邻居利用率,提高了推荐质量。在Movielens100k与Movielens1M两个数据集上的实验结果表明,所提出算法在运行时间、MAE、RMSE、F1值四个指标上均有较大提升。因此该算法在推荐系统领域具有良好的应用价值。