No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Social Weak-tie Assisted Cross-domain Short Video Recommendation
ResearchGate has not been able to resolve any citations for this publication.
Online platforms can be divided into information-oriented and social-oriented domains. The former refers to forums or E-commerce sites that emphasize user-item interactions, like Trip.com and Amazon; whereas the latter refers to social networking services (SNSs) that have rich user-user connections, such as Facebook and Twitter. Despite their heterogeneity, these two domains can be bridged by a few overlapping users, dubbed as bridge users. In this work, we address the problem of cross-domain social recommendation, i.e., recommending relevant items of information domains to potential users of social networks. To our knowledge, this is a new problem that has rarely been studied before.
Existing cross-domain recommender systems are unsuitable for this task since they have either focused on homogeneous information domains or assumed that users are fully overlapped. Towards this end, we present a novel Neural Social Collaborative Ranking (NSCR) approach, which seamlessly sews up the user-item interactions in information domains and user-user connections in SNSs. In the information domain part, the attributes of users and items are leveraged to strengthen the embedding learning of users and items. In the SNS part, the embeddings of bridge users are propagated to learn the embeddings of other non-bridge users. Extensive experiments on two real-world datasets demonstrate the effectiveness and rationality of our NSCR method.
With the explosive growth of online social networks, it is now well understood that social information is highly helpful to recommender systems. Social recommendation methods are capable of battling the critical cold-start issue, and thus can greatly improve prediction accuracy. The main intuition is that through trust and influence, users are more likely to develop affinity toward items consumed by their social ties. Despite considerable work in social recommendation, little attention has been paid to the important distinctions between strong and weak ties, two well-documented notions in social sciences. In this work, we study the effects of distinguishing strong and weak ties in social recommendation. We use neighbourhood overlap to approximate tie strength and extend the popular Bayesian Personalized Ranking (BPR) model to incorporate the distinction of strong and weak ties. We present an EM-based algorithm that simultaneously classifies strong and weak ties in a social network w.r.t. optimal recommendation accuracy and learns latent feature vectors for all users and all items. We conduct extensive empirical evaluation on four real-world datasets and demonstrate that our proposed method significantly outperforms state-of-the-art pairwise ranking methods in a variety of accuracy metrics.
Cross-domain recommender systems are usually able to suggest items, which are not in the same domain, where users provided ratings. For this reason, cross-domain recommendation has attracted more and more attention in recent years. However, most studies propose to make cross-domain recommendation in the scenario, where there are common ratings between different domains. The scenario without common ratings is seldom considered. In this paper, we propose a novel method to solve the cross-domain recommendation problem in such a scenario. We first apply trust relations to the cross-domain scenario for predicting coarse ratings pertaining to cross-domain items. Then, we build a new rating matrix, including known ratings and predicted ratings of items from different domains, and transform a user-item matrix into an item-item association matrix. Finally, we compute the similarities of items belonging to different domains and use item-based collaborative filtering to generate recommendations. Through relevant experiments on a real-world data set, we compare our method to a trust-aware recommendation method and demonstrate its effectiveness in terms of prediction accuracy, recall, and coverage.
Collaborative filtering (CF) has been the most popular approach for recommender systems in recent years. In order to analyze the property of a ranking-oriented CF algorithm directly and be able to improve its performance, this paper investigates the ranking-oriented CF from the perspective of loss function. To gain the insight into the popular bias problem, we also study the tendency of a CF algorithm in recommending the most popular items, and show that such popularity tendency can be adjusted through setting different parameters in our models. After analyzing two state-of-the-art algorithms, we propose in this paper two models using the generalized logistic loss function and the hinge loss function, respectively. The experimental results show that the proposed methods outperform the state-of-the-art algorithms on two real data sets.
In the context of social media services, data shortage has severally hindered accurate user modeling and practical personalized applications. This paper is motivated to leverage the user data distributed in disparate online social networks (OSN) to make up for the data shortage in user modeling, which we refer to as “cross-OSN user modeling.” Generally, the data that the same user distributes in different OSNs consist of both behavior data (i.e., interaction with multimedia items) and social data (i.e., interaction between users). This paper focuses on the following two challenges: 1) how to aggregate the users’ cross-OSN interactions with multimedia items of the same modality, which we call cross-OSN homogeneous behaviors, and 2) how to integrate users’ cross-OSN social data with behavior data. Our proposed solution to address the challenges consist of two corresponding components as follows. 1) Homogeneous behavior quantification, where homogeneous user behaviors are quantified based on their importance in reflecting user preferences. After quantification, the examined cross-OSN user behaviors are aggregated to construct a unified user-item interaction matrix. 2) Local social regularization, where the cross-OSN social data is integrated as regularization in matrix factorization-based user modeling at local topic level. The proposed cross-OSN user modeling solution is evaluated in the application of personalized video recommendation. Carefully designed experiments on self-collected Google+ and YouTube datasets have validated its effectiveness and the advantage over single-OSN-based methods.
Recommender systems can suffer from data sparsity and cold start issues. However, social networks, which enable users to build relationships and create different types of items, present an unprecedented opportunity to alleviate these issues. In this paper, we represent a social network as a star-structured hybrid graph centered on a social domain, which connects with other item domains. With this innovative representation, useful knowledge from an auxiliary domain can be transferred through the social domain to a target domain. Various factors of item transferability, including popularity and behavioral consistency, are determined. We propose a novel Hybrid Random Walk (HRW) method, which incorporates such factors, to select transferable items in auxiliary domains, bridge cross-domain knowledge with the social domain, and accurately predict user-item links in a target domain. Extensive experiments on a real social dataset demonstrate that HRW significantly outperforms existing approaches.
Traditional personalized video recommendation methods focus on utilizing user profile or user history behaviors to model user interests, which follows a static strategy and fails to capture the swift shift of the short-term interests of users. According to our cross-platform data analysis, the information emergence and propagation is faster in social textual stream-based platforms than that in multimedia sharing platforms at micro user level. Inspired by this, we propose a dynamic user modeling strategy to tackle personalized video recommendation issues in the multimedia sharing platform YouTube, by transferring knowledge from the social textual stream-based platform Twitter. In particular, the cross-platform video recommendation strategy is divided into two steps. (1) Real-time hot topic detection: the hot topics that users are currently following are extracted from users' tweets, which are utilized to obtain the related videos in YouTube. (2) Time-aware video recommendation: for the target user in YouTube, the obtained videos are ranked by considering the user profile in YouTube, time factor, and quality factor to generate the final recommendation list. In this way, the short-term (hot topics) and long-term (user profile) interests of users are jointly considered. Carefully designed experiments have demonstrated the advantages of the proposed method.
One-class collaborative filtering or collaborative ranking with implicit feedback has been steadily receiving more attention, mostly due to the "one-class" characteristics of data in various services, e.g., "like" in Facebook and "bought" in Amazon. Previous works for solving this problem include pointwise regression methods based on absolute rating assumptions and pairwise ranking methods with relative score assumptions, where the latter was empirically found performing much better because it models users' ranking-related preferences more directly. However, the two fundamental assumptions made in the pairwise ranking methods, (1) individual pairwise preference over two items and (2) independence between two users, may not always hold. As a response, we propose a new and improved assumption, group Bayesian personalized ranking (GBPR), via introducing richer interactions among users. In particular, we introduce group preference, to relax the aforementioned individual and independence assumptions. We then design a novel algorithm correspondingly, which can recommend items more accurately as shown by various ranking-oriented evaluation metrics on four real-world datasets in our experiments.
The effectiveness of existing top-N recommendation methods decreases as the sparsity of the datasets increases. To alleviate this problem, we present an item-based method for generating top-N recommendations that learns the item-item similarity matrix as the product of two low dimensional latent factor matrices. These matrices are learned using a structural equation modeling approach, wherein the value being estimated is not used for its own estimation. A comprehensive set of experiments on multiple datasets at three different sparsity levels indicate that the proposed methods can handle sparse datasets effectively and outperforms other state-of-the-art top-N recommendation methods. The experimental results also show that the relative performance gains compared to competing methods increase as the data gets sparser.
Recommendation systems have received great attention for their commercial value in today's online business world. However, most recommendation systems encounter the data sparsity problem and the cold-start problem. To improve recommendation accuracy in this circumstance, additional sources of information about the users and items should be incorporated in recommendation systems. In this paper, we modify the model in Bayesian Probabilistic Matrix Factorization, and propose two recommendation approaches fusing social relations and item contents with user ratings in a novel way. The proposed approach is computationally efficient and can be applied to trust-aware or content-aware recommendation systems with very large dataset. Experimental results on three real world datasets show that our method gets more accurate recommendation results with faster converging speed than other matrix factorization based methods. We also verify our method in cold-start settings, and our method gets more accurate recommendation results than the compared approaches.
The explosive growth of the world-wide-web and the emergence of e-commerce has led to the development of recommender systems---a personalized information filtering technology used to identify a set of items that will be of interest to a certain user. User-based collaborative filtering is the most successful technology for building recommender systems to date and is extensively used in many commercial recommender systems. Unfortunately, the computational complexity of these methods grows linearly with the number of customers, which in typical commercial applications can be several millions. To address these scalability concerns model-based recommendation techniques have been developed. These techniques analyze the user--item matrix to discover relations between the different items and use these relations to compute the list of recommendations.In this article, we present one such class of model-based recommendation algorithms that first determines the similarities between the various items and then uses them to identify the set of items to be recommended. The key steps in this class of algorithms are (i) the method used to compute the similarity between the items, and (ii) the method used to combine these similarities in order to compute the similarity between a basket of items and a candidate recommender item. Our experimental evaluation on eight real datasets shows that these item-based algorithms are up to two orders of magnitude faster than the traditional user-neighborhood based recommender systems and provide recommendations with comparable or better quality.
Online social networks have become an essential part of people's daily life, and an increasing number of users are now using multiple Online Social Networks (OSNs) for different social media services simultaneously. As a result, user's interests and preferences usually distribute in different OSNs. While most of the existing work mainly aggregates the distributed user behaviors or features directly, recently very few efforts have been focused on understanding the cross-OSN association from collective user behaviors. In this paper, we go one step further to consider the dynamic characteristic of user behaviors and propose a dynamic cross-OSN association mining framework. In this framework, dynamic user modeling is first conducted to capture the drift of user interest in each OSN. A session-based factorization method is then proposed to establish the cross-OSN association in a dynamic manner, by progressively updating the derived association each time a new session of data arrives. Based on the derived dynamic association, we finally design a cold-start YouTube video recommendation application, by only utilizing users' behaviors in Twitter. Experiments are conducted using real-world user data from Twitter and YouTube. The results demonstrate the effectiveness of this proposed framework in capturing the underlying association between different OSNs and achieving superior cold-start recommendation performance.
Recommending a ranked list of interesting venues to users based on their preferences has become a key functionality in Location-Based Social Networks (LBSNs) such as Yelp and Gowalla. Bayesian Personalised Ranking (BPR) is a popular pairwise recommendation technique that is used to generate the ranked list of venues of interest to a user, by leveraging the user's implicit feedback such as their check-ins as instances of positive feedback, while randomly sampling other venues as negative instances. To alleviate the sparsity that affects the usefulness of recommendations by BPR for users with few check-ins, various approaches have been proposed in the literature to incorporate additional sources of information such as the social links between users, the textual content of comments, as well as the geographical location of the venues. However, such approaches can only readily leverage one source of additional information for negative sampling. Instead, we propose a novel Personalised Ranking Framework with Multiple sampling Criteria (PRFMC) that leverages both geographical influence and social correlation to enhance the effectiveness of BPR. In particular, we apply a multi-centre Gaussian model and a power-law distribution method, to capture geographical influence and social correlation when sampling negative venues, respectively. Finally, we conduct comprehensive experiments using three large-scale datasets from the Yelp, Gowalla and Brightkite LBSNs. The experimental results demonstrate the effectiveness of fusing both geographical influence and social correlation in our proposed PRFMC framework and its superiority in comparison to BPR-based and other similar ranking approaches. Indeed, our PRFMC approach attains a 37% improvement in MRR over a recently proposed approach that identifies negative venues only from social links.
In recent years, deep neural networks have yielded immense success on speech recognition, computer vision and natural language processing. However, the exploration of deep neural networks on recommender systems has received relatively less scrutiny. In this work, we strive to develop techniques based on neural networks to tackle the key problem in recommendation -- collaborative filtering -- on the basis of implicit feedback. Although some recent work has employed deep learning for recommendation, they primarily used it to model auxiliary information, such as textual descriptions of items and acoustic features of musics. When it comes to model the key factor in collaborative filtering -- the interaction between user and item features, they still resorted to matrix factorization and applied an inner product on the latent features of users and items. By replacing the inner product with a neural architecture that can learn an arbitrary function from data, we present a general framework named NCF, short for Neural network-based Collaborative Filtering. NCF is generic and can express and generalize matrix factorization under its framework. To supercharge NCF modelling with non-linearities, we propose to leverage a multi-layer perceptron to learn the user-item interaction function. Extensive experiments on two real-world datasets show significant improvements of our proposed NCF framework over the state-of-the-art methods. Empirical evidence shows that using deeper layers of neural networks offers better recommendation performance.
Next Point-of-interest (POI) recommendation has become an important task for location-based social networks (LBSNs). However, previous efforts suffer from the high computational complexity and the transition pattern between POIs has not been well studied. In this paper, we propose a two-fold approach for next POI recommendation. First, the preferred next category is predicted by using a third-rank tensor optimized by a Listwise Bayesian Personalized Ranking (LBPR) approach. Specifically we introduce two functions, namely Plackett-Luce model and cross entropy, to generate the likelihood of ranking list for posterior computation. Then POI candidates filtered by the predicated category are ranked based on the spatial influence and category ranking influence. Extensive experiments on two real-world datasets demonstrate the significant improvements of our methods over several state-of-the-art methods.
In recent years, deep neural networks have yielded immense success on speech recognition, computer vision and natural language processing. However, the exploration of deep neural networks on recommender systems has received relatively less scrutiny. In this work, we strive to develop techniques based on neural networks to tackle the key problem in recommendation --- collaborative filtering --- on the basis of implicit feedback.
Although some recent work has employed deep learning for recommendation, they primarily used it to model auxiliary information, such as textual descriptions of items and acoustic features of musics. When it comes to model the key factor in collaborative filtering --- the interaction between user and item features, they still resorted to matrix factorization and applied an inner product on the latent features of users and items.
By replacing the inner product with a neural architecture that can learn an arbitrary function from data, we present a general framework named NCF, short for Neural network-based Collaborative Filtering. NCF is generic and can express and generalize matrix factorization under its framework. To supercharge NCF modelling with non-linearities, we propose to leverage a multi-layer perceptron to learn the user-item interaction function. Extensive experiments on two real-world datasets show significant improvements of our proposed NCF framework over the state-of-the-art methods. Empirical evidence shows that using deeper layers of neural networks offers better recommendation performance.
Group recommendation plays a significant role in today’s social media systems, where users form social groups to receive multimedia content together and interact with each other, instead of consuming the online content individually. Limitations of traditional group recommendation approaches are that: (1) They usually infer group members’ preferences by their historical behaviors, failing to capture inactive users’ preferences from the sparse historical data. (2) Relationship between group members is not studied by these approaches, which fail to capture the inherent personality of members in a group. To address this issue, we propose a social-aware group recommendation framework which jointly utilize both social relationship and social behaviors to not only infer a group’s preference, but also model the tolerance and altruism characteristics of group members. Based on the observation that the following relationship in the online social network reflects common interests of users, we propose a group preference model based on external experts of group members. Furthermore, we model users’ tolerance (willingness to receive content not preferred) and altruism (willingness to receive content preferred by friends). Finally, based on the group preference model, we design recommendation algorithms for users under different social contexts. Experimental results demonstrate the effectiveness of our approach, which significantly improves the recommendation accuracy against traditional approaches, especially in the cases of inactive group members.
Cross-domain data analysis is one of the most important tasks in social multimedia. It has a wide range of real-world applications, including cross-platform event analysis, cross-domain multi-event tracking, cross-domain video recommendation, etc. It is also very challenging because the data have multi-modal and multi-domain properties, and there are no explicit correlations to link different domains. To deal with these issues, we propose a generic Cross-Domain Collaborative Learning (CDCL) framework based on non-parametric Bayesian dictionary learning model for cross-domain data analysis. In the proposed CDCL model, it can make use of the shared domain priors and modality priors to collaboratively learn the data's representations by considering the domain discrepancy and the multi-modal property. As a result, our CDCL model can effectively explore the virtues of different information sources to complement and enhance each other for cross-domain data analysis. To evaluate the proposed model, we apply it for two different applications: cross-platform event recognition and cross-network video recommendation. The extensive experimental evaluations well demonstrate the effectiveness of the proposed algorithm for cross-domain data analysis.
Pairwise learning-to-rank algorithms have been shown to allow recommender systems to leverage unary user feedback. We propose Multi-feedback Bayesian Personalized Ranking (MF-BPR), a pairwise method that exploits different types of feedback with an extended sampling method. The feedback types are drawn from different "channels", in which users interact with items (e.g., clicks, likes, listens, follows, and purchases). We build on the insight that different kinds of feedback, e.g., a click versus a like, reflect different levels of commitment or preference. Our approach differs from previous work in that it exploits multiple sources of feedback simultaneously during the training process. The novelty of MF-BPR is an extended sampling method that equates feedback sources with "levels" that reflect the expected contribution of the signal. We demonstrate the effectiveness of our approach with a series of experiments carried out on three datasets containing multiple types of feedback. Our experimental results demonstrate that with a right sampling method, MF-BPR outperforms BPR in terms of accuracy. We find that the advantage of MF-BPR lies in its ability to leverage level information when sampling negative items.
Online video sharing sites are increasingly encouraging their users to connect to the social network venues such as Facebook and Twitter, with goals to boost user interaction and better disseminate the high-quality video content. This in turn provides huge possibilities to conduct cross-network collaboration for personalized video recommendation. However, very few efforts have been devoted to leveraging users' social media profiles in the auxiliary network to capture and personalize their video preferences, so as to recommend videos of interest. In this article, we propose a unified YouTube video recommendation solution by transferring and integrating users' rich social and content information in Twitter network. While general recommender systems often suffer from typical problems like cold-start and data sparsity, our proposed recommendation solution is able to effectively learn from users' abundant auxiliary information on Twitter for enhanced user modeling and well address the typical problems in a unified framework. In this framework, two stages are mainly involved: (1) auxiliary-network data transfer, where user preferences are transferred from an auxiliary network by learning cross-network knowledge associations; and (2) cross-network data integration, where transferred user preferences are integrated with the observed behaviors on a target network in an adaptive fashion. Experimental results show that the proposed cross-network collaborative solution achieves superior performance not only in terms of accuracy, but also in improving the diversity and novelty of the recommended videos.
Online shopping has become the buzzword in this information age. Users want to purchase the best possible item and services at the shortest span of time. In this information age Recommender system is a very useful tool, because it has the capability of filtering the information according to user interest and provide personalized suggestion. One of the major drawbacks of the classical recommender system is that, they deal with the only single domain. In real world scenario domains could be related to each other by some common information. There are many approaches available for cross domain recommendation, but they are not able to provide better accuracy of high dimensional data and these approaches are suffering from data sparsity problem. In this paper, we deal with cross domain recommendation where we exploit knowledge from auxiliary domains (e.g., movies) which contains additional user preference data to improve recommendation on the target domain (e.g., books). In order to achieve a high level of accuracy, we make use of semantic similarity measure of common information by which domains are related and Tensor decomposition to exploiting the latent factor for high dimensional data. Tensor decomposition with semantic similarity is used for making cross domain recommendation where in the data sparsity problem is avoided by normalizing and clustering the data in auxiliary domain. We provide experimental results on real world data sets and compared our proposed method with other similar approaches based on hit ratio and the results show that we achieve a better hit ratio.
Recommendation systems adopt various techniques to recommend ranked lists of items to help users in identifying items that fit their personal tastes best. Among various recommendation algorithms, user and item-based collaborative filtering methods have been very successful in both industry and academia. More recently, the rapid growth of the Internet and E-commerce applications results in great challenges for recommendation systems as the number of users and the amount of available online information have been growing too fast. These challenges include performing high quality recommendations per second for millions of users and items, achieving high coverage under the circumstance of data sparsity and increasing the scalability of recommendation systems. To obtain higher quality recommendations under the circumstance of data sparsity, in this paper, we propose a novel method to compute the similarity of different users based on the side information which is beyond user-item rating information from various online recommendation and review sites. Furthermore, we take the special interests of users into consideration and combine three types of information (users, items, user-items) to predict the ratings of items. Then FUIR, a novel recommendation algorithm which fuses user and item information, is proposed to generate recommendation results for target users. We evaluate our proposed FUIR algorithm on three data sets and the experimental results demonstrate that our FUIR algorithm is effective against sparse rating data and can produce higher quality recommendations.
Both researchers and practitioners in the field of collaborative filtering have shown keen interest to user behaviors of the “one-class” feedback form such as transactions in e-commerce and “likes” in social networks. This recommendation problem is termed as one-class collaborative filtering (OCCF). In most of the previous work, a pairwise preference assumption called Bayesian personalized ranking (BPR) was empirically proved to be able to exploit such one-class data well. In one of the most recent work, an upgraded model called group preference based BPR (GBPR) leverages the group preference and obtains better performance.
In this paper, we go one step beyond GBPR, and propose a new and generic assumption, i.e., group Bayesian personalized ranking with rich interactions (GBPR⁺). In our GBPR⁺, we adopt a set of items instead of one single item as used in GBPR, which is expected to introduce rich interactions. GBPR is a special case of our GPBR⁺ when the item set contains only one single item. We study the empirical performance of our GBPR⁺ with several state-of-the-art methods on four real-world datasets, and find that our GPBR⁺ can generate more accurate recommendations.
Model-based approach to collaborative filtering (CF), such as latent factor models, has improved both accuracy and efficiency of predictions on large and sparse dataset. However, most existing methods still face two major problems: (1) The recommendation results derived from user and item vectors of a set of unobserved factors are lack of explanation; (2) Cold start users and items out of user-item rating matrix cannot be handled accurately. In this paper, we propose a hybrid method for addressing the problems by incorporating content-based information (i.e, users' tags and items' keywords) and social information. The main idea behind our method is to build content association based on three factors-user interest in selected tags, tag-keyword relation and item correlation with extracted keywords, and then recommend the items with high similarity in content to users. Two novel methods-neighbor based approach and 3-factor matrix factorization are proposed for building tag-keyword relation matrix and learning user interest vector for selected tags and item correlation vector for extracted keywords. Besides, we introduce a social regularization term to help shape user interest vector. Analysis shows our method can generate explainable recommendation results with simple descriptions, and experiments on real dataset demonstrate our method improves recommendation accuracy of state-of-the-art CF models for previous users and items with few ratings, as well as cold start users and items with no rating.
Recommending products to users means estimating their preferences for certain items over others. This can be cast either as a problem of estimating the rating that each user will give to each item, or as a problem of estimating users' relative preferences in the form of a ranking. Although collaborative-filtering approaches can be used to identify users who rate and rank products similarly, another source of data that informs us about users' preferences is their set of social connections. Both rating- and ranking-based paradigms are important in real-world recommendation settings, though rankings are especially important in settings where explicit feedback in the form of a numerical rating may not be available. Although many existing works have studied how social connections can be used to build better models for rating prediction, few have used social connections as a means to derive more accurate ranking-based models. Using social connections to better estimate users' rankings of products is the task we consider in this paper. We develop a model, SBPR (Social Bayesian Personalized Ranking), based on the simple observation that users tend to assign higher ranks to items that their friends prefer. We perform experiments on four real-world recommendation data sets, and show that SBPR outperforms alternatives in ranking prediction both in warm- and cold-start settings.
In many application domains of recommender systems, explicit rating information is sparse or non-existent. The preferences of the current user have therefore to be approximated by interpreting his or her behavior, i.e., the implicit user feedback. In the literature, a number of algorithm proposals have been made that rely solely on such implicit feedback, among them Bayesian Personalized Ranking (BPR).
In the BPR approach, pairwise comparisons between the items are made in the training phase and an item i is considered to be preferred over item j if the user interacted in some form with i but not with j. In real-world applications, however, implicit feedback is not necessarily limited to such binary decisions as there are, e.g., different types of user actions like item views, cart or purchase actions and there can exist several actions for an item over time.
In this paper we show how BPR can be extended to deal with such more fine-granular, graded preference relations. An empirical analysis shows that this extension can help to measurably increase the predictive accuracy of BPR on realistic e-commerce datasets.
Although video recommender systems have become the predominant way for people to obtain video information, their performances are far from satisfactory in that (1) the recommended videos are often mismatched with the users' interests and (2) the recommendation results are, in most cases, hardly understandable for users and therefore cannot persuade them to engage. In this paper, we attempt to address the above problems in data representation level, and aim to learn a common attributed representation for users and videos in social media with good interpretability, stability and an appropriate level of granularity. The basic idea is to represent videos with users' social attributes, and represent users with content attributes of videos, such that both users and videos can be represented in a common space concatenated by social attributes and content attributes. The video recommendation problem can then be converted into a similarity matching problem in the common space. However, it is still challenging to balance the roles of social attributes and content attributes, learn such a common representation in sparse user-video interactions and deal with the cold-start problem. In this paper, we propose a REgularized Dual-fActor Regression (REDAR) method based on matrix factorization. In this method, social attributes and content attributes are flexibly combined, and social and content information are effectively exploited to alleviate the sparsity problem. An incremental version of REDAR is designed to solve the cold-start problem. We extensively evaluate the proposed method for video recommendation application in real social network dataset, and the results show that, in most cases, the proposed method can achieve a relative improvement of more than 20% compared to state-of-the-art baseline methods.
Pairwise algorithms are popular for learning recommender systems from implicit feedback. For each user, or more generally context, they try to discriminate between a small set of selected items and the large set of remaining (irrelevant) items. Learning is typically based on stochastic gradient descent (SGD) with uniformly drawn pairs. In this work, we show that convergence of such SGD learning algorithms slows down considerably if the item popularity has a tailed distribution. We propose a non-uniform item sampler to overcome this problem. The proposed sampler is context-dependent and oversamples informative pairs to speed up convergence. An efficient implementation with constant amortized runtime costs is developed. Furthermore, it is shown how the proposed learning algorithm can be applied to a large class of recommender models. The properties of the new learning algorithm are studied empirically on two real-world recommender system problems. The experiments indicate that the proposed adaptive sampler improves the state-of-the art learning algorithm largely in convergence without negative effects on prediction quality or iteration runtime.
This paper investigates the classification of short user-generated videos (UGVs) using the accompanied audio data since short UGVs accounts for a great proportion of the Internet UGVs and many short UGVs are accompanied by single-category soundtracks. We define seven types of UGVs corresponding to seven audio categories respectively. We also investigate three modeling approaches for audio feature representation, namely, single Gaussian (1G), Gaussian mixture (GMM) and Bag-of-Audio-Word (BoAW) models. Then using Support Vector Machine (SVM) with three different distance measurements corresponding to three feature representations, classifiers are trained to categorize the UGVs. The accompanying evaluation results show that these approaches are effective for categorizing the short UGVs based on their audio track. Experimental results show that a GMM representation with approximated Bhattacharyya distance (ABD) measurement produces the best performance, and BoAW representation with chi_square kernel also reports comparable results.
Item recommendation is the task of predicting a personalized ranking on a set
of items (e.g. websites, movies, products). In this paper, we investigate the
most common scenario with implicit feedback (e.g. clicks, purchases). There are
many methods for item recommendation from implicit feedback like matrix
factorization (MF) or adaptive knearest-neighbor (kNN). Even though these
methods are designed for the item prediction task of personalized ranking, none
of them is directly optimized for ranking. In this paper we present a generic
optimization criterion BPR-Opt for personalized ranking that is the maximum
posterior estimator derived from a Bayesian analysis of the problem. We also
provide a generic learning algorithm for optimizing models with respect to
BPR-Opt. The learning method is based on stochastic gradient descent with
bootstrap sampling. We show how to apply our method to two state-of-the-art
recommender models: matrix factorization and adaptive kNN. Our experiments
indicate that for the task of personalized ranking our optimization method
outperforms the standard learning techniques for MF and kNN. The results show
the importance of optimizing models for the right criterion.
Relational learning can be used to augment one data source with other
correlated sources of information, to improve predictive accuracy. We frame a
large class of relational learning problems as matrix factorization problems,
and propose a hierarchical Bayesian model. Training our Bayesian model using
random-walk Metropolis-Hastings is impractically slow, and so we develop a
block Metropolis-Hastings sampler which uses the gradient and Hessian of the
likelihood to dynamically tune the proposal. We demonstrate that a predictive
model of brain response to stimuli can be improved by augmenting it with side
information about the stimuli.
Relational learning is concerned with predicting unknown values of a relation, given a database of entities and ob- served relations among entities. An example of relational learning is movie rating prediction, where entities could in- clude users, movies, genres, and actors. Relations encode users' ratings of movies, movies' genres, and actors' roles in movies. A common prediction technique given one pairwise relation, for example a #users #movies ratings matrix, is low-rank matrix factorization. In domains with multiple relations, represented as multiple matrices, we may improve predictive accuracy by exploiting information from one re- lation while predicting another. To this end, we propose a collective matrix factorization model: we simultaneously factor several matrices, sharing parameters among factors when an entity participates in multiple relations. Each rela- tion can have a dierent value type and error distribution; so, we allow nonlinear relationships between the parameters and outputs, using Bregman divergences to measure error. We extend standard alternating projection algorithms to our model, and derive an ecient Newton update for the pro- jection. Furthermore, we propose stochastic optimization methods to deal with large, sparse matrices. Our model gen- eralizes several existing matrix factorization methods, and therefore yields new large-scale optimization algorithms for these problems. Our model can handle any pairwise re- lational schema and a wide variety of error models. We demonstrate its eciency,
A key element of the social networks on the internet such as Facebook and Flickr is that they encourage users to create connections between themselves, other users and objects.
One important task that has been approached in the literature that deals with such data is to use social graphs to predict user behavior (e.g. joining a group of interest). More specifically, we study the cold-start problem, where users only participate in some relations, which we will call social relations, but not in the relation on which the predictions are made, which we will refer to as target relations.
We propose a formalization of the problem and a principled approach to it based on multi-relational factorization techniques. Furthermore, we derive a principled feature extraction scheme from the social data to extract predictors for a classifier on the target relation. Experiments conducted on real world datasets show that our approach outperforms current methods.
Improving a recommender system by collective matrix factorization with tag information
- bu