基于情境相似度和二次聚类的协同过滤推荐算法

随着移动互联网规模的不断扩大,传统推荐系统因较少考虑多种情境因素和用户置信度对用户偏好预测的综合影响,造成了推荐算法预测结果的偏差。针对此问题,将情境信息引入个性化推荐的过程中,提出一种基于情境相似度和二次聚类的协同过滤算法。该算法首先根据用户情境的相似度对用户进行初始聚类,再基于评分矩阵计算用户评分置信度,将用户分为核心用户和非核心用户;然后根据核心用户评分对初始聚类的簇心进行调整,并对簇中非核心用户进行重聚类,形成新的聚簇;最终根据情境相似度对用户偏好进行预测。该算法可以在一定程度上降低评分矩阵中的噪点对聚类结果的影响,提高了推荐结果的准确性。基于实际数据集的仿真实验表明,该算法与传统协同过滤算法相比能够有效提高用户偏好预测的准确性,增加协同过滤推荐算法的精确度。     

融合社交信息的局部潜在空间推荐方法

随着社交网络的发展,越来越多的研究利用社交信息来改进传统推荐算法的性能,然而现有的推荐算法大多忽略了用户兴趣的多样化,未考虑用户在不同社交维度中关心的层面不同,导致推荐质量较差.为了解决这个问题,提出了一种同时考虑全局潜在因子和不同子集特定潜在因子的推荐方法LSFS,使得推荐过程既考虑了用户共享偏好又考虑了用户在不同子集中的特定偏好.考虑到参与到不同社交维度的用户对不同的项目感兴趣,首先根据用户的社交关系将用户划分到不同的子集中;其次通过截断奇异值分解技术建模用户对项目的评分,其中全局潜在因子捕获用户共享的层面,而不同用户子集的特定潜在因子捕获用户关心的特定层面;最后,结合全局与局部潜在因子预测用户对未评分项目的评分.实验结果表明该方法可行且有效.     

基于聚类和用户偏好的协同过滤推荐算法

协同过滤推荐算法使用评分数据作为学习的数据源,针对协同过滤推荐算法中存在的评分数据稀疏以及算法的可拓展性问题,提出了一种基于聚类和用户偏好的协同过滤推荐算法。为了挖掘用户的偏好,该算法引入了用户对项目类型的平均评分到评分矩阵中,并加入了基于用户自身属性的相似度;同时,为了降低数据稀疏性,该算法使用Weighted Slope One算法填充评分数据中的未评分项,并通过融入密度和距离优化初始聚类中心的K-means算法聚类填充后的评分数据中的用户,缩小了相似用户的搜索空间;最后在聚类后的数据集中使用传统的协同过滤推荐算法生成目标用户的推荐结果。通过使用MovieLens100K数据集实验证明,提出的算法对推荐效果有所改善。     

Social recommendation based on trust and influence in SNS environments

The development of social media provides convenience to people’s lives. People’s social relationship and influence on each other is an important factor in a variety of social activities. It is obviously important for the recommendation, while social relationship and user influence are rarely taken into account in traditional recommendation algorithms. In this paper, we propose a new approach to personalized recommendation on social media in order to make use of such a kind of information, and introduce and define a set of new measures to evaluate trust and influence based on users’ social relationship and rating information. We develop a social recommendation algorithm based on modeling of users’ social trust and influence combined with collaborative filtering. The optimal linear relation between them will be reached by the proposed method, because the importance of users’ social trust and influence varies with the data. Our experimental results show that the proposed algorithm outperforms traditional recommendation in terms of recommendation accuracy and stability.

     

Personalized Tag Recommendation Using Social Influence

Tag recommendation encourages users to add more tags in bridging the semantic gap between human concept and the features of media object,which provides a feasible solution for content-based multimedia information retrieval.In this paper,we study personalized tag recommendation in a popular online photo sharing site - Flickr.Social relationship information of users is collected to generate an online social network.From the perspective of network topology,we propose node topological potential to characterize user’s social influence.With this metric,we distinguish different social relations between users and find out those who really have influence on the target users.Tag recommendations are based on tagging history and the latent personalized preference learned from those who have most influence in user’s social network.We evaluate our method on large scale real-world data.The experimental results demonstrate that our method can outperform the non-personalized global co-occurrence method and other two state-of-the-art personalized approaches using social networks.We also analyze the further usage of our approach for the cold-start problem of tag recommendation.     

基于注意力门控神经网络的社会化推荐算法

针对社会化推荐算法中存在的推荐准确率不高的问题,提出了一种多头注意力门控神经网络(MAGN)算法.具体来说,采用门控神经网络对输入的用户和用户-朋友对进行融合得到联合嵌入,利用注意力记忆网络来获取不同朋友在不同方面对用户的影响,利用多头注意力来获取在不同方面对用户影响程度偏高的几位朋友.采用门控神经网络将朋友影响和用户...     

基于遗忘函数的均值贝叶斯个性化排序算法研究

针对贝叶斯个性化排序算法未能充分应用用户的行为信息,导致算法在数据稀疏情况下推荐性能以及鲁棒性均大幅度降低的问题,提出了均值贝叶斯个性化排序(MBPR)算法,来进一步挖掘用户对隐式反馈信息的偏好关系。考虑到用户兴趣随时间变化的特征,又将遗忘函数引入MBPR算法中。该算法首先对用户的历史评分记录进行预处理;然后根据用户的评分信息对项目进行正负反馈的划分,对每名用户进行个性化建模,挖掘用户对未参与项目的喜好程度,生成推荐列表。为验证提出算法的推荐性能,在公开数据集MovieLens及Yahoo上进行分析和对比实验。实验结果表明该算法的推荐性能及鲁棒性较对比算法均有显著提高。     

Preference-based user rating correction process for interactive recommendation systems

In most of the recommendation systems, user rating is an important user activity that reflects their opinions. Once the users return their ratings about items the systems have suggested, the user ratings can be used to adjust the recommendation process.However, while rating the items users can make some mistakes (e.g., natural noises). As the recommendation systems receive more incorrect ratings, the performance of such systems may decrease. In this paper, we focus on an interactive recommendation system which can help users to correct their own ratings. Thereby, we propose a method to determine whether the ratings from users are consistent to their own preferences (represented as a set of dominant attribute values) or not and eventually to correct these ratings to improve recommendation. The proposed interactive recommendation system has been particularly applied to two user rating datasets (e.g., MovieLens and Netflix) and it has shown better recommendation performance (i.e., lower error ratings).     

Health-guided recipe recommendation over knowledge graphs

While the availability of large-scale online recipe collections presents opportunities for health consumers to access a wide variety of recipes, it can be challenging for them to discover relevant recipes. Whereas most recommender systems are designed to offer selections consistent with users’ past behavior, it remains an open problem to offer selections that can help users’ transition from one type of behavior to another, intentionally. In this paper, we introduce health-guided recipe recommendation as a way to incrementally shift users towards healthier recipe options while respecting the preferences reflected in their past choices. Introducing a knowledge graph (KG) into recommender systems as side information has attracted great interest, but its use in recipe recommendation has not been studied. To fill this gap, we consider the task of recipe recommendation over knowledge graphs. In particular, we jointly learn recipe representations via graph neural networks over two graphs extracted from a large-scale Food KG, which capture different semantic relationships, namely, user preferences and recipe healthiness, respectively. To integrate the nutritional aspects into recipe representations and the recommendation task, instead of simple fusion, we utilize a knowledge transfer scheme to enable the transfer of useful semantic information across the preferences and healthiness aspects. Experimental results on two large real-world recipe datasets showcase our model’s ability to recommend tasty as well as healthy recipes to users.     

EARS: Emotion-aware recommender system based on hybrid information fusion

Recommender systems suggest items that users might like according to their explicit and implicit feedback information, such as ratings, reviews, and clicks. However, most recommender systems focus mainly on the relationships between items and the user’s final purchasing behavior while ignoring the user’s emotional changes, which play an essential role in consumption activity. To address the challenge of improving the quality of recommender services, this paper proposes an emotion-aware recommender system based on hybrid information fusion in which three representative types of information are fused to comprehensively analyze the user’s features: user rating data as explicit information, user social network data as implicit information and sentiment from user reviews as emotional information. The experimental results verify that the proposed approach provides a higher prediction rating and significantly increases the recommendation accuracy.     

A semantic social network-based expert recommender system

This research work presents a framework to build a hybrid expert recommendation system that integrates the characteristics of content-based recommendation algorithms into a social network-based collaborative filtering system. The proposed method aims at improving the accuracy of recommendation prediction by considering the social aspect of experts’ behaviors. For this purpose, content-based profiles of experts are first constructed by crawling online resources. A semantic kernel is built by using the background knowledge derived from Wikipedia repository. The semantic kernel is employed to enrich the experts’ profiles. Experts’ social communities are detected by applying the social network analysis and using factors such as experience, background, knowledge level, and personal preferences. By this way, hidden social relationships can be discovered among individuals. Identifying communities is used for determining a particular member’s value according to the general pattern behavior of the community that the individual belongs to. Representative members of a community are then identified using the eigenvector centrality measure. Finally, a recommendation is made to relate an information item, for which a user is seeking an expert, to the representatives of the most relevant community. Such a semantic social network-based expert recommendation system can provide benefits to both experts and users if one looks at the recommendation from two perspectives. From the user’s perspective, she/he is provided with a group of experts who can help the user with her/his information needs. From the expert’s perspective she/he has been assigned to work on relevant information items that fall under her/his expertise and interests.     

A Novel Approach Based on Multi-View Content Analysis and Semi-Supervised Enrichment for Movie Recommendation

Although many existing movie recommender systems have investigated recommendation based on information such as clicks and tags, much less efforts have been made to explore the multimedia content of movies, which has potential information for the elicitation of the user’s visual and musical preferences.In this paper, we explore the content from three media types (image, text, audio) and propose a novel multi-view semi-supervised movie recommendation method, which represents each media type as a view space for movies.The three views of movies are integrated to predict the rating values under the multi-view framework.Furthermore, our method considers the casual users who rate limited movies.The algorithm enriches the user profile with a semi-supervised way when there are only few rating histories.Experiments indicate that the multimedia content analysis reveals the user’s profile in a more comprehensive way.Different media types can be a complement to each other for movie recommendation.And the experimental results validate that our semi-supervised method can effectively enrich the user profile for recommendation with limited rating history.     

基于对抗采样的社交推荐算法

推荐系统利用用户的历史记录、物品的基础信息等数据进行建模来捕获用户的偏好,有效缓解了信息过载等问题,虽然其已应用广泛,但整个推荐领域面临的挑战却依旧存在,其中数据稀疏这一问题对于推荐性能有举足轻重的影响。近年来,大量研究表明基于社交信息的推荐算法能够有效缓解数据稀疏问题,但它们也仍然存在一定的局限。线上的社交网络是非常稀疏的,并且线上社交网络中的“朋友”通常包括同学、同事、亲戚等,因此,拥有显式朋友关系的用户不一定拥有相似的偏好,即直接利用显式朋友的兴趣偏好进行推荐会存在噪声问题。此外,大部分基于隐式反馈的算法通常直接对用户没有交互过的物品进行随机采样,然后将其作为用户实际交互过的物品的负样本来优化模型,然而用户没有交互过的物品并不代表用户不喜欢,这种粗粒度的采样策略忽略了用户的真实偏好,同样也带来了一定程度的噪声。生成对抗网络（GANs）因其在训练中捕获复杂数据分布的能力以及强大的鲁棒性被广泛应用到推荐系统中,为了减弱上述噪声问题带来的影响,本文基于生成对抗网络提出了一种细粒度的对抗采样推荐模型（ASGAN）,包括一个生成器和判别器。其中,生成器首先利用图表示学习技术初始化社交网络,接着为用户生成一个与其偏好相似的朋友,然后再从该朋友喜欢的物品集中同时生成该用户喜欢的物品和用户不喜欢的物品。判别器则尽可能区分出用户实际交互过的物品和生成器生成的两类物品。随着对抗训练的进行,生成器能更有效地进行社交朋友采样和物品采样,而判别器能够良好地捕获用户的真实偏好分布。最后,在三个公开的真实数据集上与现有的六个工作进行对比,实验结果证明：ASGAN拥有更好的推荐性能,通过重构社交网络和细粒度采样有效缓解了社交信息和物品采样策略带来的噪声问题。     

Facebook single and cross domain data for recommendation systems

The emergence of social networks and the vast amount of data that they contain about their users make them a valuable source for personal information about users for recommender systems. In this paper we investigate the feasibility and effectiveness of utilizing existing available data from social networks for the recommendation process, specifically from Facebook. The data may replace or enrich explicit user ratings. We extract from Facebook content published by users on their personal pages about their favorite items and preferences in the domain of recommendation, and data about preferences related to other domains to allow cross-domain recommendation. We study several methods for integrating Facebook data with the recommendation process and compare the performance of these methods with that of traditional collaborative filtering that utilizes user ratings. In a field study that we conducted, recommendations obtained using Facebook data were tested and compared for 95 subjects and their crawled Facebook friends. Encouraging results show that when data is sparse or not available for a new user, recommendation results relying solely on Facebook data are at least equally as accurate as results obtained from user ratings. The experimental study also indicates that enriching sparse rating data by adding Facebook data can significantly improve results. Moreover, our findings highlight the benefits of utilizing cross domain Facebook data to achieve improvement in recommendation performance.     

Not Even Past: Information Aging and Temporal Privacy in Online Social Networks

Online social networks (OSNs) make information accessible for unlimited periods and provide easy access to past information by arranging information in time lines or by providing sophisticated search mechanisms. Despite increased concerns over the privacy threat that is posed by digital memory, there is little knowledge about retrospective privacy: the extent to which the age of the exposed information affects sharing preferences. In this article, we investigate how information aging impacts users’ sharing preferences on Facebook. Our findings are based on a between-subjects experiment (n = 272), in which we measured the impact of time since first publishing an OSN post on its sharing preferences. Our results quantify how willingness to share is lower for older Facebook posts and show that older posts have lower relevancy to the user’s social network and are less representative of the user’s identity. We show that changes in the user’s social circles, the occurrence of significant life changes and a user’s young age are correlated with a further decrease in the willingness to keep sharing past information. We discuss our findings by juxtaposing digital memory theories and privacy theories and suggest a vision for mechanisms that can help users manage longitudinal privacy.     

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.     

融合信任关系与评论文本的矩阵分解推荐算法

针对推荐系统中用户评分数据稀疏所导致推荐结果不精确的问题,本文尝试将用户评分、信任关系和项目评论文本信息融合在概率矩阵分解方法中以缓解评分数据稀疏性问题.首先以共同好友数目及项目流行度改进皮尔逊用户偏好相似程度并将其作为用户间的直接信任值,然后考虑用户间信任传播过程中所有路径的影响构建新的信任网络;其次通过BERT预训练(Pre-training of Deep Bidirectional Transformers for Language Understanding)模型提取项目的评论文本向量,构造项目的评论文本特征矩阵;最后基于概率矩阵分解(Probabilistic Matrix Factorization,PMF)模型融合用户的评分数据、用户的信任关系以及项目的评论文本信息进行推荐.通过不断的理论分析并在真实的Yelp数据集上进行实验验证,均表明本文算法的有效性.     

一种优化聚类的协同过滤推荐算法

针对传统的协同过滤推荐算法存在评分数据稀疏和推荐准确率偏低的问题,提出了一种优化聚类的协同过滤推荐算法。根据用户的评分差异对原始评分矩阵进行预处理,再将得到的用户项目评分矩阵以及项目类型矩阵构造用户类别偏好矩阵,更好反映用户的兴趣偏好,缓解数据的稀疏性。在该矩阵上利用花朵授粉优化的模糊聚类算法对用户聚类,增强用户的聚类效果,并将项目偏好信息的相似度与项目评分矩阵的相似度进行加权求和,得到多个最近邻居。融合时间因素对目标用户进行项目评分预测,改善用户兴趣变化对推荐效果的影响。通过在MovieLens 100k数据集上实验结果表明,提出的算法缓解了数据的稀疏性问题,提高了推荐的准确性。     

Exploiting interactions of review text,hidden user communities and item groups,and time for collaborative filtering

Rich side information concerning users and items are valuable for collaborative filtering (CF) algorithms for recommendation. For example, rating score is often associated with a piece of review text, which is capable of providing valuable information to reveal the reasons why a user gives a certain rating. Moreover, the underlying community and group relationship buried in users and items are potentially useful for CF. In this paper, we develop a new model to tackle the CF problem which predicts user’s ratings on previously unrated items by effectively exploiting interactions among review texts as well as the hidden user community and item group information. We call this model CMR (co-clustering collaborative filtering model with review text). Specifically, we employ the co-clustering technique to model the user community and item group, and each community–group pair corresponds to a co-cluster, which is characterized by a rating distribution in exponential family and a topic distribution. We have conducted extensive experiments on 22 real-world datasets, and our proposed model CMR outperforms the state-of-the-art latent factor models. Furthermore, both the user’s preference and item profile are drifting over time. Dynamic modeling the temporal changes in user’s preference and item profiles are desirable for improving a recommendation system. We extend CMR and propose an enhanced model called TCMR to consider time information and exploit the temporal interactions among review texts and co-clusters of user communities and item groups. In this TCMR model, each community–group co-cluster is characterized by an additional beta distribution for time modeling. To evaluate our TCMR model, we have conducted another set of experiments on 22 larger datasets with wider time span. Our proposed model TCMR performs better than CMR and the standard time-aware recommendation model on the rating score prediction tasks. We also investigate the temporal effect on the user–item co-clusters.     

Collaborative filtering and deep learning based recommendation system for cold start items

Recommender system is a specific type of intelligent systems, which exploits historical user ratings on items and/or auxiliary information to make recommendations on items to the users. It plays a critical role in a wide range of online shopping, e-commercial services and social networking applications. Collaborative filtering (CF) is the most popular approaches used for recommender systems, but it suffers from complete cold start (CCS) problem where no rating record are available and incomplete cold start (ICS) problem where only a small number of rating records are available for some new items or users in the system. In this paper, we propose two recommendation models to solve the CCS and ICS problems for new items, which are based on a framework of tightly coupled CF approach and deep learning neural network. A specific deep neural network SADE is used to extract the content features of the items. The state of the art CF model, timeSVD++, which models and utilizes temporal dynamics of user preferences and item features, is modified to take the content features into prediction of ratings for cold start items. Extensive experiments on a large Netflix rating dataset of movies are performed, which show that our proposed recommendation models largely outperform the baseline models for rating prediction of cold start items. The two proposed recommendation models are also evaluated and compared on ICS items, and a flexible scheme of model retraining and switching is proposed to deal with the transition of items from cold start to non-cold start status. The experiment results on Netflix movie recommendation show the tight coupling of CF approach and deep learning neural network is feasible and very effective for cold start item recommendation. The design is general and can be applied to many other recommender systems for online shopping and social networking applications. The solution of cold start item problem can largely improve user experience and trust of recommender systems, and effectively promote cold start items.