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.     

LDA-CF: 一种混合协同过滤方法

推荐系统是一种克服信息过载的重要工具,其中最流行的方法是协同过滤。该文提出一种结合潜在因素模型和邻域方法的混合协同过滤方法LDA-CF。我们首先将评分矩阵转换成伪文档集合,使用LDA(Latent Dirichlet Allocation)主题模型发现用户和物品潜在因素向量;然后在低维潜在因素空间计算用户和物品相似度;最后采用邻域方法预测未知评分。在MovieLens 100k数据集上的实验表明: 在评分预测任务中,LDA-CF取得的MAE性能指标优于传统的邻域方法。因此,LDA可以有效地从评分矩阵中发现对计算相似度十分有用的用户和物品低维特征表示,在一定程度上缓解了数据稀疏问题。     

CROC: A New Evaluation Criterion for Recommender Systems

Evaluation of a recommender system algorithm is a challenging task due to the many possible scenarios in which such systems may be deployed. We have designed a new performance plot called the CROC curve with an associated statistic: the area under the curve. Our CROC curve supplements the widely used ROC curve in recommender system evaluation by discovering performance characteristics that standard ROC evaluation often ignores. Empirical studies on two domains and including several recommender system algorithms demonstrate that combining ROC and CROC curves in evaluation can lead to a more informed characterization of performance than using either curve alone.     

广域推荐:社会网络与协同过滤(英文)

商务企业应用数据挖掘技术向潜在客户推荐产品。大多数推荐系统聚焦研究兴趣于特定的领域,如电影或书籍。使用用户相似度或产品相似度的推荐算法通常可以达到较好效果。然而,当面临其他领域问题时,推荐常变得非常困难,因为数据过于稀疏,难以仅基于购买历史发现用户或产品间的相似性。为解决此问题,提出使用社会网络数据,通过对历史的观察提高产品推荐有效性。利用人工协同过滤和基于社会网络的推荐算法的最新进展进行领域推荐工作。研究显示社会网络的应用对于产品推荐具有很强的指导作用,但是,高的推荐精度需以牺牲召回率为代价。数据的稀疏性意味着社会网络并不总是可用,在这种情况下提出一种解决方案,很好地利用了社会网络的有效信息。     

Generalization of recommender systems: Collaborative filtering extended to groups of users and restricted to groups of items

In this paper we present a collaborative filtering method which opens up the possibilities of traditional collaborative filtering in two aspects: (1) it enables joint recommendations to groups of users and (2) it enables the recommendations to be restricted to items similar to a set of reference items. By way of example, a group of four friends could request joint recommendations of films similar to “Avatar” or “Titanic”. In the paper, using experiments, we show that the traditional approach of collaborative filtering does not satisfactorily resolve the new possibilities contemplated; we also provide a detailed formulation of the method proposed and an extensive set of experiments and comparative results which show the superiority of designed collaborative filtering compared to traditional collaborative filtering in: (a) number of recommendations obtained, (b) quality of the predictions, (c) quality of the recommendations. The experiments have been carried out on the databases Movielens and Netflix.     

EFFICIENT PERSONALIZATION OF E‐LEARNING ACTIVITIES USING A MULTI‐DEVICE DECENTRALIZED RECOMMENDER SYSTEM

Personalization is becoming a key issue in designing effective e‐learning systems and, in this context, a promising solution is represented by software agents. Usually, these systems provide the student with a student agent that interacts with a site agent associated with each e‐learning site. However, in presence of a large number of students and of e‐learning sites, the tasks of the agents are often onerous, even more if the student agents run on devices with limited resources. To face this problem, we propose a new multiagent learning system, called ISABEL. Our system provides each student, that are using a specific device, with a device agent able to autonomously monitor the student's behavior when accessing e‐learning Web sites. Each site is associated, in its turn, with a teacher agent. When a student visits an e‐learning site, the teacher agent collaborates with some tutor agents associated with the student, to provide him with useful recommendations. We present both theoretical and experimental results to show that this distributed approach introduces significant advantages in quality and efficiency of the recommendation activity with respect to the performances of other past recommenders.     

基于云计算技术的个性化推荐系统

传统的协同过滤推荐技术在大数据环境下存在一定的不足。针对该问题,提出了一种基于云计算技术的个性化推荐方法：将大数据集和推荐计算分解到多台计算机上并行处理。在对经典ItemCF算法MapReduce化后,建立了一个基于Hadoop开源框架的并行推荐引擎,并通过在已商用的英语训练平台上进行学习推荐工作验证了该系统的有效性。实验结果表明,在集群中使用云计算技术处理海量数据,可以大大提高推荐系统的可扩展性。     

协同深度学习推荐算法研究

针对当用户评分较少时,推荐系统由于数据稀疏推荐性能显著降低这一问题,介绍了协同深度学习算法（Collaborative In Deep Learning,CIDL）.本算法首先对大量数据进行深度学习,然后对数据文本进行挖掘提取词汇表,最后对评级（反馈）矩阵进行协同过滤,从而得出对用户的推荐项目.本文使用真实的电影数据进行实验,与另外四种优秀算法进行对比,证明该算法可以真实有效得解决由于数据稀疏使得性能降低的问题,并提高推荐的准确度.     

协同过滤推荐中一种改进的信息核提取方法

针对协同过滤推荐算法中存在的可扩展性问题,在原有基于频率（frequency-based,FB）和排名（rank-based,RB）的信息核提取方法的基础上,提出了改进的提取信息核方法IFB（IFrequency-based）和IRB（IRank-based,IRB）,在寻找最相似邻居环节中提出了一个优化集的概念,在优化集上为每个用户寻找最相似的邻居。从实验结果看出,通过该方法能够得到更加准确的推荐结果,有效降低了绝对平均误差（MAE）,同时具有更高的准确率和召回率,推荐效果更优。     

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。     

针对隐式反馈推荐系统的表征学习方法

基于邻域的top-N推荐算法利用隐式反馈数据建立排序模型,其算法性能严重依赖于相似度函数的表现。传统相似性度量函数在隐式反馈数据上会遇到数据过于稀疏和维数过高两个问题,稀疏数据不利于推荐模型选取光滑的邻域,过高的数据维数会导致维数灾难问题,导致推荐算法表现较差。为此提出一种基于表征学习方法的推荐算法,改进算法实现了基于二部图网络的多目标节点表征学习方法,在节点表征中通过嵌入不同层次的网络结构信息和适合推荐任务的次序信息来提升推荐性能。三个不同规模真实数据集上的实验结果表明,该算法相较于常用的基于隐式反馈的推荐模型具有更高的准确率和召回率,特别是针对大规模数据集能够有效缓解矩阵稀疏性问题和维数灾难问题,提高推荐性能。     

融合结构邻居和语义邻居的解耦图对比学习推荐模型

基于GCN的协同过滤模型在推荐领域取得了较好的效果,但现有的图协同过滤学习方法通常不区分用户和项目的交互关系,不易挖掘用户行为的潜在意图.因此,提出了一种融合结构邻居和语义邻居的解耦图对比学习推荐模型.首先,将用户和项目嵌入投影到独立空间进行意图解耦;其次,在图传播阶段,依据用户和项目的意图特征挖掘其潜在语义邻居,根据意图相似性对结构邻居和语义邻居进行解耦表征学习,生成用户和项目的完整高阶表示.在对比学习阶段,对节点进行随机扰动并生成对比视图,构建结构和语义的对比学习任务;最后,根据多任务策略,对监督任务和对比学习任务进行联合优化.在真实数据集Yelp2018 和Amazon-Book上的实验表明,提出的模型相比最优基准模型NCL在两个数据集上的Recall@20 指标提高了 7.54%、5.65%,NDCG@20 指标提高了8.57%、6.28%.     

SCMF:一种融合多源数据的软约束矩阵分解推荐算法

数据稀疏是推荐系统面临的主要挑战之一。近年来,多源数据融合为解决数据稀疏问题提供了新思路。然而,现有方法大多假设对象在不同数据源中具有相同的表示,这种硬约束方式无法刻画对象在不同数据源中的差异性。该文提出一种基于软约束矩阵分解的推荐算法,通过约束不同数据源中对象的隐因子向量,能够同时刻画同一对象表示的共性及其在不同数据源中的差异性。在两个数据集上的实验表明,该文提出的软约束矩阵分解算法在准确率方面优于现有的单数据源推荐算法和多源数据硬约束融合推荐算法,可以有效解决推荐系统面临的数据稀疏问题。     

聚类与协同过滤相结合的隐式推荐系统

针对目前大多推荐系统中使用的协同过滤算法都需要有显示的用户反馈的问题,提出一种在隐式反馈推荐系统中使用聚类与矩阵分解技术相结合的方法,为用户提供更好地推荐结果。其结果是由基于用户历史购买记录的隐式反馈产生的,不需任何显式反馈提供的数据。采用高维的、无参数的分裂层次聚类技术产生聚类结果,根据聚类的结果为每个用户提供高兴趣度的个性化推荐。实验结果表明,在隐式反馈的情况下该方法也能有效获得用户偏好,产生大量的高准确度推荐。     

Collaborative filtering based on information-theoretic co-clustering

Collaborative filtering is one of the most popular recommendation techniques, which provides personalised recommendations based on users’ tastes. In spite of its huge success, it suffers from a range of problems, the most fundamental being that of data sparsity. Sparsity in ratings makes the formation of inaccurate neighbourhood, thereby resulting in poor recommendations. To address this issue, in this article, we propose a novel collaborative filtering approach based on information-theoretic co-clustering. The proposed approach computes two types of similarities: cluster preference and rating, and combines them. Based on the combined similarity, the user-based and item-based approaches are adopted, respectively, to obtain individual predictions for an unknown target rating. Finally, the proposed approach fuses these resultant predictions. Experimental results show that the proposed approach is superior to existing alternatives.     

Studying Recommendation Algorithms by Graph Analysis

We present a novel framework for studying recommendation algorithms in terms of the 'jumps' that they make to connect people to artifacts. This approach emphasizes reachability via an algorithm within the implicit graph structure underlying a recommender dataset and allows us to consider questions relating algorithmic parameters to properties of the datasets. For instance, given a particular algorithm 'jump,' what is the average path length from a person to an artifact? Or, what choices of minimum ratings and jumps maintain a connected graph? We illustrate the approach with a common jump called the 'hammock' using movie recommender datasets.