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.     

Personalized recommendation based on review topics

The traditional collaborative filtering algorithm is a successful recommendation technology. The core idea of this algorithm is to calculate user or item similarity based on user ratings and then to predict ratings and recommend items based on similar users’ or similar items’ ratings. However, real applications face a problem of data sparsity because most users provide only a few ratings, such that the traditional collaborative filtering algorithm cannot produce satisfactory results. This paper proposes a new topic model-based similarity and two recommendation algorithms: user-based collaborative filtering with topic model algorithm (UCFTM, in this paper) and item-based collaborative filtering with topic model algorithm (ICFTM, in this paper). Each review is processed using the topic model to generate review topic allocations representing a user’s preference for a product’s different features. The UCFTM algorithm aggregates all topic allocations of reviews by the same user and calculates the user most valued features representing product features that the user most values. User similarity is calculated based on user most valued features, whereas ratings are predicted from similar users’ ratings. The ICFTM algorithm aggregates all topic allocations of reviews for the same product, and item most valued features representing the most valued features of the product are calculated. Item similarity is calculated based on item most valued features, whereas ratings are predicted from similar items’ ratings. Experiments on six data sets from Amazon indicate that when most users give only one review and one rating, our algorithms exhibit better prediction accuracy than other traditional collaborative filtering and state-of-the-art topic model-based recommendation algorithms.     

The impact of available information on negotiation results

Negotiation is the most famous tool for reaching an agreement between parties. Usually, the different parties can be modeled as a buyer and a seller, who negotiate about the price of a given item. In most cases, the parties have incomplete information about one another, but they can invest money and efforts in order to acquire information about each other. This leads to the question of how much each party will be willing to invest on information about its opponent, prior to the negotiation process. In this paper, we consider the profitability of automated negotiators acquiring information on their opponents. In our model, a buyer and a seller negotiate on the price of a given item. Time is costly, and incomplete information exists about the reservation price of both parties. The reservation price of the buyer is the maximum price it is willing to pay for an item or service, and the reservation price of the seller is the minimum price it is willing to receive in order to sell the item or service. Our research is based on Cramton’s symmetrical protocol of negotiation that provides the agents with stable and symmetric strategies, and involves a delay in proposing an offer for signaling. The parties in Cramton’s model delay their offers in order to signal their strength, and then an agreement is reached after one or two offers. We determine the Nash equilibrium for agents that prefer to purchase information. Then, in addition to the theoretical background, we used simulations to check which type of equilibrium will actually be obtained. We found that in most of the cases, each agent will prefer to purchase information only if its opponent does. The reason for these results lies in the fact that an agent that prefers to purchase information according to a one-side method, signals its weakness and thereby reduces its position in the negotiation. Our results demonstrate the efficiency of joint information acquisition by both agents, but they also show that one-sided information purchasing may be inefficient, if the acquisition activity is revealed by the opponent, which causes it to infer that the informed agent is relatively weak.     

A probabilistic inference model for recommender systems

Recommendation is an important application that is employed on the Web. In this paper, we propose a method for recommending items to a user by extending a probabilistic inference model in information retrieval. We regard the user’s preference as the query, an item as a document, and explicit and implicit factors as index terms. Additional information sources can be added to the probabilistic inference model, particularly belief networks. The proposed method also uses the belief network model to recommend items by combining expert information. Experimental results on real-world data sets show that the proposed method can improve recommendation effectiveness.     

Product recommendation with temporal dynamics

In many E-commerce recommender systems, a special class of recommendation involves recommending items to users in a life cycle. For example, customers who have babies will shop on Diapers.com within a relatively long period, and purchase different products for babies within different growth stages. Traditional recommendation algorithms produce recommendation lists similar to items that the target user has accessed before (content filtering), or compute recommendation by analyzing the items purchased by the users who are similar to the target user (collaborative filtering). Such recommendation paradigms cannot effectively resolve the situation with a life cycle, i.e., the need of customers within different stages might vary significantly. In this paper, we model users’ behavior with life cycles by employing hand-crafted item taxonomies, of which the background knowledge can be tailored for the computation of personalized recommendation. In particular, our method first formalizes a user’s long-term behavior using the item taxonomy, and then identifies the exact stage of the user. By incorporating collaborative filtering into recommendation, we can easily provide a personalized item list to the user through other similar users within the same stage. An empirical evaluation conducted on a purchasing data collection obtained from Diapers.com demonstrates the efficacy of our proposed method.     

一种融合标签和知识图谱的推荐方法

随着信息的海量增长,推荐系统成为我们日常生活中一种重要的应用。传统的推荐系统根据用户和物品的交互行为进行推荐并利用用户对物品的评分来体现用户的喜好,但是数据的稀疏性会影响推荐结果的准确度,并且简单地评分数字也难以体现用户偏好的主观性以及用户选择的可解释性。因此,该文提出了一种融合标签和知识图谱的推荐方法,其中标签是一种文本信息,其包含的丰富内容和潜在的语义信息可以体现用户对物品的主观评价,对推荐起着关键作用。而知识图谱作为一种有效的推荐辅助技术,其包含的大量实体能为物品提供更多有效的特征信息。此外,该文还提出了一种融合注意力和自注意力的混合注意力模型,通过标签和实体为物品特征分配混合注意力权重,从而提高了推荐性能。实验结果表明,在MovieLens和Last.FM数据集上,该模型的推荐性能较其他推荐算法有所提升。     

Item recommendation using tag emotion in social cataloging services

Due to the overload of contents, the user suffers from difficulty in selecting items. The social cataloging services allow users to consume items and share their opinions, which influences in not only oneself but other users to choose new items. The recommendation system reduces the problem of the choice by recommending the items considering the behavior of the people and the characteristics of the items.In this study, we propose a tag-based recommendation method considering the emotions reflected in the user’s tags. Since the user’s estimation of the item is made after consuming the item, the feelings of the user obtained during consuming are directly reflected in ratings and tags. The rating has overall valence on the item, and the tag represents the detailed feelings. Therefore, we assume that the user’s rating for an item is the basic emotion of the tag attached to the item, and the emotion of tag is adjusted by the unique emotion value of the tag. We represent the relationships between users, items, and tags as a three-order tensor and apply tensor factorization. The experimental results show that the proposed method achieves better recommendation performance than baselines.     

基于会话的多粒度图神经网络推荐模型

基于会话的推荐旨在根据当前用户的匿名会话的点击序列信息来预测用户的下一次点击行为。现有方法多数都是通过对用户会话点击序列的物品信息进行建模,并学习物品的向量表示,进而进行推荐。而作为一种粗粒度的信息,物品的类别信息对物品有聚合作用,可作为物品信息的重要补充。基于此,提出了基于会话的多粒度图神经网络推荐模型（SRMGNN）。首先,使用图神经网络（GNN）得到会话序列中的物品和物品类别的嵌入向量表示,并使用注意力网络捕捉用户的注意力信息;然后,将赋予了不同注意力权重值的物品和物品类别信息进行融合后,输入到门限循环单元（GRU）里;最后,通过GRU学习会话序列的物品时序信息,并给出推荐列表。在公开的Yoochoose数据集和Diginetica数据集上进行实验,实验结果验证了该模型在增加了物品类别信息后的优势,且实验结果表明了在Precision@20和MRR@20这2种评价指标上,该模型相较于短期注意力/记忆优先级（STAMP）模型、神经注意力（NARM）模型、GRU4REC等8种模型均有更好的效果。     

Virtual item sales as a revenue model: identifying attributes that drive purchase decisions

The global market for virtual items, characters and currencies was estimated to exceed 2.1 Billion USD in 2007. Selling virtual goods for real money is an increasingly common revenue model not only for online games and virtual worlds, but for social networking sites and other mainstream online services as well. What drives consumer spending on virtual items is an increasingly relevant question, but little research has been devoted to the topic so far. Previous literature suggests that demand for virtual items is based on the items’ ability to confer gameplay advantages on one hand, and on the items’ decorative value on the other hand. In this paper, I adopt a perspective from the sociology of consumption and analyse examples from 14 virtual asset platforms to suggest a more detailed set of item attributes that drive virtual item purchase decisions, consisting of functional, hedonic and social attributes.     

Exploit latent Dirichlet allocation for collaborative filtering

Previous work on the one-class collaborative filtering (OCCF) problem can be roughly categorized into pointwise methods, pairwise methods, and content-based methods. A fundamental assumption of these approaches is that all missing values in the user-item rating matrix are considered negative. However, this assumption may not hold because the missing values may contain negative and positive examples. For example, a user who fails to give positive feedback about an item may not necessarily dislike it; he may simply be unfamiliar with it. Meanwhile, content-based methods, e.g. collaborative topic regression (CTR), usually require textual content information of the items, and thus their applicability is largely limited when the text information is not available. In this paper, we propose to apply the latent Dirichlet allocation (LDA) model on OCCF to address the above-mentioned problems. The basic idea of this approach is that items are regarded as words, users are considered as documents, and the user-item feedback matrix constitutes the corpus. Our model drops the strong assumption that missing values are all negative and only utilizes the observed data to predict a user’s interest. Additionally, the proposed model does not need content information of the items. Experimental results indicate that the proposed method outperforms previous methods on various ranking-oriented evaluation metrics. We further combine this method with a matrix factorization-based method to tackle the multi-class collaborative filtering (MCCF) problem, which also achieves better performance on predicting user ratings.     

一种基于CW-RNN的多时间尺度序列建模推荐算法

序列化推荐试图利用用户与物品的历史交互序列,预测下次即将交互的物品。针对序列化推荐中推荐物品依赖于用户的长时间全局兴趣、中时间兴趣还是短时间局部兴趣的不确定性,该文提出了一种基于CW-RNN的多时间尺度序列建模推荐算法。首先,该算法引入CW-RNN层,从用户与物品的历史交互序列中抽取多个时间尺度的用户兴趣特征。然后,通过尺度维卷积来建模对不同时间尺度的用户兴趣特征的依赖,生成多时间尺度用户兴趣特征的统一表示。最后,利用全连接层建模统一的多尺度用户兴趣特征和隐式物品特征的交互关系。在MovieLens-1M和Amazon Movies and TV两个公开数据集上的实验结果表明,相比于现有最优的序列推荐算法,该文提出的算法在准确率上分别提升了3.80%和8.63%。     

Dynamic affinity-based classification of multi-class imbalanced data with one-versus-one decomposition: a fuzzy rough set approach

Although the relevance problem in recommender systems, which typically refers to the similarity between the preference of the user and the items the system recommends, has been well studied, the issue of making recommendations in right orders has barely been mentioned before. The order defines the way in which items should be better consumed in relation to each other. The analysis of real-world data sets demonstrates strong item consumption patterns in the degree of order. In this paper, we propose a novel method to tackle the right-order recommendation problem based on a graph structure that incorporates the item consumption orders. The proposed method can combine relevance and order effects in recommendations. We attempt to recommend the relevant items in the consecutive steps within a user session so that the user’s selected items in the consecutive steps can be stringed together in a right order. The experimental evaluation conducted on three real-world data sets shows that the recommendation accuracy is improved by considering item consumption orders compared with the baseline recommendation methods. In addition, the study on right-order recommendation contributes to the exploration on recommendation appropriateness besides relevance.     

基于多维特征交叉的深度协同过滤算法

基于深度学习的推荐算法最初以用户和物品的ID信息作为输入,但是ID无法很好地表现用户与物品的特征。在原始数据中,用户对物品的评分数据在一定程度上能表现出用户和物品的特征,但是未考虑用户的评分偏好以及物品的热门程度。在评分任务中使用隐式反馈和ID信息作为用户与物品的特征,在消除用户主观性对特征造成的噪声的同时在一定程度上缓解冷启动问题,利用单层神经网络对原始高维稀疏特征降维,使用特征交叉得到用户与物品的低阶交互,再利用神经网络捕获用户与物品的高阶交互,有效提取了特征间的高低阶交互。在四个公开数据集上的实验表明,该算法能有效提高推荐精度。     

基于加权二部图的个性化推荐算法

针对网络推断(NBI)算法的二部图实现算法忽略二部图权重而导致实际评分值高的项目没有得到优先推荐这一问题,提出加权网络推断(WNBI)算法的加权二部图实现算法。该算法以项目的评分作为二部图中用户与项目的边权,按照用户-项目间边权占该节点权重和的比例分配资源,从而实现评分值高的项目得到优先推荐。通过在数据集MovieLens上的实验表明,相比NBI算法,WNBI算法命中高评分值项目数目增多,同时在推荐列表长度小于20的情况下,命中项目的数量和命中高评分项目数量均有明显增加。     

Modeling and broadening temporal user interest in personalized news recommendation

User profiling is an important step for solving the problem of personalized news recommendation. Traditional user profiling techniques often construct profiles of users based on static historical data accessed by users. However, due to the frequent updating of news repository, it is possible that a user’s fine-grained reading preference would evolve over time while his/her long-term interest remains stable. Therefore, it is imperative to reason on such preference evaluation for user profiling in news recommenders. Besides, in content-based news recommenders, a user’s preference tends to be stable due to the mechanism of selecting similar content-wise news articles with respect to the user’s profile. To activate users’ reading motivations, a successful recommender needs to introduce “somewhat novel” articles to users.In this paper, we initially provide an experimental study on the evolution of user interests in real-world news recommender systems, and then propose a novel recommendation approach, in which the long-term and short-term reading preferences of users are seamlessly integrated when recommending news items. Given a hierarchy of newly-published news articles, news groups that a user might prefer are differentiated using the long-term profile, and then in each selected news group, a list of news items are chosen as the recommended candidates based on the short-term user profile. We further propose to select news items from the user–item affinity graph using absorbing random walk model to increase the diversity of the recommended news list. Extensive empirical experiments on a collection of news data obtained from various popular news websites demonstrate the effectiveness of our method.     

Factors influencing impulse buying during an online purchase

Using actual purchasing behavior by visitors to a High School Reunion web store, this study examines the factors that lead to an increased willingness by on-line consumers to purchase impulse items. Consistent with mental accounting and the psychophysics of prices, we find that people’s purchase of the impulse item increases with the total amount spent on other items. We also find that linking a US $1 donation to the impulse item, thereby providing a reason to purchase, increases the frequency of the impulse purchase.     

改进的哈希学习高效推荐算法

哈希技术能有效地解决推荐系统面临的存储和检索效率的问题。然而,现存的哈希推荐方法存在一个问题,推荐关注于建模用户对项目的偏好,而哈希学习关心的是相似性。为此,提出了一种改进的哈希推荐方法。计算每个用户、项目相对评分系统的均值作为偏置。对用户评分矩阵进行去偏置处理,将评分映射到相似性区间。以保持相似性为目标,提出了两种方式来分解相似性矩阵得到用户和项目的二进制码。在三个真实数据集上的实验结果表明,与其他方法对比,提出的方法在检索精度上有一定的优势。     

基于生成对抗模型的序列推荐算法

针对传统序列推荐算法时间信息和项目内容信息运用不充分的问题,该文提出基于生成对抗模型的序列推荐算法。通过生成对抗模型将序列建模与时间、内容信息建模分离,充分挖掘用户项目交互的序列信息和项目内容信息。运用卷积神经网络作为生成对抗模型的生成器,捕获用户项目交互的序列模式。运用注意力机制作为生成对抗模型的判别器,捕获交互序列的时间信息和项目内容信息。针对传统序列推荐算法时间信息建模不充分的问题,提出一种改进的时间嵌入方式,充分建模用户项目交互关于时间的周期性模式。利用生成对抗模型同时建模用户的稳定偏好和动态偏好,提升推荐系统的用户体验,并在公开数据集MovieLens-1M和Amazon-Beauty上与现有的优秀算法做比较。实验证明,该文所提出的算法在评价指标HR@N和NDCG@N上较基线方法均有一定提升。     

基于知识图谱的互补项目推荐

在缺乏用户交互互补项目方面数据的情况下,将用户对项目的偏好融合到只考虑项目关系的互补项目推荐中,提高推荐模型的性能。提出一种基于知识图谱的互补项目推荐方法,在用户历史交互项目集中推测用户交互的互补项目,基于知识图谱提取用户对互补项目的偏好,利用图像与文本学习项目之间的互补关系,最后基于神经网络实现二者的共同学习。提出的方法在Amazon数据集上与次优的基线方法相比,ACC提升了7%,precision提升了3%,这说明提出的方法性能优异。该算法共同学习用户对项目的偏好与项目之间的互补关系,提升了推荐性能。     

基于评分偏好和项目属性的协同过滤算法

针对传统的协同过滤推荐由于数据稀疏性导致物品间相似性计算不准确、推荐准确度不高的问题,文中提出了一种基于用户评分偏好模型、融合时间因素和物品属性的协同过滤算法,通过改进物品相似度度量公式来提高推荐的准确度。首先考虑到不同用户的评分习惯存在差异这一客观现象,引入评分偏好模型,通过模型计算出用户对评分类别的偏好,以用户对评分类别的偏好来代替用户对物品的评分,重建用户-物品评分矩阵;其次基于时间效应,引入时间权重因子,将时间因素纳入评分相似度计算中;然后结合物品的属性,将物品属性相似度和评分相似度进行加权,完成物品最终相似度的计算;最后通过用户偏好公式来计算用户对候选物品的偏好,依据偏好对用户进行top-N推荐。在MovieLens-100K和MovieLens-Latest-Small数据集上进行了充分实验。结果表明,相比已有的经典的协同过滤算法,所提算法的准确率和召回率在MovieLens-100K数据集上提高了9%~27%,在MovieLens-Latest-Small数据集上提高了16%~28%。因此,改进的协同过滤算法能有效提高推荐的准确度,有效缓解数据稀疏性问题。