top-N Recommendation Research Articles

A collaborative filtering recommendation framework based on Wasserstein GAN

Compared with the original GAN, Wasserstein GAN minimizes the Wasserstein Distance between the generative distribution and the real distribution, can well capture the potential distribution of data and has achieved excellent results in image generation. However, the exploration of Wasserstein GAN on recommendation systems has received relatively less scrutiny. In this paper, we propose a collaborative filtering recommendation framework based on Wasserstein GAN called CFWGAN to improve recommendation accuracy. By learning the real user distribution, we can mine the potential nonlinear interactions between users and items, and capture users’ preferences for all items. Besides, we combine two positive and negative item sampling methods and add the reconstruction loss to the generator’s loss. This can well handle the problem of discrete data in recommendation (relative to the continuity of image data). By continuously approximating the generative distribution to the real user distribution, we can finally obtain better users’ preference information and provide higher accuracy in recommendation. We evaluate the CFWGAN model on three real-world datasets, and the empirical results show that our method is competitive with or superior to state-of-the-art approaches on the benchmark top-N recommendation task.

Trust‐aware generative adversarial network with recurrent neural network for recommender systems

Recently recommender systems become more and more significant in the daily life. Although the recommender systems based on the generative adversarial network (GAN) are competent, the user trust information is seldom taken into consideration. In this paper, we propose a Trust-Aware Generative adversarial network with recurrent neural network for RECommender systems named TagRec, which makes use of the user trust information for top-N recommendation. In the framework, the discriminative model is a multi-layer perceptron to distinguish whether a sample is from the real data or fake data. The discriminator helps to guide the training of the generative model to make it fit the data distribution of the user trust information. The generative model is a recurrent neural network (RNN) with long short-term memory (LSTM) cells, aiming to confuse the discriminative model by generating samples as similar as possible to the real data. Through the adversarial training between the discriminative and generative models, the user trust information can be fully used to improve the recommendation performance. We conduct extensive experiments on real-word datasets to validate the effectiveness of the TagRec.

Top-N recommendation algorithm integrated neural network

With the gradual popularization of social informatization, people’s information security is gradually threatened. The development of the Internet has gradually exposed people’s privacy, and the protection of the privacy of people in the era of Internet information has become a topic of concern to all people. This research mainly discusses the research of the Top-N recommendation algorithm with integrated neural network. The purpose of protecting people’s privacy is achieved by interfering with the Top-N recommendation algorithm on the Internet signal. In response to people’s concerns, the Top-N recommendation algorithm with integrated neural network was used during the experiment. The experimenters were randomly selected from netizens who frequently used computers to measure the privacy and security of each group of researchers and the signal of the Top-N recommendation algorithm. Interference level. The use of the Top-N recommendation algorithm is divided into six levels, and the experimentally measured information protection rate is 88% when the use level of the Top-N recommendation algorithm is F level. In the case of signal interference, the interference intensity is divided into five levels. Similarly, when the signal interference intensity is 5, the information leakage rate is at least 10%. The selection of personnel throughout the experiment is random and the interference during the experiment and the use of the Top-N recommendation algorithm with integrated neural network are divided according to levels. The research results show that when the signal interference intensity is 5 and the recommended algorithm is F, the privacy protection of netizens is the best. The Top-N recommendation algorithm with integrated neural network has important potential value in protecting people’s privacy.

Leveraging pointwise prediction with learning to rank for top-N recommendation

Pointwise prediction and Learning to Rank (L2R) are two hot strategies to model user preference in recommender systems. Currently, these two types of approaches are often considered independently, and most existing efforts utilize them separately. Unfortunately, pointwise prediction tends to cause the problem of overfitting, while L2R is more prone to higher variance. On the other hand, the advantages of multi-task learning and ensemble learning inspire us to utilize multiple approaches jointly so that methods can promote together synergistically. Therefore, we propose a new framework called CPL, where pointwise prediction and L2R are inherently combined to improve the performance of top-N recommendations. To verify the effectiveness of CPL, an instantiation of CPL, which is named CPLmg, is introduced. CPLmg is based on two components, i.e., Factorized SLIM (Sparse LInear Method) and GAPfm (Graded Average Precision factor model), to perform pointwise prediction and L2R, respectively. Different from the original version of SLIM, FSLIM reconstructs a denser representation both for users and items. Moreover, the low-rank users’ and item’s latent factor matrices act as a bridge between FSLIM and GAPfm. Extensive experiments on four real-world datasets show that CPLmg significantly outperforms the compared methods. To explore other possible combinations for CPL further, we selected another pointwise method, i.e., FunkSVD, and L2R approach, i.e., BPR, to implement CPLdb. The experimental results demonstrate the superiority of CPL again as it can help improve the performance of its pointwise prediction and L2R components.

Neural graph personalized ranking for Top-N Recommendation

Personalized recommendation has been widely applied to many real-world services. Many of recent studies focus on collaborative filtering (CF) by deep neural networks, which pursue to predict users’ preference on items based on the past user–item interactions (e.g., a user rates an item). A general CF approach consists of two key modules, embedding representation learning and interaction modeling. In most existing methods, the embedding module is followed by the interaction modeling module, and the user–item interaction information is only emploited in interaction modeling directly. Existing methods, however, defectively overlook the correlation between users and items, as well as the inherent connection between embedding learning and the interaction information. To fill this gap, we propose neural graph personalized ranking (NGPR) which directly makes use of the user–item interaction information in embedding learning by incorporating the user–item interaction graph in embedding learning. Specifically, we construct the user–item interaction graph using de facto interaction between a user and an item. Correlation between users and items can also be reserved by concatenating representations of users and items in the entire procedure of embedding learning. Moreover, more complicated structures like multilayer perceptron (MLP) can be used in interaction modeling to make the most use of the representations, rather than simple linear transformation. We conduct extensive experiments on three public benchmarks and demonstrate the superior performance of the proposed NGPR model on personalized ranking task. In addition, our ablation studies verify that our novel design to incorporate the user–item interaction graph in embedding learning is effective.

Item Cold-Start Recommendation with Personalized Feature Selection

The problem of recommending new items to users (often referred to as item cold-start recommendation) remains a challenge due to the absence of users’ past preferences for these items. Item features from side information are typically leveraged to tackle the problem. Existing methods formulate regression methods, taking item features as input and user ratings as output. These methods are confronted with the issue of overfitting when item features are high-dimensional, which greatly impedes the recommendation experience. Availing of high-dimensional item features, in this work, we opt for feature selection to solve the problem of recommending top-N new items. Existing feature selection methods find a common set of features for all users, which fails to differentiate users’ preferences over item features. To personalize feature selection, we propose to select item features discriminately for different users. We study the personalization of feature selection at the level of the user or user group. We fulfill the task by proposing two embedded feature selection models. The process of personalized feature selection filters out the dimensions that are irrelevant to recommendations or unappealing to users. Experimental results on real-life datasets with high-dimensional side information reveal that the proposed method is effective in singling out features that are crucial to top-N recommendation and hence improving performance.

Boosting Item-based Collaborative Filtering via Nearly Uncoupled Random Walks

Item-based models are among the most popular collaborative filtering approaches for building recommender systems. Random walks can provide a powerful tool for harvesting the rich network of interactions captured within these models. They can exploit indirect relations between the items, mitigate the effects of sparsity, ensure wider itemspace coverage, as well as increase the diversity of recommendation lists. Their potential however, can be hindered by the tendency of the walks to rapidly concentrate towards the central nodes of the graph, thereby significantly restricting the range of K -step distributions that can be exploited for personalized recommendations. In this work, we introduce RecWalk ; a novel random walk-based method that leverages the spectral properties of nearly uncoupled Markov chains to provably lift this limitation and prolong the influence of users’ past preferences on the successive steps of the walk—thereby allowing the walker to explore the underlying network more fruitfully. A comprehensive set of experiments on real-world datasets verify the theoretically predicted properties of the proposed approach and indicate that they are directly linked to significant improvements in top- n recommendation accuracy. They also highlight RecWalk’s potential in providing a framework for boosting the performance of item-based models. RecWalk achieves state-of-the-art top- n recommendation quality outperforming several competing approaches, including recently proposed methods that rely on deep neural networks.

AIRec: Attentive intersection model for tag-aware recommendation

Tag-aware recommender systems (TRS) utilize rich tagging information to better depict user portraits and item features. Recently, many efforts have been done to improve TRS with neural networks. However, existing methods construct user representations through either explicit tagging behaviors or implicit interacted items, which is inadequate to capture multi-aspect user preferences. Besides, there are still lacks of investigation about the intersection between user and item tags, which is crucial for better recommendation.In this paper, we propose AIRec, an attentive intersection model for TRS, to address the above issues. More precisely, we first project the sparse tag vectors into a latent space through multi-layer perceptron (MLP). Then, the user representations are constructed with a hierarchical attention network, where the item-level attention differentiates the contributions of interacted items and the preference-level attention discriminates the saliencies between explicit and implicit preferences. After that, the intersection between user and item tags is exploited to enhance the learning of conjunct features. Finally, the user and item representations are concatenated and fed to factorization machines (FM) for score prediction. We conduct extensive experiments on two real-world datasets, demonstrating significant improvements of AIRec over state-of-the-art methods for tag-aware top-n recommendation.

Block-Aware Item Similarity Models for Top- N Recommendation

Top- N recommendations have been studied extensively. Promising results have been achieved by recent item-based collaborative filtering (ICF) methods. The key to ICF lies in the estimation of item similarities. Observing the block-diagonal structure of the item similarities in practice, we propose a block-diagonal regularization (BDR) over item similarities for ICF. The intuitions behind BDR are as follows: (1) with BDR, item clustering is embedded into the learning of ICF methods; (2) BDR induces sparsity of item similarities, which guarantees recommendation efficiency; and (3) BDR captures in-block transitivity to overcome rating sparsity. By regularizing the item similarity matrix of item similarity models with BDR, we obtain a block-aware item similarity model. Our experimental evaluations on a large number of datasets show that the block-diagonal structure is crucial to the performance of top- N recommendation.

STS: Spatial–Temporal–Semantic Personalized Location Recommendation

The rapidly growing location-based social network (LBSN) has become a promising platform for studying users’ mobility patterns. Many online applications can be built based on such studies, among which, recommending locations is of particular interest. Previous studies have shown the importance of spatial and temporal influences on location recommendation; however, most existing approaches build a universal spatial–temporal model for all users despite the fact that users always demonstrate heterogeneous check-in behavior patterns. In order to realize truly personalized location recommendations, we propose a Gaussian process based model for each user to systematically and non-linearly combine temporal and spatial information to predict the user’s displacement from their currently checked-in location to the next one. The locations whose distances to the user’s current checked-in location are the closest to the predicted displacement are recommended. We also propose an enhancement to take into account category information of locations for semantic-aware recommendation. A unified recommendation framework called spatial–temporal–semantic (STS) is introduced to combine displacement prediction and the semantic-aware enhancement to provide final top-N recommendation. Extensive experiments over real datasets show that the proposed STS framework significantly outperforms the state-of-the-art location recommendation models in terms of precision and mean reciprocal rank (MRR).

Anaffective Aware Pseudo Association Method to Connect Disjoint Users Across Multiple Datasets – An Enhanced Validation Method Fortext-based Emotion Aware Recommender

We derive a method to enhance the evaluation for a text-based Emotion Aware Recommender that we have developed. However, we did not implement a suitable way to assess the top-N recommendations subjectively. In this study, we introduce an emotion-aware Pseudo Association Method to interconnect disjointed users across different datasets so data files can be combined to form a more extensive data file. Users with the same user IDs found in separate data files in the same dataset are often the same users. However, users with the same user ID may not be the same user across different datasets. We advocate an emotion aware Pseudo Association Method to associate users across different datasets. The approach interconnects users with different user IDs across different datasets through the most similar users' emotion vectors (UVECs). We found the method improved the evaluation process of assessing the top-N recommendations objectively.

A novel Collaborative Filtering recommendation approach based on Soft Co-Clustering

Collaborative Filtering (CF) recommendation algorithm has been widely applied into recommender systems. Many CF algorithms associate a user/an item with one of subgroups by explicit or implicit features. However, considering that users may have multiple personalities and items may have diverse attributes, it is more reasonable to associate a user/an item with more than one group. In this paper, we propose the Soft K-indicators Alternative Projection (SKAP) algorithm, which can efficiently resolve soft clustering problem with high dimensions, to generate a sparse partition matrix and further a Top-N recommendation list is given. Unlike fuzzy C-means clustering, the SKAP algorithm is independent on the selection of initial values. In addition to that, we integrate the item type information into recommender systems to improve recommendation accuracy. Experimental results show that the proposed approach behaves superior performance in Top-N recommendation in terms of classical metrics and further show that multi-label classification framework is a better description than classical Co-Clustering framework.

Local ranking and global fusion for personalized recommendation

Most of top-n personalized recommendation algorithms perform a kind of global learning-to-rank based on the original user–item rating matrix, yet with an implicit presumption of a unified feature space for all users (all items). However, we argue that such a presumption is a limiting factor for better recommendation, because features learned in a global space could be too coarse-grained. As such, ranking a recommendation list against a unified space might be dominated by only a few significant aspects of users’ interests.In this paper, we propose a novel local ranking and global fusion (LRGF) framework for personalized recommendation, which allows using multiple fine-grained feature spaces for learning-to-rank. In LRGF, we first adopt two-stage random walk on the established user graph and item graph for selecting those mostly related users and items, so as to form multiple local user–item groups. Next, we propose a pairwise gap-aware Bayesian personalized ranking to obtain local ranking lists for each local user–item group. Finally, a global decision fusion is proposed to obtain the final top-n recommendation list for each user only from her local ranking lists. Experiments on the three real-world datasets show that the proposed LRGF outperforms the state-of-the-art schemes in terms of higher mean average precision and normalized discounted cumulative gain.

Exploring multiple spatio-temporal information for point-of-interest recommendation

Traditional collaborative filtering methods perform poorly in providing location recommendations due to the high sparsity of users’ check-in data, prompting the development of new location recommendation approaches that can integrate situational factors such as time and location. Using long short-term memory (LSTM) neural networks and kernel density estimation (KDE), this paper integrates the impact of point-of-interest (POI) location and category on users’ check-in behavior according to check-in sequence data. First, LSTM neural networks are used to model users’ periodic and repetitive daily activities for a sequence-based prediction of the probability of whether the user will visit a candidate POI. Second, the user’s geographical preference in the two-dimensional space is represented by KDE and used to make a location-based check-in probability prediction. Next, the user’s category preference is used to predict the check-in probability of a candidate POI. Finally, a user preference model is constructed from three perspectives of time, location, and category, and the comprehensive check-in probability is used for Top-N recommendation. The validation experiments on Foursquare dataset verifies that, in terms of recommendation precision and recall, the proposed recommendation method is superior to both the basic LSTM approach and the method that uses only location information. In addition, it is experimentally confirmed that the geographical preference, which is reflected by “clustering” of a user’s check-in locations, is stable, but the user’s category preference is prone to drift.

Capsmf: a novel product recommender system using deep learning based text analysis model

Researchers and data scientists have developed different Recommender System Algorithms such as Content-Based and Collaborative-Based in order to filter a large amount of information available on the internet and hence, recommend only the relevant and essential content based on the personalized interests of users. Information acquired explicitly by collecting users’ ratings for an item lead to the problem of data sparsity. Many researchers have been currently working towards the improvement of rating prediction accuracy by integrating the auxiliary information along with the ratings provided by the users. This paper proposes a novel product recommender system called as “CapsMF”, it applies the advanced neural network architecture Capsule Networks (Caps) for document representation, and MF represents Matrix factorization. In the proposed approach, we have enhanced a deep neural network text analysis model by adding a newly discovered neural network architecture; Capsule Networks stacked on bi-directional Recurrent Neural Network (Bi-RNN) for the robust representation of textual descriptions of items and users. The Deep Neural Network text analysis model is integrated with the Probabilistic Matrix Factorization to generate improved recommendations. The experiment has been performed on two real amazon datasets resulting in the enhancement of rating prediction accuracy, the recall, and the precision of top-n recommendations, in comparison to the basic and hybrid Recommendation System Algorithms. Also, text analysis model involving Capsule Networks stacked with Recurrent Neural Networks (RNNs) have outperformed the baseline models that have single Convolutional Neural Networks (CNN) or CNN combined with Bi-RNN in text analysis.

BRS cS: a hybrid recommendation model fusing multi-source heterogeneous data

Recommendation systems are often used to solve the problem of information overload on the Internet. Many types of data can be used for recommendations, and fusing different types of data can make recommendations more accurate. Most existing fusion recommendation models simply combine the recommendation results from different data instead of fully fusing multi-source heterogeneous data to make recommendations. Furthermore, users’ choices are usually affected by their direct and even indirect friends’ preferences. This paper proposes a hybrid recommendation model BRS cS (an acronym for BPR-Review-Score-Social). It fully fuses social data, score, and review together; uses improved BPR model to optimize the ranking; and trains them in a joint representation learning framework to get the top-N recommendations. User trust model is used to introduce social relationships into the rating and review data, PV-DBOW model is used to process the review data, and fully connected neural network is used to process the rating data. Experiments on Yelp public dataset show that the BRS cS algorithm proposed outperforms other recommendation algorithms such as BRS c, UserCF, and HRS c. The BRS cS model is also scalable and can fuse new types of data easily.

KG2Rec: LSH-CF recommendation method based on knowledge graph for cloud services

With the rapid development of cloud computing paradigm and mobile devices, the number of cloud services and user interest data are growing explosively. It becomes harder for users to find suitable cloud services within satisfying response time. Thus a suitable cloud service automatic recommendation system is needed to solve the above problem. In this work, we propose KG2Rec, a novel recommendation method based on Knowledge Graph to provide users with appropriate recommendations to meet their real-time needs. The property-specific knowledge graph model is designed to model user–item and item–item bipartite relations. Based on the property-specific knowledge graph model, property-specific user–item relation features can be mined from it. Then the user–item relation features will be integrated into an improved collaborative filtering algorithm based on the local sensitive hash technique for Top-N item recommendation. Finally, we evaluate the proposed method in terms of Top-N recommendation on the MovieLens dataset, and prove it outperforms numbers of state-of-the-art recommendation systems. In addition, we prove it has well performance in term of long tail recommendation and big data processing.

Memory-aware gated factorization machine for top-N recommendation

Factorization machine (FM) has recently become one of the most popular methods in collaborative filtering due to its flexibility of incorporating auxiliary information, e.g., user demographics and item genres. However, standard FM method and its deep variants (e.g., NFM and DeepFM) suffer from two key issues: (1) failing to effectively leverage user historical records, i.e., all historical records are treated equally without considering the relevance to the targeted user–item pair and (2) failing to adaptively weigh the importance of auxiliary information, i.e., auxiliary information may have negative effects on the accuracy in certain cases but existing methods cannot effectively detect and eliminate the negative effects. To this end, this paper proposes a memory-aware gated factorization machine (MAGFM), which improves the FM method by introducing two new components: (1) an external user memory matrix is introduced to each user, which can enrich the representation capacity by leveraging user historical items and the auxiliary information associated with the historical items and (2) gated filtering units are applied on top of the embedding of user/item auxiliary information, which can adaptively filter out the features with negative effects to achieve higher accuracy. Experimental studies on real-world datasets demonstrate that MAGFM can substantially outperform FM, NFM and DeepFM methods by 0.31% – 12.77% relatively in top-N recommendation.

Efficient Heterogeneous Collaborative Filtering without Negative Sampling for Recommendation

Recent studies on recommendation have largely focused on exploring state-of-the-art neural networks to improve the expressiveness of models, while typically apply the Negative Sampling (NS) strategy for efficient learning. Despite effectiveness, two important issues have not been well-considered in existing methods: 1) NS suffers from dramatic fluctuation, making sampling-based methods difficult to achieve the optimal ranking performance in practical applications; 2) although heterogeneous feedback (e.g., view, click, and purchase) is widespread in many online systems, most existing methods leverage only one primary type of user feedback such as purchase. In this work, we propose a novel non-sampling transfer learning solution, named Efficient Heterogeneous Collaborative Filtering (EHCF) for Top-N recommendation. It can not only model fine-grained user-item relations, but also efficiently learn model parameters from the whole heterogeneous data (including all unlabeled data) with a rather low time complexity. Extensive experiments on three real-world datasets show that EHCF significantly outperforms state-of-the-art recommendation methods in both traditional (single-behavior) and heterogeneous scenarios. Moreover, EHCF shows significant improvements in training efficiency, making it more applicable to real-world large-scale systems. Our implementation has been released 1 to facilitate further developments on efficient whole-data based neural methods.

Genetic Algorithm Influenced Top-N Recommender System to Alleviate New User Cold Start Problem

Most recommender systems are based on the familiar collaborative filtering algorithm to suggest items. Quite often, collaborative filtering algorithm fails in generating recommendations due to the lack of adequate user information resulting in new user cold start problem. The cold start problem is one among the prevailing issue in recommendation system where the system fails to render recommendations. To overcome the new user cold start issue, demographical information of the user is utilised as the user information source. Among the demographical information, the impact of the user gender is less explored when compared with other information like age, profession, region, etc. In this work, a genetic algorithm-influenced gender-based top-n recommender algorithm is proposed to address the new user cold start problem. The algorithm utilises the evolution concepts of the genetic algorithm to render top-n recommendations to a new user. The evaluation of the proposed algorithm using real world datasets proved that the algorithm has a better efficiency than the state of art approaches.