Search

Music has been a means of emotional expression throughout human history. Different cultures and times have offered unique ways of expressing the emotions experienced by individuals through music. Technological advancements have greatly enhanced our interaction with music, making access easier thanks to digital platforms. While traditional music recommendation systems are based on the user's listening habits, this study aims to provide music recommendations by evaluating the user's immediate emotional state. We utilize a CNN-based emotion recognition system for this purpose. This model predicts one of seven emotions by analyzing facial images. The results of the Confusion Matrix (CM) in the training phase indicate that the model is proficient at classifying emotions. ROC curve analysis shows the model's high capability in emotion prediction. Notably, the model's overall accuracy stands at 92.7%, highlighting its effectiveness. This research demonstrates the potential of emotion-based music recommendation systems in enhancing the user experience. [ABSTRACT FROM AUTHOR]

Social influence prediction has permeated many domains, including marketing, behavior prediction, recommendation systems, and more. However, traditional methods of predicting social influence not only require domain expertise, they also rely on extracting user features, which can be very tedious. Additionally, graph convolutional networks (GCNs), which deals with graph data in non-Euclidean space, are not directly applicable to Euclidean space. To overcome these problems, we extended DeepInf such that it can predict the social influence of COVID-19 via the transition probability of the page rank domain. Furthermore, our implementation gives rise to a deep learning-based personalized propagation algorithm, called DeepPP. The resulting algorithm combines the personalized propagation of a neural prediction model with the approximate personalized propagation of a neural prediction model from page rank analysis. Four social networks from different domains as well as two COVID-19 datasets were used to analyze the proposed algorithm's efficiency and effectiveness. Compared to other baseline methods, DeepPP provides more accurate social influence predictions. Further, experiments demonstrate that DeepPP can be applied to real-world prediction data for COVID-19. [ABSTRACT FROM AUTHOR]

With the explosion of information, recommendation systems have become important for users to find their interested information. Existing recommendation methods mainly utilize user historical interaction with items or user ratings to capture user past preferences. However, there is ignorance of various personalized reasons for users preferring an item, in which the reasons always dominate users' preference strengths on the item. In addition, the linear nature of traditional recommendation methods makes them less effective in dealing with complex data. With the development of deep learning methods, graph neural networks provide an unprecedented opportunity for recommendations, since the user-item interactions can be naturally represented as a graph and the method can extract high-order complex relationships between users and items. In this paper, we propose a novel method leveraging the FIne-Grained user preferences with Graph Neural Networks (FigGNN) for recommendation to tackle these issues. More specifically, user-item interactions with user annotated tags and user ratings are constructed as a graph. In the process of graph message propagation, the user annotated tags are incorporated for understanding user preference reasons on items, and heterogeneous user rating levels are utilized for recognizing user preference strengths on items. Experiments have been conducted on the MovieLens dataset and the results show a superior performance of FigGNN over baselines in terms of precision and recall, which demonstrates the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

With the explosive growth of online information, many recommendation methods have been proposed. This research direction is boosted with deep learning architectures, especially the recently proposed graph convolutional networks (GCNs). GCNs have shown tremendous potential in graph embedding learning thanks to its inductive inference property. However, most of the existing GCN-based methods focus on solving tasks in the homogeneous graph settings, and none of them considers heterogeneous graph settings. In this paper, we bridge the gap by developing a novel framework called HeteGraph based on the GCN principles. HeteGraph can handle heterogeneous graphs in the recommender systems. Specifically, we propose a sampling technique and a graph convolutional operation to learn high-quality graph's node embeddings, which differs from the traditional GCN approaches where a full graph adjacency matrix is needed for the embedding learning. We design two models based on the HeteGraph framework to evaluate two important recommendation tasks, namely item rating prediction and diversified item recommendations. Extensive experiments show the encouraging performance of HeteGraph on the first task and the state-of-the-art performance on the second task. [ABSTRACT FROM AUTHOR]

For skeleton-based action recognition, graph convolutional networks (GCN) have absolute advantages. Existing state-of-the-art (SOTA) methods tended to focus on extracting and identifying features from all bones and joints. However, they ignored many new input features which could be discovered. Moreover, many GCN-based action recognition models did not pay sufficient attention to the extraction of temporal features. In addition, most models had swollen structures due to too many parameters. In order to solve the problems mentioned above, a temporal feature cross-extraction graph convolutional network (TFC-GCN) is proposed, which has a small number of parameters. Firstly, we propose the feature extraction strategy of the relative displacements of joints, which is fitted for the relative displacement between its previous and subsequent frames. Then, TFC-GCN uses a temporal feature cross-extraction block with gated information filtering to excavate high-level representations for human actions. Finally, we propose a stitching spatial–temporal attention (SST-Att) block for different joints to be given different weights so as to obtain favorable results for classification. FLOPs and the number of parameters of TFC-GCN reach 1.90 G and 0.18 M, respectively. The superiority has been verified on three large-scale public datasets, namely NTU RGB + D60, NTU RGB + D120 and UAV-Human. [ABSTRACT FROM AUTHOR]

Course recommendation aims at finding proper and attractive courses from massive candidates for students based on their needs, and it plays a significant role in the curricula-variable system. However, nearly all students nowadays need help selecting appropriate courses from abundant ones. The emergence and application of personalized course recommendations can release students from that cognitive overload problem. However, it still needs to mature and improve its scalability, sparsity, and cold start problems resulting in poor quality recommendations. Therefore, this paper proposes a novel personalized course recommendation system based on deep factorization machine (DeepFM), namely Deep PersOnalized couRse RecommendatIon System (DORIS), which selects the most appropriate courses for students according to their basic information, interests and the details of all courses. The experimental results illustrate that our proposed method outperforms other approaches. [ABSTRACT FROM AUTHOR]

Web service recommender systems have a fundamental role in the selection, composition and substitution of services. Indeed, they are used in several application areas such as Web APIs and Cloud Computing. Likewise, Deep Learning techniques have brought undeniable advantages and solutions to the challenges faced by recommendations in all areas. Unfortunately, the field of Web services has not yet benefited well from these deep methods, moreover, the works using these methods for Web services domain are very recent compared to the works of other fields. Thus, the objective of this paper is to study and analyze state-of-the-art work on Web services recommender systems based on Deep Learning techniques. This analysis will help readers wishing to work in this field, and allows us to direct our future work concerning the Web services recommendation by exploiting the advantages of Deep Learning techniques. [ABSTRACT FROM AUTHOR]

Recommender systems based on sentiment analysis become challenging due to the presence of enormous data available over the internet. With the lack of proper data cleaning and analysis methods, existing machine learning (ML) techniques fail to generate accurate recommendations. To overcome this issue, this paper proposes a Light Deep Learning (LightDL)-based recommender system that uses Twitter-based reviews. First, the data is collected from Twitter and cleaned by subsequent data cleaning processes. Then, this pre-processed data is fed into the LightDL model, which learns the important features like hashtags, unigrams, multigrams, etc. from each piece of data. Here, we have learned about four groups of features, including semantic features, syntactic features, symbolic features, and tweet-based features. Finally, the data is classified into positive, negative, and neutral categories according to the learned features. On the basis of classified sentiment, the review is generated to the users. Finally, the model is evaluated in terms of accuracy, precision, recall, f-measure, and error rate through extensive experiments in Matlab. The proposed LightDL model outperforms in all performance measures; specifically, it achieves 95% accuracy for the Twitter dataset. [ABSTRACT FROM AUTHOR]

In today's online market, recommendation systems have become universal and are an aspect of any online shopping portal. The traditional approach uses the subscriber's historical knowledge, and this technique is not adequate for resolving problems with a cold start. These issues include recommendations for non-registered users or newly added customers and new items added. Session-based recommendations based on recurrent neural networks are gaining popularity for product recommendations. This is due to recurrent neural networks' ability to record sequential feature data more effectively throughout the current session, which results in more similarity between consumer behaviour sequences. Nevertheless, most state-of-the-art recurring neural networking systems completely ignore the long-term details of multiple sessions and concentrate solely on short-term communication in a single session. This paper presents a hybrid time-centric prediction model to address research issues that learn the customers' short and long-term behaviours. Experiments on the recsys challenge data set are carried out to assess the efficiency of the hybrid time-centric prediction models over the existing hybrid models in terms of HitRate and Mean-Reciprocal Rate. [ABSTRACT FROM AUTHOR]

Recommendation system is one of the most common and basic applications in E-Commerce, search engines, etc. However, the user-item rating matrix data in the recommendation system is extremely sparse, resulting in poor recommendation results. This paper proposes a dual auto encoders matrix factorization (DAE-MF) model for the sparseness of the scoring matrix. This model uses a two-way parallel network structure, which combines a convolutional auto encoder and stacked denoising autoencoder and matrix factorization technology, fully combining the advantages of the two modules, effectively improving the model’s data completion capabilities. Through experiments on MovieLens and AIV datasets, it is verified that the DAE-MF method is superior to the existing models in rating prediction tasks. [ABSTRACT FROM AUTHOR]

With the rapid development of the patent marketplace, patent trading recommendation is required to mitigate the technology searching cost of patent buyers. Current research focuses on the recommendation based on existing patents of a company; a few studies take into account the sequential pattern of patent acquisition activities and the possible diversity of a company's business interests. Moreover, the profiling of patents based on solely patent documents fails to capture the high-order information of patents. To bridge the gap, we propose a knowledge-aware attentional bidirectional long short-term memory network (KBiLSTM) method for patent trading recommendation. KBiLSTM uses knowledge graph embeddings to profile patents with rich patent information. It introduces bidirectional long short-term memory network (BiLSTM) to capture the sequential pattern in a company's historical records. In addition, to address a company's diverse technology interests, we design an attention mechanism to aggregate the company's historical patents given a candidate patent. Experimental results on the United States Patent and Trademark Office (USPTO) data set show that KBiLSTM outperforms state-of-the-art baselines for patent trading recommendation in terms of F1 and normalised discounted cumulative gain (nDCG). The attention visualisation of randomly selected company intuitively demonstrates the recommendation effectiveness. [ABSTRACT FROM AUTHOR]

Recommending personalized content from massive data for users is the key function of the recommendation system. The recommendation process of the traditional recommendation systems is often regarded as static, which cannot reflect the changes of user's real-time interest. This paper addressed this problem and presented a recommendation model that leverages the ability of deep learning methods to effectively deal with decision-making problems. In this model, a state generation module containing gate recurrent unit (GRU) and attention network was designed to obtain user's long and short-term preferences as well as history scores. Then, an actor-critic algorithm was employed to imitate the real-time recommendations. We trained the proposed model and evaluated it on four wellknown public datasets. It is proved that the proposed model is superior to existing recommendation models. [ABSTRACT FROM AUTHOR]

With the advancement of computer hardware and communication technologies, deep learning technology has made significant progress, enabling the development of systems that can accurately estimate human emotions. Factors such as facial expressions, gender, age, and the environment influence human emotions, making it crucial to understand and capture these intricate factors. Our system aims to recommend personalized images by accurately estimating human emotions, age, and gender in real time. The primary objective of our system is to enhance user experiences by recommending images that align with their current emotional state and characteristics. To achieve this, our system collects environmental information, including weather conditions and user-specific environment data through APIs and smartphone sensors. Additionally, we employ deep learning algorithms for real-time classification of eight types of facial expressions, age, and gender. By combining this facial information with the environmental data, we categorize the user's current situation into positive, neutral, and negative stages. Based on this categorization, our system recommends natural landscape images that are colorized using Generative Adversarial Networks (GANs). These recommendations are personalized to match the user's current emotional state and preferences, providing a more engaging and tailored experience. Through rigorous testing and user evaluations, we assessed the effectiveness and user-friendliness of our system. Users expressed satisfaction with the system's ability to generate appropriate images based on the surrounding environment, emotional state, and demographic factors such as age and gender. The visual output of our system significantly impacted users' emotional responses, resulting in a positive mood change for most users. Moreover, the system's scalability was positively received, with users acknowledging its potential benefits when installed outdoors and expressing a willingness to continue using it. Compared to other recommender systems, our integration of age, gender, and weather information provides personalized recommendations, contextual relevance, increased engagement, and a deeper understanding of user preferences, thereby enhancing the overall user experience. The system's ability to comprehend and capture intricate factors that influence human emotions holds promise in various domains, including human–computer interaction, psychology, and social sciences. [ABSTRACT FROM AUTHOR]

In recent years, service-oriented computing technology has developed rapidly. The growing number of services increases the choice burden of software developers when developing service-based systems, such as mashups or applications. How to recommend appropriate services for developers to create mashups has become a basic problem in service-oriented recommendation systems. To solve this problem, people have proposed various methods to recommend services to match the requirements of the new mashups and achieved great success. However, there are also some challenges in feature utilization and text requirement understanding. Therefore, we propose a Mashup-oriented Service Recommendation framework based on Deep Learning, called MSRDL. A content component was designed in MSRDL to generate the representation of mashups and services. Besides, an interaction component was created in MSRDL to model the invocation records between mashups and services. The output features of the two parts are further integrated into MLP to obtain the service recommendation lists. Experimental results on ProgrammableWeb datasets show that our method is superior to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Expert recommendation is one of the research hotspots in the online Q&A community. However, the majority of the currently used algorithms concentrate on matching user's static interest and question information, ignoring the efficient capture of user' s dynamic interest representation information, which results in a lack of recommendation accuracy. In order to address the aforementioned issues, this paper proposed an expert recommendation algorithm for the Q&A community that incorporated multi-head self-attention. Firstly, this algorithm built out question encoders of convolutional neural networks and attention mechanisms to process target question and user' s historically answered questions, in order to extract the corresponding question representations. Secondly, it treated the sequence of user's historically answered questions as a time series and used the multi-head self-attention mechanism to learn the dynamic interest representations hidden in the sequence, and combined with the user' s static interest representations to obtain the user' s comprehensive interest representations. Finally, it calculated the similarity between the target question representation and the comprehensive user representation to generate recommendation results. Based on the real data from the Zhihu Q&A community, the experimental results show that the performance of this algorithm is significantly better than the current popular expert recommendation algorithm. [ABSTRACT FROM AUTHOR]

With the rapid development of Internet medical information technology, a large amount of medical data appeared on the Internet, however, how to extract effective information from the massive and complex medical data to provide professional medical services and suggestions to users has become a hot spot for this research. The recommendation system can effectively solve the problem of accurate matching of complex medical data resources; however, the cold start, data sparsity and user interest migration of the system in the complex data environment have a large impact on the recommendation effect; therefore, this paper proposes a weighted neural matrix decomposition improved health management recommendation scheme incorporating deep learning techniques. The scheme first uses an implicit feedback method to improve the prediction scores and improve the linear model performance of the matrix decomposition algorithm to form a weighted neural matrix decomposition health management recommendation algorithm. Second, the improved method and deep neural network are fused to improve the performance of the nonlinear model part of the algorithm by using the structural properties of the neural network. Finally, this paper's method is compared with the mainstream six recommendation algorithms on four publicly available real datasets. The experimental results show that the root mean square error (RMSE) of the WENMF algorithm is smaller than that of the comparison algorithm on all four datasets, and the convergence speed is faster. The hit rate (HR) and normalized discounted cumulative gain (NDCG) of the WENMF algorithm are higher than those of the comparison algorithm on all four datasets, and the maximum difference is 0.04. Therefore, the recommendation accuracy and ranking quality of the WENMF algorithm in the recommendation system are verified, and the cold start and data sparsity problems of the recommendation system are effectively alleviated. [ABSTRACT FROM AUTHOR]

Although reinforcement learning (RL) can solve complex tasks after training, it's difficult to extend trained agents to target environments with perturbations. This lack of generalization ability has caused the dilemma of large-scale application of RL. Rich neural networks have brought huge improvements to deep learning, but these aren't adapted to RL and even bring negative effects. Therefore, RL algorithms have few choices of neural network, which greatly limits the representation and generalization ability of RL. To overcome these limitations, we propose a deep semi-dense compression network (DSCN) to improve RL generalization ability. First, we perform a structural extension on a general network model for RL. Then based on the information theory, we propose a semi-dense connection to enhance the neural information flow (NIF) of initial features. Finally, drawing on the ideas of course learning, we propose a channel compression approach to filter the redundant information of the deep network. In addition, we innovatively extend the experimental environment and evaluation metrics of the existing platform, which can fully evaluate the performance of DSCN. The experimental results show that our model achieves stable and significant generalization performance improvement. [ABSTRACT FROM AUTHOR]

Convolutional neural-network-based autoencoders, which can integrate the spatial correlation between pixels well, have been broadly used for hyperspectral unmixing and obtained excellent performance. Nevertheless, these methods are hindered in their performance by the fact that they treat all spectral bands and spatial information equally in the unmixing procedure. In this article, we propose an adaptive spectral–spatial attention autoencoder network, called SSANet, to solve the mixing pixel problem of the hyperspectral image. First, we design an adaptive spectral–spatial attention module, which refines spectral–spatial features by sequentially superimposing the spectral attention module and spatial attention module. The spectral attention module is built to select useful spectral bands, and the spatial attention module is designed to filter spatial information. Second, SSANet exploits the geometric properties of endmembers in the hyperspectral image while considering abundance sparsity. We significantly improve the endmember and abundance results by introducing minimum volume and sparsity regularization terms into the loss function. We evaluate the proposed SSANet on one synthetic dataset and four real hyperspectral scenes, i.e., Samson, Jasper Ridge, Houston, and Urban. The results indicate that the proposed SSANet achieved competitive unmixing results compared with several conventional and advanced unmixing approaches with respect to the root mean square error and spectral angle distance. [ABSTRACT FROM AUTHOR]

Deep learning, which is a subfield of machine learning, has opened a new era for the development of neural networks. The auto-encoder is a key component of deep structure, which can be used to realize transfer learning and plays an important role in both unsupervised learning and non-linear feature extraction. By highlighting the contributions and challenges of recent research papers, this work aims to review state-of-the-art auto-encoder algorithms. Firstly, we introduce the basic auto-encoder as well as its basic concept and structure. Secondly, we present a comprehensive summarization of different variants of the auto-encoder. Thirdly, we analyze and study auto-encoders from three different perspectives. We also discuss the relationships between auto-encoders, shallow models and other deep learning models. The auto-encoder and its variants have successfully been applied in a wide range of fields, such as pattern recognition, computer vision, data generation, recommender systems, etc. Then, we focus on the available toolkits for auto-encoders. Finally, this paper summarizes the future trends and challenges in designing and training auto-encoders. We hope that this survey will provide a good reference when using and designing AE models. [ABSTRACT FROM AUTHOR]

As a powerful data modeling tool, Heterogeneous Information Network (HIN) has been successfully used in auxiliary information exploitation to boost recommendation performance. For HIN based recommendation, it is challenging to extract and fuse useful features of user preferences and item attributes under different semantic paths in HINs. Existing methods leverage a pre-defined fusion function to integrate different semantics for recommendation, which cannot characterize the complex nonlinear interactions between users and items. In this paper, we present a general framework named MNRec, short for Meta-path fusion based Neural Recommendation, to extract and fuse user and item embeddings under different meta-paths for recommendation. Under the framework, we propose an instantiation of MNRec with Multi-Layer Perceptron (MLP) structure. It consists of two major steps, i.e., meta-path based heterogeneous network embedding and deep learning based rating prediction. Concretely, appropriate meta-paths are first designed according to domain knowledge. Then the embeddings of users and items are obtained through a meta-path and commuting matrix based heterogeneous network embedding method. Finally, in light of the powerful nonlinear modeling capabilities of deep neural networks, the learned embeddings under different meta-paths are integrated into a two-pathway MLP structure for rating prediction. Experimental results on three real-world datasets demonstrate the superiority and effectiveness of MNRec compared with state-of-the-art baselines in rating prediction. [ABSTRACT FROM AUTHOR]

Multi-task learning in deep neural networks has become a topic of growing importance in many research fields, including drug discovery. However, applying multi-task learning poses new challenges in improving prediction performance. This study investigated the potential of training data enrichment to enhance multi-task model prediction quality in drug discovery. The study evaluated four scenarios with varying degrees of information capacity of the training data and applied two types of test data to evaluate prediction performance. We used three datasets: ViralChEMBL, which consisted of binary activities of compounds against viral species, was applied for the classification task; pQSAR(159) and pQSAR(4267), which consisted of bio-activities of compounds and assays from the research of the profile-QSAR method, were applied for regression tasks. We built multi-task models based on the feed-forward DNNs using the PyTorch framework. Our findings showed that training data enrichment could be an effective means of enhancing prediction performance in multi-task learning, but the degree of improvement depends on the quality of the training data. The more unique compounds and targets the training data included, the more new compound-target interactions are required for prediction improvement. Also, we found out that even using multi-task learning, one could not predict the interactions of compounds that are highly dissimilar from those used for model training. The study provides some recommendations for effectively employing multi-task learning in drug discovery to improve prediction accuracy and facilitate the discovery of novel drug candidates. [ABSTRACT FROM AUTHOR]

In the futuristic Industry framework, user interactions with the product are seamlessly integrated with the product life cycle which results in Information overload. The shopbots were proposed in Industry 4.0 where the more focus was on process automation. In this research work, we propose recommender system for industry 5.0 which is based on collaboration of human beings and machines with more focus on user's personalization. Recommender system can be considered as an information filtering tool that provides suggestions to users about products, music, friend, topic, etc. This suggestion is based on the interest of users. Several research works have been carried out to improve recommendation accuracy by using matrix factorization, trust-based, hybrid-based, machine learning, and deep learning techniques. However, very few existing works have leveraged textual opinions for the recommendation to the best of our knowledge. Existing research works have focused only on numerical ratings, which do not reflect actual user behaviour. In this research work, sentiments of textual opinions are analysed for an in-depth analysis of user's behaviour. Recommendation accuracy is improved by using the proposed score Recop which is calculated from opinion sentiments. Furthermore, the sparsity issue is resolved by using our proposed approach. Amazon and Yelp review datasets are used for Experiment analysis. Mean absolute error (MAE) and root mean square error (RMSE) values are improved significantly using the proposed approach compared to the existing approaches. MAE and RMSE scores on the Yelp dataset are 0.85 and 1.51, respectively. Additionally, MAE and RMSE scores on the Amazon dataset are 0.66 and 0.93, respectively, which clearly reflects the significant contribution of our proposed approach. [ABSTRACT FROM AUTHOR]

Traditional recommendation algorithms map text content through topic model or mean value of word vectorization. For the issue that existing methods cannot make full use of text information or ignore word order information, this study proposes a multi-task learning recommendation method for scientific literature. Based on the multi-task learning framework, an encoder is designed and a GL model is established. The GL model is trained to combine content recommendation and text metadata prediction, which improves the sparsity of traditional collaborative filtering and regularizes the collaborative filtering model. Finally, an evaluation test is carried out on public and private data sets respectively, and the superiority of the proposed method is demonstrated by comparing with the existing classical methods. [ABSTRACT FROM AUTHOR]

Hybrid recommendation algorithms perform well in improving the accuracy of recommendation systems. However, in specific applications, they still cannot reach the requirements of the recommendation target due to the gap between the design of the algorithms and data characteristics. In this paper, in order to learn higher-order feature interactions more efficiently and to distinguish the importance of different feature interactions better on the prediction results of recommendation algorithms, we propose a light and FM deep neural network (LFDNN), a hybrid recommendation model including four modules. The LightGBM module applies gradient boosting decision trees for feature processing, which improves LFDNN's ability to handle dense numerical features; the shallow model introduces the FM model for explicitly modeling the finite-order feature crosses, which strengthens the expressive ability of the model; the deep neural network module uses a fully connected feedforward neural network to allow the model to obtain more high-order feature crosses information and mine more data patterns in the features; finally, the Fusion module allows the shallow model and the deep model to obtain a better fusion effect. The results of comparison, parameter influence and ablation experiments on two real advertisement datasets shows that the LFDNN reaches better performance than the representative recommendation models. [ABSTRACT FROM AUTHOR]

The choice of learning rate (LR) functions and policies has evolved from a simple fixed LR to the decaying LR and the cyclic LR, aiming to improve the accuracy and reduce the training time of Deep Neural Networks (DNNs). This article presents a systematic approach to selecting and composing an LR policy for effective DNN training to meet desired target accuracy and reduce training time within the pre-defined training iterations. It makes three original contributions. First, we develop an LR tuning mechanism for auto-verification of a given LR policy with respect to the desired accuracy goal under the pre-defined training time constraint. Second, we develop an LR policy recommendation system (LRBench) to select and compose good LR policies from the same and/or different LR functions through dynamic tuning, and avoid bad choices, for a given learning task, DNN model, and dataset. Third, we extend LRBench by supporting different DNN optimizers and show the significant mutual impact of different LR policies and different optimizers. Evaluated using popular benchmark datasets and different DNN models (LeNet, CNN3, ResNet), we show that our approach can effectively deliver high DNN test accuracy, outperform the existing recommended default LR policies, and reduce the DNN training time by 1.6-6.7× to meet a targeted model accuracy. [ABSTRACT FROM AUTHOR]

Deep learning has yielded success in many research fields. In the last few years, deep learning techniques have been applied in recommender systems to solve cold start and data sparsity problems. However, only a few attempts have been made in social-based recommender systems. In this study, we address this issue and propose a novel recommendation model called Sentiment based Social Neural Hybrid Filtering (S-SNHF). This model combines collaborative and content-based filtering with social information using a deep neural architecture based on Generalized Matrix Factorization (GMF) and Hybrid Multilayer Perceptron (HybMLP). Furthermore, for achieving higher recommendation reliability, the hybrid sentiment analysis model is integrated to analyse users' opinions and infer their preferences. The results of the empirical study performed with three popular datasets show the contribution of both, social information and sentiment analysis on the recommendation performance and that our approach achieves significantly better recommendation accuracy, compared with state-of-the-art recommendation methods. [ABSTRACT FROM AUTHOR]

Deep learning provides accurate collaborative filtering models to improve recommender system results. Deep matrix factorization and their related collaborative neural networks are the state of the art in the field; nevertheless, both models lack the necessary stochasticity to create the robust, continuous, and structured latent spaces that variational autoencoders exhibit. On the other hand, data augmentation through variational autoencoder does not provide accurate results in the collaborative filtering field due to the high sparsity of recommender systems. Our proposed models apply the variational concept to inject stochasticity in the latent space of the deep architecture, introducing the variational technique in the neural collaborative filtering field. This method does not depend on the particular model used to generate the latent representation. In this way, this approach can be applied as a plugin to any current and future specific models. The proposed models have been tested using four representative open datasets, three different quality measures, and state-of-the-art baselines. The results show the superiority of the proposed approach in scenarios where the variational enrichment exceeds the injected noise effect. Additionally, a framework is provided to enable the reproducibility of the conducted experiments. [ABSTRACT FROM AUTHOR]

In recent years, there has been a growing interest in developing next point-of-interest (POI) recommendation systems in both industry and academia. However, current POI recommendation strategies suffer from the lack of sufficient mixing of details of the features related to individual users and their corresponding contexts. To overcome this issue, we propose a deep learning model based on an attention mechanism in this study. The suggested technique employs an attention mechanism that focuses on the pattern's friendship, which is responsible for concentrating on the relevant features related to individual users. To compute context-aware similarities among diverse users, our model employs six features of each user as inputs, including user ID, hour, month, day, minute, and second of visiting time, which explore the influences of both spatial and temporal features for the users. In addition, we incorporate geographical information into our attention mechanism by creating an eccentricity score. Specifically, we map the trajectory of each user to a shape, such as a circle, triangle, or rectangle, each of which has a different eccentricity value. This attention-based mechanism is evaluated on two widely used datasets, and experimental outcomes prove a noteworthy improvement of our model over the state-of-the-art strategies for POI recommendation. [ABSTRACT FROM AUTHOR]

Traditional recommendation systems are faced with two long-standing obstacles, namely data sparsity and cold-start problems, which promote the emergence and development of Cross-Domain Recommendation (CDR). The core idea of CDR is to leverage information collected from other domains to alleviate the two problems in one domain. Since the early 2010s, many efforts have been engaged for cross-domain recommendation. Recently, with the development of deep learning and neural networks, a large number of methods have emerged. However, there is a limited number of systematic surveys on CDR, especially regarding the latest proposed methods as well as the recommendation scenarios and recommendation tasks they address. In this survey article, we first proposed a two-level taxonomy of cross-domain recommendation that classifies different recommendation scenarios and recommendation tasks. We then introduce and summarize existing cross-domain recommendation approaches under different recommendation scenarios in a structured manner. We also organize datasets commonly used. We conclude this survey by providing several potential research directions about this field. [ABSTRACT FROM AUTHOR]

In recommender systems, top-N recommendation is an important task with implicit feedback data. Although the recent success of deep learning largely pushes forward the research on top-N recommendation, there are increasing concerns on appropriate evaluation of recommendation algorithms. It therefore is important to study how recommendation algorithms can be reliably evaluated and thoroughly verified. This work presents a large-scale, systematic study on six important factors from three aspects for evaluating recommender systems. We carefully select 12 top-N recommendation algorithms and eight recommendation datasets. Our experiments are carefully designed and extensively conducted with these algorithms and datasets. In particular, all the experiments in our work are implemented based on an open sourced recommendation library, Recbole [139], which ensures the reproducibility and reliability of our results. Based on the large-scale experiments and detailed analysis, we derive several key findings on the experimental settings for evaluating recommender systems. Our findings show that some settings can lead to substantial or significant differences in performance ranking of the compared algorithms. In response to recent evaluation concerns, we also provide several suggested settings that are specially important for performance comparison. [ABSTRACT FROM AUTHOR]

Click-through rate (CTR) prediction is a research point for measuring recommendation systems and calculating AD traffic. Existing studies have proved that deep learning performs very well in prediction tasks, but most of the existing studies are based on deterministic models, and there is a big gap in capturing uncertainty. Modeling uncertainty is a major challenge when using machine learning solutions to solve real-world problems in various domains. In order to quantify the uncertainty of the model and achieve accurate and reliable prediction results. This paper designs a CTR prediction framework combining feature selection and feature interaction. In this framework, a CTR prediction model based on Bayesian deep learning is proposed to quantify the uncertainty in the prediction model. On the squeeze network and DNN parallel prediction model framework, the approximate posterior parameter distribution of the model is obtained using the Monte Carlo dropout, and obtains the integrated prediction results. Epistemic and aleatoric uncertainty are defined and adopt information entropy to calculate the sum of the two kinds of uncertainties. Epistemic uncertainty could be measured by mutual information. Experimental results show that the model proposed is superior to other models in terms of prediction performance and has the ability to quantify uncertainty. [ABSTRACT FROM AUTHOR]

Recommender systems for requirements are typically built on the assumption that similar requirements can be used as proxies to retrieve similar software. When a stakeholder proposes a new requirement, natural language processing (NLP)-based similarity metrics can be exploited to retrieve existing requirements, and in turn, identify previously developed code. Several NLP approaches for similarity computation between requirements are available. However, there is little empirical evidence on their effectiveness for code retrieval. This study compares different NLP approaches, from lexical ones to semantic, deep-learning techniques, and correlates the similarity among requirements with the similarity of their associated software. The evaluation is conducted on real-world requirements from two industrial projects from a railway company. Specifically, the most similar pairs of requirements across two industrial projects are automatically identified using six language models. Then, the trace links between requirements and software are used to identify the software pairs associated with each requirements pair. The software similarity between pairs is then automatically computed with JPLag. Finally, the correlation between requirements similarity and software similarity is evaluated to see which language model shows the highest correlation and is thus more appropriate for code retrieval. In addition, we perform a focus group with members of the company to collect qualitative data. Results show a moderately positive correlation between requirements similarity and software similarity, with the pre-trained deep learning-based BERT language model with preprocessing outperforming the other models. Practitioners confirm that requirements similarity is generally regarded as a proxy for software similarity. However, they also highlight that additional aspect comes into play when deciding software reuse, e.g., domain/project knowledge, information coming from test cases, and trace links. Our work is among the first ones to explore the relationship between requirements and software similarity from a quantitative and qualitative standpoint. This can be useful not only in recommender systems but also in other requirements engineering tasks in which similarity computation is relevant, such as tracing and change impact analysis. [ABSTRACT FROM AUTHOR]

The internet and social networks produce an increasing amount of data. There is a serious necessity for a recommendation system because exploring through the huge collection is time-consuming and difficult. In this study, a multi-modal classifier is introduced which makes use of the output from dual deep neural networks: GRU for text analysis and Faster R-CNN for image analysis. These two networks reduce overall complexity with minimal computational time while retaining accuracy. More precisely, the GRU network is utilized to process movie reviews and the Faster RCNN is used to recognize each frames of the movie trailers. Gated Recurrent Unit (GRU) is a well-known variety of RNN that computes sequential data across recurrent structures. Faster RCNN is an enhanced version of Fast RCNN, it combines with the rectangular region proposals and with the features is extract by the ResNet-101. Initially, the trailer of the movie is manually splitted into frames and these frames are pre-processed using fuzzy elliptical filter for image analysis and the movie reviews are also tokenized for text analysis. The pre-processed text is taken as an input for GRU to classify offensive and non-offensive movies and the pre-processed images are taken as an input for Faster R-CNN to classify violence and non- violence movies based on the extracted features from the movie trailer. Afterwards, the four classified outputs are given as input for fuzzy decision-making unit for recommending best movies based on the Mamdani fuzzy inference system with gauss membership functions. The performance of the dual deep neural networks was evaluated using the specific parameters like specificity, precision, recall, accuracy and F1 score measures. The proposed GRU yields accuracy range of 97.73% for reviews and FRCNN yields the accuracy range of 98.42% for movie trailer. [ABSTRACT FROM AUTHOR]

In modern days, making recommendation for news articles poses a great challenge due to vast amount of online information. However, providing personalized recommendations from news articles, which are the sources of condense textual information is not a trivial task. A recommendation system needs to understand both the textual information of a news article, and the user contexts in terms of long-term and temporary preferences via the user's historic records. Unfortunately, many existing methods do not possess the capability to meet such need. In this work, we propose a neural deep news recommendation model called CupMar, that not only is able to learn the user-profile representation in different contexts, but also is able to leverage the multi-aspects properties of a news article to provide accurate, personalized news recommendations to users. The main components of our CupMar approach include the News Encoder and the User-Profile Encoder. Specifically, the News Encoder uses multiple properties such as news category, knowledge entity, title and body content with advanced neural network layers to derive informative news representation, while the User-Profile Encoder looks through a user's browsed news, infers both of her long-term and recent preference contexts to encode a user representation, and finds the most relevant candidate news for her. We evaluate our CupMar model with extensive experiments on the popular Microsoft News Dataset (MIND), and demonstrate the strong performance of our approach. [ABSTRACT FROM AUTHOR]

Recommendation systems, the best way to deal with information overload, are widely utilized to provide users with personalized content and services with high efficiency. Many recommendation algorithms have been researched and deployed extensively in various e-commerce applications, including the movie streaming services over the last decade. However, sparse data cold-start problems are often encountered in many movie recommendation systems. In this paper, we reported a personalized multimodal movie recommendation system based on multimodal data analysis and deep learning. The real-world MovieLens datasets were selected to test the effectiveness of our new recommendation algorithm. With the input information, the hidden features of the movies and the users were mined using deep learning to build a deep-learning network algorithm model for training to further predict movie scores. With a learning rate of 0.001, the root mean squared error (RMSE) scores achieved 0.9908 and 0.9096 for test sets of MovieLens 100 K and 1 M datasets, respectively. The scoring prediction results show improved accuracy after incorporating the potential features and connections in multimodal data with deep-learning technology. Compared with the traditional collaborative filtering algorithms, such as user-based collaborative filtering (User-CF), item-based content-based filtering (Item-CF), and singular-value decomposition (SVD) approaches, the multimodal movie recommendation system using deep learning could provide better personalized recommendation results. Meanwhile, the sparse data problem was alleviated to a certain degree. We suggest that the recommendation system can be improved through the combination of the deep-learning technology and the multimodal data analysis. [ABSTRACT FROM AUTHOR]

Air quality plays a vital role in people's health, and air quality forecasting can assist in decision making for government planning and sustainable development. In contrast, it is challenging to multi-step forecast accurately due to its complex and nonlinear caused by both temporal and spatial dimensions. Deep models, with their ability to model strong nonlinearities, have become the primary methods for air quality forecasting. However, because of the lack of mechanism-based analysis, uninterpretability forecasting makes decisions risky, especially when the government makes decisions. This paper proposes an interpretable variational Bayesian deep learning model with information self-screening for PM2.5 forecasting. Firstly, based on factors related to PM2.5 concentration, e.g., temperature, humidity, wind speed, spatial distribution, etc., an interpretable multivariate data screening structure for PM2.5 forecasting was established to catch as much helpful information as possible. Secondly, the self-screening layer was implanted in the deep learning network to optimize the selection of input variables. Further, following implantation of the screening layer, a variational Bayesian gated recurrent unit (GRU) network was constructed to overcome the complex distribution of PM2.5 and achieve accurate multi-step forecasting. The high accuracy of the proposed method is verified by PM2.5 data in Beijing, China, which provides an effective way, with multiple factors for PM2.5 forecasting determined using deep learning technology. [ABSTRACT FROM AUTHOR]

Recommender system or recommendation system is becoming an increasingly important technology on e-commerce websites to help users find products that suit their preferences. However, the growing number of users and products makes finding the right product difficult. Therefore, this study aims to develop a recommender system on e-commerce websites using collaborative filtering and deep learning techniques. Collaborative filtering is used to find similarities between users based on their preferences, while deep learning is used to improve the performance of the recommender system in generating more accurate recommendations. The test method is carried out by comparing the performance of the recommender system developed with the recommender system that already exists on the e-commerce website. The results of the test show that the recommender system developed is able to provide recommendations that are more accurate and more in line with user preferences compared to the existing recommender system. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the existing Point of Interest (POI) recommendation model in social network big data is difficult to extract deep feature information, a POI recommendation model based on deep learning in social networks and big data is proposed in this article. The input data are all gathered through intelligent sensors to apply some raw data pre-processing tasks and thus reduce the computational burden on the model. First, a POI static feature extraction method based on symmetric matrix decomposition is designed to capture the geographical location and POI category features in Location-Based Social Networking (LBSN). Second, the improved Continuous Bags-of-Words (CBOW) model is used to extract the semantic features in the user comment information, and realize the implicit vector representation of POI in geographic, category, semantic and temporal feature space. Finally, by adaptively selecting relevant check-in activities from the check-in history to learn and change user preferences, the Geographical-Spatiotemporal Gated Recurrent Unit Network (GSGRUN) can distinguish the user preferences of different check-in. Experiments show that when the length of the recommendation list is 15, the precision of the proposed algorithm on the loc-Gowalla data set is 0.0686, the recall is 0.0769, and the precision on the loc-Brightkite data set is 0.0659, the recall is 0.0835, both of which are better than the comparative recommendation methods. Therefore, compared with the comparison methods, the proposed method can significantly improve the performance of the POI recommendation system. [ABSTRACT FROM AUTHOR]

Recommender system can find information that satisfies the user′s individual needs from huge amount of information. With the development of deep learning, deep learning has been widely applied in recommender systems. CTR prediction plays an important role in recommender system and has been widely used in many fields such as personalized recommendation, information retrieval, online advertising and so on. For the issue of large and sparse data in recommender system, this study merges xDeepFM model with attention network, and proposes a new CTR prediction model based on deep learning, which is called Atte-xDeepFM model. This model can solve the issue of feature scarcity, effectively learn the interactions relationship between features, and does not need to manually extract useful information in feature engineering. The comparative experiments on Avazu and Criteo data sets prove the effectiveness of the proposed model. Compared with the algorithm model commonly used in CTR prediction, the proposed model has better recommendation effect. [ABSTRACT FROM AUTHOR]

Though relatively good effect has been achieved by the image de‐blurring method based on deep learning, the existing methods still suffer from the problem of unclear restoration of the edges. Therefore, brain‐inspired image restoration model based on human attention and "fine vision" is proposed to improve the blind restoration quality of the image in this paper according to the response mechanism of the different cerebral cortices for high and low spatial resolutions. The designed brain‐inspired model consists of dual‐channel network available to realize the function of feature merger for low and high resolutions, which is used to extract the image edges with detailed information filtered out. Confirmatory experiment is implemented based on the blurred image in the data set of GOPRO, LIVE and set14. As per the result, the model proposed is available for relatively good restoration of blurred image and super‐resolution, as well as looking results by visual inspection. [ABSTRACT FROM AUTHOR]

Generative Adversarial Networks (GANs) has achieved great success in computer vision like Image Inpainting, Image Super-Resolution. Many researchers apply it to improve the effectiveness of recommendation system. However, GANs-based methods obtain users' preferences using a single Neural Network framework in generative model, which may not be fully mined. Furthermore, most GANs-based algorithms adopt cross-entropy loss to get pair-wise bias, but these methods don't reveal global data distribution loss when data are sparse. Those problems will influence the performance of the algorithm and result in poor accuracy. To address these problems, we introduce Wide & Deep Generative Adversarial Networks for Recommendation System (a.k.a W & DGAN) in this paper. On the one hand, we employ Wide & Deep Learning as a generative model capable of extracting both explicit and implicit information of user preferences. Furthermore, we combine Cross-Entropy loss in G with Wasserstein loss in D to get data distribution, then, the joint loss will be to receive the training information feedback from data distribution. Empirical results on three public benchmarks show that W&DGAN significantly outperforms state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The sequential recommendation (also known as the next-item recommendation), which aims to predict the following item to recommend in a session according to users’ historical behavior, plays a critical role in improving session-based recommender systems. Most of the existing deep learning-based approaches utilize the recurrent neural network architecture or self-attention to model the sequential patterns and temporal influence among a user’s historical behavior and learn the user’s preference at a specific time. However, these methods have two main drawbacks. First, they focus on modeling users’ dynamic states from a user-centric perspective and always neglect the dynamics of items over time. Second, most of them deal with only the first-order user-item interactions and do not consider the high-order connectivity between users and items, which has recently been proved helpful for the sequential recommendation. To address the above problems, in this article, we attempt to model user-item interactions by a bipartite graph structure and propose a new recommendation approach based on a Position-enhanced and Time-aware Graph Convolutional Network (PTGCN) for the sequential recommendation. PTGCN models the sequential patterns and temporal dynamics between user-item interactions by defining a position-enhanced and time-aware graph convolution operation and learning the dynamic representations of users and items simultaneously on the bipartite graph with a self-attention aggregator. Also, it realizes the high-order connectivity between users and items by stacking multi-layer graph convolutions. To demonstrate the effectiveness of PTGCN, we carried out a comprehensive evaluation of PTGCN on three real-world datasets of different sizes compared with a few competitive baselines. Experimental results indicate that PTGCN outperforms several state-of-the-art sequential recommendation models in terms of two commonly-used evaluation metrics for ranking. In particular, it can make a better trade-off between recommendation performance and model training efficiency, which holds great potential for online session-based recommendation scenarios in the future. [ABSTRACT FROM AUTHOR]

To address the problem of poor recommendation accuracy in the session recommendation model, this paper proposed a capsule-based graph convolutional disentanglement for session-aware recommendation (CGCD) method.Specifically, it used disentangled learning technique to transition item embedding into factor embedding based on multiple sub-channels. Then, it used a capsule dynamic fusion strategy to aggregate different factors to obtain a new item embedding.In addition, it used a multi-head attention mechanism to assign weights to each item in the session. Finally, it aggregated the item embedding with other items in the current session according to the assigned weights, and then generated an accurate session representation to achieve item recommendation. Experiments on two publicly available real datasets show that the proposed model can improve the Pre@10, Pre@20, MRR@10 and MRR@20 by 5.17%, 2.99%, 6.56% and 2.94% on average of the recommended performance, which verifies the effectiveness and efficiency of the method in this paper. [ABSTRACT FROM AUTHOR]

In order to improve the safety of pedestrian interaction in complex traffic scenes, this paper proposes a pedestrian trajectory prediction algorithm SAN-GAN (Social Angle Norm-GAN) based on Social-GAN network (Social-Generative Adversarial Network). The algorithm first takes the pedestrian’s historical location information and head information as input, obtains the hidden feature information of the pedestrian through the trajectory generator LSTM network (Long Short Term Memory Networks), captures the dynamic changes of the pedestrian’s visual field based on the pedestrian visual field module, builds a fan-shaped visual field for all pedestrians and filters the valid information, thus driving the neural network model to predict future trajectory changes of pedestrians. By comparing SAN-GAN with LSTM, Social-LSTM (Social-Long Short Term Memory Networks), Social-GAN and other trajectory prediction algorithms, The results show that the SAN-GAN algorithm reduces the ADE by an average of 65.8%, 51.2% and 10.7%, and the FDE by an average of 73.6%, 60.9% and 10.4%, respectively, in predicting the pedestrian trajectory for 3.2 s. The SAN-GAN is effective in predicting the future trajectory of pedestrians interacting in complex traffic environments compared to other algorithms. [ABSTRACT FROM AUTHOR]

As a technique for accelerating and stabilizing training, the batch normalization (BN) is widely used in deep learning. However, BN cannot effectively estimate the mean and the variance of samples when training/fine-tuning with small batches of data on resource-constrained devices. It will lead to a decrease in the accuracy of the deep learning model. In the fruit fly olfactory system, the algorithm based on the "negative image" habituation model can filter redundant information and improve numerical stability. Inspired by the circuit mechanism, we propose a novel normalization method, the habituation normalization (HN). HN first eliminates the "negative image" obtained by habituation and then calculates the statistics for normalizing. It solves the problem of accuracy degradation of BN when the batch size is small. The experiment results show that HN can speed up neural network training and improve the model accuracy on vanilla LeNet-5, VGG16, and ResNet-50 in the Fashion MNIST and CIFAR10 datasets. Compared with four standard normalization methods, HN keeps stable and high accuracy in different batch sizes, which shows that HN has strong robustness. Finally, the applying HN to the deep learning-based EEG signal application system indicates that HN is suitable for the network fine-tuning and neural network applications under limited computing power and memory. [ABSTRACT FROM AUTHOR]

Next basket recommendation is a critical task in market basket data analysis. It is particularly important in grocery shopping, where grocery lists are an essential part of shopping habits of many customers. In this work, we first present a new grocery Recommender System available on the MyGroceryTour platform. Our online system uses different traditional machine learning (ML) and deep learning (DL) algorithms, and provides recommendations to users in a real-time manner. It aims to help Canadian customers create their personalized intelligent weekly grocery lists based on their individual purchase histories, weekly specials offered in local stores, and product cost and availability information. We perform clustering analysis to partition given customer profiles into four non-overlapping clusters according to their grocery shopping habits. Then, we conduct computational experiments to compare several traditional ML algorithms and our new DL algorithm based on the use of a gated recurrent unit (GRU)-based recurrent neural network (RNN) architecture. Our DL algorithm can be viewed as an extension of DREAM (Dynamic REcurrent bAsket Model) adapted to multi-class (i.e. multi-store) classification, since a given user can purchase recommended products in different grocery stores in which these products are available. Among traditional ML algorithms, the highest average F-score of 0.516 for the considered data set of 831 customers was obtained using Random Forest, whereas our proposed DL algorithm yielded the average F-score of 0.559 for this data set. The main advantage of the presented Recommender System is that our intelligent recommendation is personalized, since a separate traditional ML or DL model is built for each customer considered. Such a personalized approach allows us to outperform the prediction results provided by general state-of-the-art DL models. [ABSTRACT FROM AUTHOR]

Path-based relational reasoning over knowledge graphs has become increasingly popular due to a variety of downstream applications such as question answering in dialogue systems, fact prediction, and recommendation systems. In recent years, reinforcement learning (RL) based solutions for knowledge graphs have been demonstrated to be more interpretable and explainable than other deep learning models. However, the current solutions still struggle with performance issues due to incomplete state representations and large action spaces for the RL agent. We address these problems by developing HRRL (Heterogeneous Relational reasoning with Reinforcement Learning), a type-enhanced RL agent that utilizes the local heterogeneous neighborhood information for efficient path-based reasoning over knowledge graphs. HRRL improves the state representation using a graph neural network (GNN) for encoding the neighborhood information and utilizes entity type information for pruning the action space. Extensive experiments on real-world datasets show that HRRL outperforms state-of-the-art RL methods and discovers more novel paths during the training procedure, demonstrating the explorative power of our method. • Proposing a method to address the multi-hop reasoning problem. • Improving the choice of the action using the heterogeneous context. • Leveraging the neighborhood context for improving the reasoning precision. • Comprehensive evaluation of our model against state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Most web-based educational systems contain some drawbacks, as compared to traditional classrooms. Particularly, it becomes difficult for teachers to guide students to choose an appropriate learning resource due to the large number of online learning resources. Meanwhile, student decisions make it more difficult to choose educational resources according to their circumstances. In this matter, the resource recommender system can be employed as an educational environment to recommend the educational resource advice for students, so that these recommendations can be coordinated to each student's preferences and needs. This paper presents the resource recommender system as a combination of MLP, BiLSTM, and LSTM improved deep learning networks using the attention method. Compared to similar studies conducted using DBN networks and focus only on the near past interests and preferences of users, the proposed system provides higher accuracy and more appropriate recommendations considering current interests, in addition to the user's long-term past interests. The proposed recommender system with accuracy of 0.96 and a loss of 0.0822 contains a better performance to recommend resources to students compared to other methods. [ABSTRACT FROM AUTHOR]

In recent years, hybrid recommendation techniques based on feature fusion have gained extensive attention in the field of list ranking. Most of them fuse linear and nonlinear models to simultaneously learn the linear and nonlinear features of entities and jointly fit user-item interactions. These methods are based on implicit feedback, which can reduce the difficulty of data collection and the time of data preprocessing, but will lead to the lack of entity interaction depth information due to the lack of user satisfaction. This is equivalent to artificially reducing the entity interaction features, limiting the overall performance of the model. To address this problem, we propose a two-stage recommendation model named A-DNR, short for Attention-based Deep Neural Ranking. In the first stage, user short-term preferences are modeled through an attention mechanism network. Then the user short-term preferences and user long-term preferences are fused into dynamic user preferences. In the second stage, the high-order and low-order feature interactions are modeled by a matrix factorization (MF) model and a multi-layer perceptron (MLP) model, respectively. Then, the features are fused through a fully connected layer, and the vectors are mapped to scores. Finally, a ranking list is output through the scores. Experiments on three real-world datasets (Movielens100K, Movielens1M and Yahoo Movies) show that our proposed model achieves significant improvements compared to existing methods. [ABSTRACT FROM AUTHOR]

This paper introduces versatile filters to construct efficient convolutional neural networks that are widely used in various visual recognition tasks. Considering the demands of efficient deep learning techniques running on cost-effective hardware, a number of methods have been developed to learn compact neural networks. Most of these works aim to slim down filters in different ways, e.g., investigating small, sparse or quantized filters. In contrast, we treat filters from an additive perspective. A series of secondary filters can be derived from a primary filter with the help of binary masks. These secondary filters all inherit in the primary filter without occupying more storage, but once been unfolded in computation they could significantly enhance the capability of the filter by integrating information extracted from different receptive fields. Besides spatial versatile filters, we additionally investigate versatile filters from the channel perspective. Binary masks can be further customized for different primary filters under orthogonal constraints. We conduct theoretical analysis on network complexity and an efficient convolution scheme is introduced. Experimental results on benchmark datasets and neural networks demonstrate that our versatile filters are able to achieve comparable accuracy as that of original filters, but require less memory and computation cost. [ABSTRACT FROM AUTHOR]