一种解决新项目冷启动问题的推荐算法

推荐系统作为缓解信息过载问题的有效方法之一,在社交媒体中的作用日趋重要.但是,新项目冷启动和新用户冷启动问题是推荐技术面临的难题.为了解决新项目冷启动问题,提出了用户时间权重信息概念,该定义考虑到了用户评价时间与项目发布时间的时间间隔,根据用户时间权重值的大小,可以判断该用户是积极用户还是消极用户,以及用户对新项目的偏爱程度;利用三分图的形式来描述用户-项目-标签、用户-项目-属性之间的关系.在充分考虑用户、标签、项目属性、时间等信息基础上,获得个性化的预测评分值公式,提出了推荐算法.实验结果表明:所提出的方法能够实现满足不同用户、不同偏好的个性化推荐,在为用户推荐到合适项目的同时还能带来惊喜.比较实验说明,所提出的方法推荐准确度高,推荐新颖度高.交叉验证实验结果表明:该方法在解决推荐算法中的新项目冷启动问题上,无论是在推荐的准确度还是推荐项目的新颖度上都是有效的.     

一种提高推荐多样性的概率选择模型

传统的推荐算法多以优化推荐列表的精确度为目标,而忽略了推荐算法的另一个重要指标——多样性。提出了一种新的提高推荐列表多样性的方法。该方法将列表生成步骤转换为N次概率选择过程,每次概率选择通过两个步骤完成:类型选择与项目选择。在类型选择中,引入项目的类型信息,根据用户对不同项目类型的喜好计算概率矩阵,并依照该概率矩阵选择一个类型;在项目选择中,根据项目的预测评分、项目的历史流行度、项目的推荐流行度3个因素重新计算项目的最终得分,选择得分最高的项目推荐给用户。通过阈值TR来调节多样性与精确度之间的折中。最后,通过对比实验证明了该方法的有效性。     

Integrating Collaborative and Reclusive Methods for Effective Recommendations: A Fuzzy Bayesian Approach

Collaborative filtering (CF), the most successful and widely used technique, recommends items based on the preferences of similar users. The main potentials of CF are its cross‐genre recommendation ability, and that it is completely independent of representation of the items being recommended. However, CF suffers from sparsity and cold start problems. On the other hand, a highly effective variant of content‐based filtering (CBF), reclusive methods (RMs) based on the preference of the single individual for whom recommendations to be made, provides a methodology that considers uncertainty and the multivalued nature of item features as well as user preferences in a content‐based framework using fuzzy logic approaches. The adoption of RM paradigm has several advantages when compared to CF such as sparsity and new item problem, but it suffers from overspecialization and limited content analysis. In view of the complementary nature of CF and RM, we develop a hybrid recommender system (RS) that helps in alleviating aforementioned problems in each approach. First, we propose fuzzy naïve Bayesian classifier based CF (FNB‐CF) and RM (FNB‐RM) for handling correlation‐based similarity problems. To overcome individual weaknesses of FNB‐CF and FNB‐RM, we develop a hybrid RS, FNB‐CF‐RM. Effectiveness of our proposed hybrid RS is demonstrated through experimental results using the MovieLens and IMDb data sets.     

融合评分矩阵和评论文本深度神经网络推荐模型

针对个性化推荐中用户评分矩阵数据集稀疏,用户和项目描述信息未充分利用的问题,提出融合评分矩阵和评论文本的深度神经网络推荐模型(deep neural network recommendation model,DeepRec)。首先将通过数据预处理得到的用户偏好特征和项目属性特征的文本集合分别输入到卷积神经网络进行训练,得到用户和项目的深层次非线性特征,同时将评分矩阵输入多层感知机得到用户偏好隐表示,并对两种模型提取的用户偏好隐表示进行融合;其次利用多层感知机建模用户和项目隐表示对用户进行个性化推荐;最后基于三组数据集以均方根误差为评估指标进行对比实验。结果表明DeepRec的预测误差更低,有效提高了推荐的精准度。     

A framework for tag-aware recommender systems

In social tagging system, a user annotates a tag to an item. The tagging information is utilized in recommendation process. In this paper, we propose a hybrid item recommendation method to mitigate limitations of existing approaches and propose a recommendation framework for social tagging systems. The proposed framework consists of tag and item recommendations. Tag recommendation helps users annotate tags and enriches the dataset of a social tagging system. Item recommendation utilizes tags to recommend relevant items to users. We investigate association rule, bigram, tag expansion, and implicit trust relationship for providing tag and item recommendations on the framework. The experimental results show that the proposed hybrid item recommendation method generates more appropriate items than existing research studies on a real-world social tagging dataset.     

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。     

A Review of Ontology‐Based Tag Recommendation Approaches

Tag recommender schemes suggest related tags for an untagged resource and better tag suggestions to tagged resources. Tagging is very important if the user identifies the tag that is more precise to use in searching interesting blogs. There is no clear information regarding the meaning of each tag in a tagging process. An user can use various tags for the same content, and he can also use new tags for an item in a blog. When the user selects tags, the resultant metadata may comprise homonyms and synonyms. This may cause an improper relationship among items and ineffective searches for topic information. The collaborative tag recommendation allows a set of freely selected text keywords as tags assigned by users. These tags are imprecise, irrelevant, and misleading because there is no control over the tag assignment. It does not follow any formal guidelines to assist tag generation, and tags are assigned to resources based on the knowledge of the users. This causes misspelled tags, multiple tags with the same meaning, bad word encoding, and personalized words without common meaning. This problem leads to miscategorization of items, irrelevant search results, wrong prediction, and their recommendations. Tag relevancy can be judged only by a specific user. These aspects could provide new challenges and opportunities to its tag recommendation problem. This paper reviews the challenges to meet the tag recommendation problem. A brief comparison between existing works is presented, which we can identify and point out the novel research directions. The overall performance of our ontology‐based recommender systems is favorably compared to other systems in the literature.     

基于加权三部图的协同过滤推荐算法

针对基于用户的协同过滤算法推荐结果过度集中在热门物品,导致多样性和新颖性较低、覆盖率较小的问题,文中提出基于加权三部图的协同过滤推荐算法.在分析数据稀疏和附加信息较少的基础上引入标签信息,可同时反映用户兴趣和物品属性,利用用户、物品和标签三元关系构建三部图.通过三部图网络映射到单模网络的方法获得用户偏好度,构建用户偏好度加权的三部图模型.根据热传导方法在加权三部图上进行资源重分配,挖掘更多的相似关系,利用协同过滤框架预测评分并进行推荐.在真实数据集上的实验表明,文中算法可较好地挖掘长尾物品,实现个性化推荐.     

基于标签信息特征相似性的协同过滤个性化推荐

标签推荐系统旨在利用标签数据为用户提供个性化推荐。已有的基于标签的推荐方法往往忽视了用户和资源本身的特征,而且在相似性度量时仅针对项目相似性或用户相似性进行计算,并未充分考虑二者之间的有效融合,推荐结果的准确性较低。为了解决上述问题,将标签信息融入到结合用户相似性和项目相似性的协同过滤中,提出融合标签特征与相似性的协同过滤个性化推荐方法。该方法在充分考虑用户、项目以及标签信息的基础上,利用二维矩阵来定义用户-标签以及标签-项目之间的行为。构建用户和项目的标签特征表示,通过基于标签特征的相似性度量方法计算用户相似性和项目相似性。基于用户标签行为和用户与项目的相似性线性组合来预测用户对项目的偏好值,并根据预测偏好值排序,生成最终的推荐列表。在Last.fm数据集上的实验结果表明,该方法能够提高推荐的准确度,满足用户的个性化需求。     

一种融合信任和项目卷积描述信息的PMF算法

针对现有概率矩阵分解(PMF)技术的个性化推荐系统在采用社交网络中信任信息时常常忽视项目相关描述文档信息的问题,提出一种融合用户信任和通过卷积网络以获取项目描述等信息的PMF模型.首先,利用用户偏好信息和行为轨迹信息构建一种新的信任网络;然后,通过卷积神经网络从项目描述文档中提取项目潜在的特征向量;最后,在概率矩阵分解过程中同时利用评分数据、信任网络中用户的信任信息和项目的描述信息,计算用户和项目的潜在特征向量以预测评分并进行个性化推荐.为验证算法的有效性,选择3种算法在4个数据集上进行对比,实验结果表明所提出的算法在推荐精确度和鲁棒性方面优于其他3种算法.     

标签推荐方法研究综述

随着互联网信息的爆炸式增长,标签(由用户指定用来描述项目的关键词)在互联网信息检索领域中变得越来越重要.为在线内容赋予合适的标签,有利于更高效的内容组织和内容消费.而标签推荐通过辅助用户进行打标签的操作,极大地提升了标签的质量,标签推荐也因此受到了研究者们的广泛关注.总结出标签推荐任务的三大特性,即项目内容的多样性、标...     

基于TD-FP-growth的模糊关联规则挖掘算法

提出一种基于TD_FP-growth的模糊关联规则挖掘算法.首先,使用3种t-模算子以及由其产生的蕴涵算子计算模糊频繁项的支持度和规则的蕴涵度,产生的关联规则能表示模糊项间的确定性和渐近性逻辑语义;然后,以事务的惟一标识为键值,散列存储每个事务相对FP-tree中每个结点所表示模糊项的隶属度,使TD-FP-growth适用于模糊频繁项的挖掘,并分析了算法的时间和空间复杂度;最后,实验结果表明该算法比基于apriori的模糊频繁项挖掘算法在时间方面更加有效.

Abstract：

An algorithm based on TD-FP-growth is proposed for mining fuzzy association rule, which uses three kinds of t-norm operator to calculate the support degree of fuzzy frequent items, and adopts corresponding implication operator to measure implication degree of fuzzy association rule.The association rule mined by the algorithm can express the logic semantic of graduality and certainty between fuzzy items.Each transaction's membership degree versus fuzzy item denoted by FP-tree's node is stored by hash technology, and each transaction's identifier is regarded as key value, which adapts TD-FP-growth to mine fuzzy frequent items.The time and space complexity of the algorithm are analyzed.The experimental results show that the algorithm is more effective than the fuzzy frequent item mining algorithm based on apriori in term of time.     

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.     

融合循环知识图谱和协同过滤电影推荐算法

推荐系统对筛选有效信息和提高信息获取效率具有重大的意义。传统的推荐系统会面临数据稀松和冷启动等问题。利用外部评分和物品内涵知识相结合,提出一种基于循环知识图谱和协同过滤的电影推荐模型--RKGE-CF。在充分考虑物品、用户、评分之间的相关性后,利用基于物品和用户的协同过滤进行Top-[K]推荐;将物品的外部附加数据和用户偏好数据加入知识图谱,提取实体相互之间的依赖关系,构建用户和物品之间的交互信息,以便揭示实体与关系之间的语义,帮助理解用户兴趣;将多种推荐结果按不同方法融合进行对比;模型训练时使用多组不同的负样本作为对比,以优化模型;最后利用真实电影数Movielens和IMDB映射连接成新数据集进行测试。实验结果证明该模型对于推荐效果的准确率有显著的提升,同时能更好地解释推荐背后的原因。     

联合学习用户端和项目端知识图谱的个性化推荐

如何在已有的用户行为和辅助信息的基础上准确建模用户的偏好非常重要。在各种辅助信息中,知识图谱(Know-ledge Graph,KG)作为一种新型辅助信息,其节点和边包含了丰富的结构信息和语义信息,近年来受到了越来越多研究者的关注。大量研究表明,在个性化推荐中引入知识图谱可以有效地提高推荐的性能,并增强推荐的合理性和可解释性。然而,现有的方法要么是在KG上探索每个用户-项目交互对(user-item)的独立子路径,要么使用图表示学习的方法在KG中分别学习目标用户(user)或项目(item)的表示,虽然都取得了一定的效果,但是前者没有充分捕获用户-项目(user-item)在KG上的结构信息,后者在产生嵌入(embedding)表示的过程中忽略了user和item的相互影响。为了弥补上述方法的不足,提出了一种联合学习用户端和项目端知识图谱(User-end and Item-end Knowledge Graph,UIKG)的新模型。该模型通过挖掘用户和项目在各自KG中的关联属性信息,并通过联合学习有效地捕获用户的个性化偏好与项目之间的关联性。具体的操作步骤是,利用基于图卷积神经网络的方法从用户知识图谱中学习用户表示向量,再将用户表示向量引入项目知识图谱中联合学习得到项目表示向量,实现用户端KG和项目端KG的无缝统一,最后通过多层感知器进行偏好预测,得到用户对项目的偏好概率,从而更有效地挖掘KG中的高阶结构信息和语义信息来捕获用户的个性化偏好。在公开数据集上的实验结果表明,与基线方法相比,UIKG在Recall@K指标上提高了2.5%~13.6%,在AUC和F1指标上提高了0.4%~5.8%。     

Subprofile-aware diversification of recommendations

A user of a recommender system is more likely to be satisfied by one or more of the recommendations if each individual recommendation is relevant to her but additionally if the set of recommendations is diverse. The most common approach to recommendation diversification uses re-ranking: the recommender system scores a set of candidate items for relevance to the user; it then re-ranks the candidates so that the subset that it will recommend achieves a balance between relevance and diversity. Ordinarily, we expect a trade-off between relevance and diversity: the diversity of the set of recommendations increases by including items that have lower relevance scores but which are different from the items already in the set. In early work, the diversity of a set of recommendations was given by the average of their distances from one another, according to some semantic distance metric defined on item features such as movie genres. More recent intent-aware diversification methods formulate diversity in terms of coverage and relevance of aspects. The aspects are most commonly defined in terms of item features. By trying to ensure that the aspects of a set of recommended items cover the aspects of the items in the user’s profile, the level of diversity is more personalized. In offline experiments on pre-collected datasets, intent-aware diversification using item features as aspects sometimes defies the relevance/diversity trade-off: there are configurations in which the recommendations exhibits increases in both relevance and diversity. In this paper, we present a new form of intent-aware diversification, which we call SPAD (Subprofile-Aware Diversification), and a variant called RSPAD (Relevance-based SPAD). In SPAD, the aspects are not item features; they are subprofiles of the user’s profile. We present and compare a number of different ways to extract subprofiles from a user’s profile. None of them is defined in terms of item features. Therefore, SPAD is useful even in domains where item features are not available or are of low quality. On three pre-collected datasets from three different domains (movies, music artists and books), we compare SPAD and RSPAD to intent-aware methods in which aspects are item features. We find on these datasets that SPAD and RSPAD suffer even less from the relevance/diversity trade-off: across all three datasets, they increase both relevance and diversity for even more configurations than other approaches to diversification. Moreover, we find that SPAD and RSPAD are the most accurate systems across all three datasets.

     

Learning to recommend similar items from human judgments

Similar item recommendations—a common feature of many Web sites—point users to other interesting objects given a currently inspected item. A common way of computing such recommendations is to use a similarity function, which expresses how much alike two given objects are. Such similarity functions are usually designed based on the specifics of the given application domain. In this work, we explore how such functions can be learned from human judgments of similarities between objects, using two domains of “quality and taste”—cooking recipe and movie recommendation—as guiding scenarios. In our approach, we first collect a few thousand pairwise similarity assessments with the help of crowdworkers. Using these data, we then train different machine learning models that can be used as similarity functions to compare objects. Offline analyses reveal for both application domains that models that combine different types of item characteristics are the best predictors for human-perceived similarity. To further validate the usefulness of the learned models, we conducted additional user studies. In these studies, we exposed participants to similar item recommendations using a set of models that were trained with different feature subsets. The results showed that the combined models that exhibited the best offline prediction performance led to the highest user-perceived similarity, but also to recommendations that were considered useful by the participants, thus confirming the feasibility of our approach.

     

Affect analysis of text using fuzzy semantic typing

We propose a novel, convenient fusion of natural language processing and fuzzy logic techniques for analyzing the affect content in free text. Our main goals are fast analysis and visualization of affect content for decision making. The main linguistic resource for fuzzy semantic typing is the fuzzy-affect lexicon, from which other important resources, the fuzzy thesaurus and affect category groups, are generated. Free text is tagged with affect categories from the lexicon and the affect categories' centralities and intensities are combined using techniques from fuzzy logic to produce affect sets: fuzzy sets representing the affect quality of a document. We show different aspects of affect analysis using news content and movie reviews. Our experiments show a good correspondence between affect sets and human judgments of affect content. We ascribe this to the representation of ambiguity in our fuzzy affect lexicon and the ability of fuzzy logic to deal successfully with the ambiguity of words in a natural language     

基于SDAE特征表示的协同主题回归推荐模型

为解决推荐系统中的冷启动问题,在协同主题回归CTR模型的基础上引入堆叠去噪自编码器SDAE深度学习网络,用于学习用户辅助信息的隐表示,建立SDAE-CTR模型。模型应用2层SDAE网络,以用户信息为网络输入量,将编码过程获得的用户辅助信息的隐表示和解码过程获得的输入近似表示为网络的双输出量,最小化用户辅助信息和近似表示的差值来确定最优隐表示。模型融合用户-项目评分矩阵(冷启动条件无评分)、项目内容信息和用户辅助信息实现用户对未评分项目的评分预测,并在LastFM、Book Crossing和MovieLens数据集上从推荐准确度、新颖性和用户冷启动条件下的推荐效果等3方面对SDAE-CTR模型和CTR模型进行比较。结果表明,SDAE-CTR模型在冷启动或非冷启动的条件下,推荐效果都要优于CTR模型的,虽然新颖性较CTR模型稍微逊色一些,但理论上在合理的范围内,总体上SDAE-CTR模型表现较优。