Search

Click-through rate (CTR) prediction, whose goal is to predict the probability of the user to click on an item, has become increasingly significant in the recommender systems. Recently, some deep learning models with the ability to automatically extract the user interest from his/her behaviors have achieved great success. In these work, the attention mechanism is used to select the user interested items in historical behaviors, improving the performance of the CTR predictor. Normally, these attentive modules can be jointly trained with the base predictor by using gradient descents. In this paper, we regard user interest modeling as a feature selection problem, which we call user interest selection. For such a problem, we propose a novel approach under the framework of the wrapper method, which is named Meta-Wrapper. More specifically, we use a differentiable module as our wrapping operator and then recast its learning problem as a continuous bilevel optimization. Moreover, we use a meta-learning algorithm to solve the optimization and theoretically prove its convergence. Meanwhile, we also provide theoretical analysis to show that our proposed method 1) efficiencies the wrapper-based feature selection, and 2) achieves better resistance to overfitting. Finally, extensive experiments on three public datasets manifest the superiority of our method in boosting the performance of CTR prediction. [ABSTRACT FROM AUTHOR]

Collaborative filtering (CF) is a classical model for recommender systems. Though neural network-based CF models have been shown to perform well in some cases, they still suffer from the sparsity and cold-start issues. To address these issues, we propose a novel neural network-based CF model, termed the cross-domain explicit–implicit-mixed CF neural network (CEICFNet). The proposed model utilizes deep neural networks to learn latent factors not only from the explicit ratings and the implicit interactions but also in a cross-domain manner. In particular, domain-shared multilayer perception (MLP) networks are designed to learn the user rating latent factors and the user interaction latent factors from the explicit ratings and the implicit interactions, respectively, which act as bridges for transferring knowledge across domains. On the other hand, domain-specific MLP networks are designed to learn the item rating latent factors and the item interaction latent factors from the explicit ratings and the implicit interactions, respectively. Then, in each domain, based on the user rating latent factors and the item rating latent factors, the rating predictive representation for each user–item pair can be learned by an MLP. Similarly, the interaction predictive representation for each user–item pair can be learned. For integrating the explicit ratings and the implicit interactions, in each domain, a fully connected layer is used to automatically assign different weights to the rating predictive representations and the interaction predictive representations, based on which the final interaction probability can be generated. Extensive experiments are conducted on five datasets and the results have confirmed the effectiveness of our model. [ABSTRACT FROM AUTHOR]

Semantic relationships between words provide relevant information about the whole idea in the texts. Existing embedding representation models characterize each word as a vector of numbers with a fixed length. These models have been used in tasks involving text classification, such as recommendation and question–answer systems. However, the embedded information provided by semantic relationships has been neglected. Therefore, this paper proposes an approach that involves semantic relationships in embedding models for text classification, which is evaluated. Three embedding models based on semantic relations extracted from Wikipedia are presented and compared with existing word-based models. Our approach considers the following relationships: synonymy, hyponymy, and hyperonymy. They were considered since previous experiments have shown that they provide semantic knowledge. The relationships are extracted from Wikipedia using lexical-syntactic patterns identified in the literature. The extracted relationships are embedded as a vector: synonymy, hyponymy–hyperonymy, and a combination of all relationships. A Convolutional Neural Network using semantic relationship embeddings was trained for text classification. An evaluation was carried out for the proposed relationship embedding configurations and existing word-based models to compare them based on two corpora. The results were obtained with the metrics of precision, accuracy, recall, and F 1 -measure. The best results for the 20-Newsgroup corpus were obtained with the hyponymy–hyperonymy embeddings, achieving an accuracy of 0.79. For the Reuters corpus, F 1 -measure and recall of 0.87 were obtained using synonymy–hyponymy–hyperonymy. [ABSTRACT FROM AUTHOR]

Deep learning has dominated the methodology of image inpainting in recent years. However, many of the existing methods still suffer from the defects of unnatural structure and blurry textures when filling large-area holes. To address this issue, we propose a novel coarse-to-fine residual inpainting framework: we first reconstruct the downsampled low-frequency coarse profile at a low computational cost, we then solely focus on generating high-frequency details, which will be added as residuals to the coarse profile, such that structure and texture details in composited result can be better preserved. Meanwhile, we notice that the existing attention mechanism used to build long-distance dependency is unreasonable: either does not adapt to free-form holes, or lacks filtering for invalid information sources, which could compromise feature map reconstruction. To solve this, we propose an improved Selective Free-Form Attention (SFFA) module, selectively allowing the information flow from the valid sources to hole area with free-form shape. We verify the incremental effectiveness of our contributions. We further show that our model outperforms recent state-of-the-art methods both qualitatively and quantitatively on benchmark datasets. [ABSTRACT FROM AUTHOR]

Recommender systems are information filtering tools that seek to match customers with products or services of interest. Most of the prevalent collaborative filtering recommender systems, such as matrix factorization and AutoRec, suffer from the "cold‐start" problem, where they fail to provide meaningful recommendations for new users or new items due to informative‐missing from the training data. To address this problem, we propose a weighted AutoEncoding model to leverage information from other users or items that share similar characteristics. The proposed method provides an effective strategy for borrowing strength from user or item‐specific clustering structure as well as pairwise similarity in the training data, while achieving high computational efficiency and dimension reduction, and preserving nonlinear relationships between user preferences and item features. Simulation studies and applications to three real datasets show advantages in prediction accuracy of the proposed model compared to current state‐of‐the‐art approaches. [ABSTRACT FROM AUTHOR]

When a user connects to the Internet to fulfill his needs, he often encounters a huge amount of related information. Recommender systems are the techniques for massively filtering information and offering the items that users find them satisfying and interesting. The advances in machine learning methods, especially deep learning, have led to great achievements in recommender systems, although these systems still suffer from challenges such as cold‐start and sparsity problems. To solve these problems, context information such as user communication network is usually used. In this article, we have proposed a novel recommendation method based on matrix factorization and graph analysis methods, namely Louvain for community detection and HITS for finding the most important node within the trust network. In addition, we leverage deep autoencoders to initialize users and items latent factors, and the Node2vec deep embedding method gathers users' latent factors from the user trust graph. The proposed method is implemented on Ciao and Epinions standard datasets. The experimental results and comparisons demonstrate that the proposed approach is superior to the existing state‐of‐the‐art recommendation methods. Our approach outperforms other comparative methods and achieves great improvements, that is, 15.56% RMSE improvement for Epinions and 18.41% RMSE improvement for Ciao. [ABSTRACT FROM AUTHOR]

Due to increase of data in online web resources Recommendation System plays a vital role. In many platforms such as music, movie, books, videos, ecommerce for recommending the useful data. Video recommendation system provides users a very highly satisfying recommendations which in turn increase the user stickiness to the site. A personalized recommendation system with collaborative filtering technique is implemented to increase the user satisfaction. This method is used to recommend movies to users depending on their historical data on the streaming site and also by exploring the content watched by the users with similar preferences. A personalized recommendation system using collaborative filtering with multi-layer perceptron is implemented to increase the accuracy of the system. [ABSTRACT FROM AUTHOR]

Business-based decision support systems have been proposed for a few decades in the e-commerce and textile industries. However, these Decision Support Systems (DSS) have not been so productive in terms of business decision delivery. In our proposed model, we introduce a content-based image retrieval model based on a DSS and recommendations system for the textile industry, either offline or online. We used the Fashion MNIST dataset developed by Zalando to train our deep learning model. Our proposed hybrid model can demonstrate how a DSS can be integrated with a system that can separate customers based on their personal characteristics in order to tailor recommendations of products using behavioral analytics, which is trained based on MBTI personality data and Deap EEG data containing numerous pre-trained EEG brain waves. With this hybrid, a DSS can also show product usage analytics. Our proposed model has achieved the maximum accuracy compared to other proposed state-of-the-art models due to its qualitative analysis. In the first section of our analysis, we used a deep learning algorithm to train our CBIR model based on different classifiers such as VGG-net, Inception-Net, and U-net which have achieved an accuracy of 98.2% with a 2% of minimized error rate. The result was validated using different performance metrics such as F-score, F-weight, Precision, and Recall. The second part of our model has been tested on different machine learning algorithms with an accuracy rate of 89.9%. Thus, the entire model has been trained, validated, and tested separately to gain maximum efficiency. Our proposal for a DSS system, which integrates several subsystems with distinct functional sets and several model subsystems, is what makes this study special. Customer preference is one of the major problems facing merchants in the textile industry. Additionally, it can be extremely difficult for retailers to predict customer interests and preferences to create products that fulfill those needs. The three innovations presented in this work are a conceptual model for personality characterization, utilizing an amalgamation of an ECG classification model, a suggestion for a textile image retrieval model using Denoising Auto-Encoder, and a language model based on the MBTI for customer rating. Additionally, we have proposed a section showing how blockchain integration in data pre-processing can enhance its security and AI-based software quality assurance in a multi-model system. [ABSTRACT FROM AUTHOR]

Pulmonary diseases are life-threatening diseases commonly observed worldwide, and timely diagnosis of these diseases is essential. Meanwhile, increased use of Convolution Neural Networks has promoted the advancement of computer-assisted clinical recommendation systems for diagnosing diseases using chest radiographs. The texture and shape of the tissues in the diagnostic images are essential aspects of prognosis. Therefore, in the latest studies, the vast set of images with a larger resolution is paired with deep learning techniques to enhance the performance of the disease diagnosis in chest radiographs. Moreover, pulmonary diseases have irregular and different sizes; therefore, several studies sought to add new components to existing deep learning techniques for acquiring multi-scale imaging features from diagnostic chest X-rays. However, most of the attempts do not consider the computation overhead and lose the spatial details in an effort to capture the larger receptive field for obtaining the discriminative features from high-resolution chest X-rays. In this paper, we propose an explainable and lightweight Multi-Scale Chest X-ray Network (MS-CheXNet) to predict abnormal diseases from the diagnostic chest X-rays. The MS-CheXNet consists of four following main subnetworks: (1) Multi-Scale Dilation Layer (MSDL), which includes multiple and stacked dilation convolution channels that consider the larger receptive field and captures the variable sizes of pulmonary diseases by obtaining more discriminative spatial features from the input chest X-rays; (2) Depthwise Separable Convolution Neural Network (DS-CNN) is used to learn imaging features by adjusting lesser parameters compared to the conventional CNN, making the overall network lightweight and computationally inexpensive, making it suitable for mobile vision tasks; (3) a fully connected Deep Neural Network module is used for predicting abnormalities from the chest X-rays; and (4) Gradient-weighted Class Activation Mapping (Grad-CAM) technique is employed to check the decision models' transparency and understand their ability to arrive at a decision by visualizing the discriminative image regions and localizing the chest diseases. The proposed work is compared with existing disease prediction models on chest X-rays and state-of-the-art deep learning strategies to assess the effectiveness of the proposed model. The proposed model is tested with a publicly available Open-I Dataset and data collected from a private hospital. After the comprehensive assessment, it is observed that the performance of the designed approach showcased a 7% to 18% increase in accuracy compared to the existing method. [ABSTRACT FROM AUTHOR]

CTR (Click-Through Rate) prediction has attracted more and more attention from academia and industry for its significant contribution to revenue. In the last decade, learning feature interactions have become a mainstream research direction, and dozens of feature interaction-based models have been proposed for the CTR prediction task. The most common approach for existing models is to enumerate all possible feature interactions or to learn higher-order feature interactions by designing complex models. However, a simple enumeration will introduce meaningless and harmful interactions, and a complex model structure will bring a higher complexity. In this work, we propose a lightweight, yet effective model called the Gated Adaptive feature Interaction Network (GAIN). We devise a novel cross module to drop meaningless feature interactions and preserve informative ones. Our cross module consists of multiple gated units, each of which can independently learn an arbitrary-order feature interaction. We combine the cross module with a deep module into GAIN and conduct comparative experiments with state-of-the-art models on two public datasets to verify its validity. Our experimental results show that GAIN can achieve a comparable or even better performance compared to its competitors. Furthermore, in order to verify the effectiveness of the feature interactions learned by GAIN, we transfer learned interactions to other models, such as Logistic Regression (LR) and Factorization Machines (FM), and find out that their performance can be significantly improved. [ABSTRACT FROM AUTHOR]

Background: Twitter is a popular social networking site where short messages or "tweets" of users have been used extensively for research purposes. However, not much research has been done in mining the medical professions, such as detecting the occupations of users from their biographical contents. Mining such professions can be used to build efficient recommender systems for cost-effective targeted advertisements. Moreover, it is highly important to develop effective methods to identify the occupation of users since conventional classification methods rely on features developed by human intelligence. Although, the result may be favorable for the classification problem. However, it is still extremely challenging for traditional classifiers to predict the medical occupations accurately since it involves predicting multiple occupations. Hence this study emphasizes predicting the medical occupational class of users through their public biographical ("Bio") content. We have conducted our analysis by annotating the bio content of Twitter users. In this paper, we propose a method of combining word embedding with state-of-art neural network models that include: Long Short Term Memory (LSTM), Bidirectional LSTM, Gated Recurrent Unit, Bidirectional Encoder Representations from Transformers, and A lite BERT. Moreover, we have also observed that by composing the word embedding with the neural network models there is no need to construct any particular attribute or feature. By using word embedding, the bio contents are formatted as dense vectors which are fed as input into the neural network models as a sequence of vectors. Result: Performance metrics that include accuracy, precision, recall, and F1-score have shown a significant difference between our method of combining word embedding with neural network models than with the traditional methods. The scores have proved that our proposed approach has outperformed the traditional machine learning techniques for detecting medical occupations among users. ALBERT has performed the best among the deep learning networks with an F1 score of 0.90. Conclusion: In this study, we have presented a novel method of detecting the occupations of Twitter users engaged in the medical domain by merging word embedding with state-of-art neural networks. The outcomes of our approach have demonstrated that our method can further advance the process of analyzing corpora of social media without going through the trouble of developing computationally expensive features. [ABSTRACT FROM AUTHOR]

Background and Objective. Skin lesion classification by using deep learning technologies is still a considerable challenge due to high similarity among classes and large intraclass differences, serious class imbalance in data, and poor classification accuracy with low robustness. Approach. To address these issues, a two-stage framework for dermoscopy lesion classification using adversarial training and a fuzzy rank-based ensemble of multilayer feature fusion convolutional neural network (CNN) models is proposed. In the first stage, dermoscopy dataset augmentation based on generative adversarial networks is proposed to obtain realistic dermoscopy lesion images, enabling significant improvement for balancing the number of lesions in each class. In the second stage, a fuzzy rank-based ensemble of multilayer feature fusion CNN models is proposed to classify skin lesions. In addition, an efficient channel integrating spatial attention module, in which a novel dilated pyramid pooling structure is designed to extract multiscale features from an enlarged receptive field and filter meaningful information of the initial features. Combining the cross-entropy loss function with the focal loss function, a novel united loss function is designed to reduce the intraclass sample distance and to focus on difficult and error-prone samples to improve the recognition accuracy of the proposed model. Main results. In this paper, the common dataset (HAM10000) is selected to conduct simulation experiments to evaluate and verify the effectiveness of the proposed method. The subjective and objective experimental results demonstrate that the proposed method is superior over the state-of-the-art methods for skin lesion classification due to its higher accuracy, specificity and robustness. Significance. The proposed method effectively improves the classification performance of the model for skin diseases, which will help doctors make accurate and efficient diagnoses, reduce the incidence rate and improve the survival rates of patients. [ABSTRACT FROM AUTHOR]

The assortments of global retailers are composed of hundreds of thousands of products linked by several types of relationships such as style compatibility, "bought together", "watched together", etc. Graphs are a natural representation for assortments, where products are nodes and relations are edges. Style compatibility relations are produced manually and do not cover the whole graph uniformly. We propose to use inductive learning to enhance a graph encoding style compatibility of a fashion assortment, leveraging rich node information comprising textual descriptions and visual data. Then, we show how the proposed graph enhancement substantially improves the performance on transductive tasks with a minor impact on graph sparsity. Although demonstrated in a challenging and novel industrial application case, the approach we propose is general enough to be applied to any node-level or edge-level prediction task in very sparse, large-scale networks. [ABSTRACT FROM AUTHOR]

Simultaneous localization and mapping (SLAM) is a core technology for mobile robots working in unknown environments. Most existing SLAM techniques can achieve good localization accuracy in static scenes, as they are designed based on the assumption that unknown scenes are rigid. However, real-world environments are dynamic, resulting in poor performance of SLAM algorithms. Thus, to optimize the performance of SLAM techniques, we propose a new parallel processing system, named SOLO-SLAM, based on the existing ORB-SLAM3 algorithm. By improving the semantic threads and designing a new dynamic point filtering strategy, SOLO-SLAM completes the tasks of semantic and SLAM threads in parallel, thereby effectively improving the real-time performance of SLAM systems. Additionally, we further enhance the filtering effect for dynamic points using a combination of regional dynamic degree and geometric constraints. The designed system adds a new semantic constraint based on semantic attributes of map points, which solves, to some extent, the problem of fewer optimization constraints caused by dynamic information filtering. Using the publicly available TUM dataset, SOLO-SLAM is compared with other state-of-the-art schemes. Our algorithm outperforms ORB-SLAM3 in accuracy (maximum improvement is 97.16%) and achieves better results than Dyna-SLAM with respect to time efficiency (maximum improvement is 90.07%). [ABSTRACT FROM AUTHOR]

In this new era, it is no longer impossible to create a smart home environment around the household. Moreover, users are not limited to humans but also include pets such as dogs. Dogs need long-term close companionship with their owners; however, owners may occasionally need to be away from home for extended periods of time and can only monitor their dogs' behaviors through home security cameras. Some dogs are sensitive and may develop separation anxiety, which can lead to disruptive behavior. Therefore, a novel smart home solution with an affective recommendation module is proposed by developing: (1) an application to predict the behavior of dogs and, (2) a communication platform using smartphones to connect with dog friends from different households. To predict the dogs' behaviors, the dog emotion recognition and dog barking recognition methods are performed. The ResNet model and the sequential model are implemented to recognize dog emotions and dog barks. The weighted average is proposed to combine the prediction value of dog emotion and dog bark to improve the prediction output. Subsequently, the prediction output is forwarded to a recommendation module to respond to the dogs' conditions. On the other hand, the Real-Time Messaging Protocol (RTMP) server is implemented as a platform to contact a dog's friends on a list to interact with each other. Various tests were carried out and the proposed weighted average led to an improvement in the prediction accuracy. Additionally, the proposed communication platform using basic smartphones has successfully established the connection between dog friends. [ABSTRACT FROM AUTHOR]

In precision medicine, the ultimate goal is to recommend the most effective treatment to an individual patient based on patient‐specific molecular and clinical profiles, possibly high‐dimensional. To advance cancer treatment, large‐scale screenings of cancer cell lines against chemical compounds have been performed to help better understand the relationship between genomic features and drug response; existing machine learning approaches use exclusively supervised learning, including penalized regression and recommender systems. However, it would be more efficient to apply reinforcement learning to sequentially learn as data accrue, including selecting the most promising therapy for a patient given individual molecular and clinical features and then collecting and learning from the corresponding data. In this article, we propose a novel personalized ranking system called Proximal Policy Optimization Ranking (PPORank), which ranks the drugs based on their predicted effects per cell line (or patient) in the framework of deep reinforcement learning (DRL). Modeled as a Markov decision process, the proposed method learns to recommend the most suitable drugs sequentially and continuously over time. As a proof‐of‐concept, we conduct experiments on two large‐scale cancer cell line data sets in addition to simulated data. The results demonstrate that the proposed DRL‐based PPORank outperforms the state‐of‐the‐art competitors based on supervised learning. Taken together, we conclude that novel methods in the framework of DRL have great potential for precision medicine and should be further studied. [ABSTRACT FROM AUTHOR]

Recommender systems use different techniques of machine learning (ML) to suggest users and recommend service or entity in various field of application such as in health care recommender system (HRS). Due to the vast count of algorithms shown in the literature, HRS and various application sectors are now utilizing ML algorithms from the area of artificial intelligence. However, selecting an appropriate ML algorithm in the case of a health recommender system seems to be a time-consuming task. However the development of recommender system in different service domain faces problems of algorithms selection for better accuracy. This article examined the usage of ML techniques in recommender systems for health applications through a survey of the literature. The objectives of this article are (i) recognize the literature review finding of recommender system in health applications using ML and deep learning algorithms. (ii) Assist new researchers with the help of gap in previous research. The results of this study is to proposed new recommender system in health application of mosquito borne disease by using hybrid approach of ML technique. [ABSTRACT FROM AUTHOR]

With the advances in artificial intelligence, machines can now integrate information and extract its features to build a predictive model of likely outcomes, providing more insights to make better decisions. An artificial neural network, which is based on the structure and functions of the human brain, is a series of algorithms that can filter information into layers of recognizable characteristics. The computational power of neural networks is increasing in parallel with improved development of hardware devices. Artificial intelligence has become inseparable from our daily lives because it assists us in many areas. In recent years, researchers in sports have begun to incorporate the tools of artificial intelligence to complete many tasks that used to rely on substantial labor inputs. For instance, player and object trajectory tracking makes it easier to understand team play tactics and to analyze the strategies used in games. Motion recognition provides broadcasting and live-streaming programs with quick information on what skills are deployed. Data analysis can prevent athletes from sustaining injuries. Thanks to the continual improvements in artificial intelligence, more and more studies on its application in sports are expected. Language-teaching and referee robots are likely to be introduced in the near future. [ABSTRACT FROM AUTHOR]

Group recommendation has recently drawn a lot of attention to the recommender system community. Currently, several deep learning-based approaches are leveraged to learn preferences of groups for items and predict next items in which groups may be interested. Yet, their recommendation performance is still unsatisfactory due to sparse group–item interactions. To address this challenge, this study presents a novel model, called group recommendation model with two-stage deep learning (GRMTDL), which encompasses two sequential stages: 1) group representation learning (GRL) and 2) group preference learning (GPL). In GRL, we first construct an undirected tripartite graph over group–user–item interactions, and then employ it to accurately learn group semantic features through a spatial-based variational graph autoencoder network. While in GPL, we first introduce a dual PL-network that contains two structure-sharing subnetworks: 1) group PL-network employed for GPL and 2) user PL-network utilized for user preference learning. Then, we design a novel layered transfer learning (LTL) method to learn group preferences by alternately optimizing these two subnetworks. In particular, it can effectively absorb knowledge of user preferences into the process of GPL. Furthermore, extensive experiments on four real-world datasets demonstrate that the proposed GRMTDL model outperforms the state-of-the-art baselines for group recommendation. [ABSTRACT FROM AUTHOR]

The current graph-neural-network-based recommendation algorithm fully considers the interaction between users and items. It achieves better recommendation results, but due to a large amount of data, the interaction between users and items still suffers from the problem of data sparsity. To address this problem, we propose a method to alleviate the data sparsity problem by retaining user–item interactions while fully exploiting the association relationships between items and using side-information enhancement. We constructed a "twin-tower" model by combining a user–item training model and an item–item training model inspired by the knowledge distillation technique; the two sides of the structure learn from each other during the model training process. Comparative experiments were carried out on three publicly available datasets, using the recall and the normalized discounted cumulative gain as evaluation metrics; the results outperform existing related base algorithms. We also carried out extensive parameter sensitivity and ablation experiments to analyze the influence of various factors on the model. The problem of user–item interaction data sparsity is effectively addressed. [ABSTRACT FROM AUTHOR]

Recommender systems (RSs) have become increasingly vital in the modern information era and connected economy. They play a key role in business operations by generating personalized suggestions and minimizing information overload. However, the performance of traditional RSs is limited by data sparseness and cold-start issues. Though deep learning-based recommender systems (DLRSs) are very popular, they underperform when considering rating matrices with sparse entries. Despite their performance improvements, DLRSs also suffer from data sparsity, cold start, serendipity, and generalizability issues. We propose a multistage model that uses multimodal data embedding and deep transfer learning for effective and personalized product recommendations, and is designed to overcome data sparsity and cold-start issues. The proposed model includes two phases. In the first—offline—phase, a deep learning technique is implemented to learn hidden features from a large image dataset (targeting new item cold start), and a multimodal data embedding is used to produce dense user feature and item feature vectors (targeting user cold start). This phase produces three different similarity matrices that are used as inputs for the second—online—phase to generate a list of top-n relevant items for a target user. We analyzed the accuracy and effectiveness of the proposed model against the existing baseline RSs using a Brazilian E-commerce dataset. The results show that our model scored 0.5882 for MAE and 0.4011 for RMSE which is lower than baseline RSs which indicates that the model achieved an improved accuracy and was able to minimize the typical cold start and data sparseness issues during the recommendation process. [ABSTRACT FROM AUTHOR]

In recent times, nutrition recommendation system has gained increasing attention due to their need for healthy living. Current studies on the food domain deal with a recommendation system that focuses on independent users and their health problems but lack nutritional advice to individual users. The proposed system is developed to suggest nutritional food to people based on age and gender predicted from their face image. The designed methodology preprocesses the input image before performing feature extraction using the deep convolution neural network (DCNN) strategy. This network extracts D -dimensional characteristics from the source face image, followed by the feature selection strategy. The face's distinctive and identifiable traits are chosen utilizing a hybrid particle swarm optimization (HPSO) technique. Support vector machine (SVM) is used to classify a person's age and gender. The nutrition recommendation system relies on the age and gender classes. The proposed system is evaluated using classification rate, precision, and recall using Adience dataset and UTKface dataset, and real-world images exhibit excellent performance by achieving good prediction results and computation time. [ABSTRACT FROM AUTHOR]

In this paper, we present a new AI (Artificial Intelligence) edge platform, called "MiniDeep", which provides a standalone deep learning platform based on the cloud-edge architecture. This AI-Edge platform provides developers with a whole deep learning development environment to set up their deep learning life cycle processes, such as model training, model evaluation, model deployment, model inference, ground truth collecting, data pre-processing, and training data management. To the best of our knowledge, such a whole deep learning development environment has not been built before. MiniDeep uses Amazon Web Services (AWS) as the backend platform of a deep learning tuning management model. In the edge device, the OpenVino enables deep learning inference acceleration at the edge. To perform a deep learning life cycle job, MiniDeep proposes a mini deep life cycle (MDLC) system which is composed of several microservices from the cloud to the edge. MiniDeep provides Train Job Creator (TJC) for training dataset management and the models' training schedule and Model Packager (MP) for model package management. All of them are based on several AWS cloud services. On the edge device, MiniDeep provides Inference Handler (IH) to handle deep learning inference by hosting RESTful API (Application Programming Interface) requests/responses from the end device. Data Provider (DP) is responsible for ground truth collection and dataset synchronization for the cloud. With the deep learning ability, this paper uses the MiniDeep platform to implement a recommendation system for AI-QSR (Quick Service Restaurant) KIOSK (interactive kiosk) application. AI-QSR uses the MiniDeep platform to train an LSTM (Long Short-Term Memory)-based recommendation system. The LSTM-based recommendation system converts KIOSK UI (User Interface) flow to the flow sequence and performs sequential recommendations with food suggestions. At the end of this paper, the efficiency of the proposed MiniDeep is verified through real experiments. The experiment results have demonstrated that the proposed LSTM-based scheme performs better than the rule-based scheme in terms of purchase hit accuracy, categorical cross-entropy, precision, recall, and F1 score. [ABSTRACT FROM AUTHOR]

Most clothing recommendation methods have problems such as high resource consumption and inconsistent subjectively labeled clothing labels. Based on this, a multilabel classification algorithm based on deep learning (DL) theory is introduced, based on which the clothing style recognition model is constructed. Next, the concept of the decision tree algorithm is given, and the clothing recommendation model is built based on this algorithm. Moreover, the clothing style recognition model based on a multilabel classification algorithm and the clothing recommendation system based on a decision tree algorithm are tested by building simulation experiments and combining neural network technology. Finally, the application of the decision tree algorithm and DL theory in clothing recommendation design is studied through the literature collection method. The research focus is to realize the recognition of clothing through decision tree algorithm and DL method to achieve the intelligent recommendation of clothing style. The results show that: (1) the neural network technology in DL theory can realize efficient recognition and classification of clothing style by automatically extracting image features and combining with a multilabel classification algorithm. (2) The decision tree algorithm can make an initial recommendation according to users' style preferences, then make implicit recommendations through user retrieval, browsing, and other operations, and make dynamic clothing style recommendations to users. (3) When the neural network based on a multilabel classification algorithm is trained, the precision, recall rate, and F1 values are 0.73, 0.43, and 0.55, respectively. (4) After using the clothing recommendation system based on the decision tree algorithm, the subjects' average satisfaction is 86.25%, indicating that this system can give users a better clothing recommendation experience. This exploration aims to provide a crucial reference for further improving the quality of clothing recommendation services. It has important theoretical significance and practical value for the development of artificial intelligence in the field of fashion design, and is expected to provide a reference for the development of bionics. [ABSTRACT FROM AUTHOR]

Link prediction aims at inferring missing links or predicting future ones based on the currently observed network. This topic is important for many applications such as social media, bioinformatics and recommendation systems. Most existing methods focus on homogeneous settings and consider only low-order pairwise relations while ignoring either the heterogeneity or high-order complex relations among different types of nodes, which tends to lead to a sub-optimal embedding result. This paper presents a method named Heterogeneous Hypergraph Variational Autoencoder (HeteHG-VAE) for link prediction in heterogeneous information networks (HINs). It first maps a conventional HIN to a heterogeneous hypergraph with a certain kind of semantics to capture both the high-order semantics and complex relations among nodes, while preserving the low-order pairwise topology information of the original HIN. Then, deep latent representations of nodes and hyperedges are learned by a Bayesian deep generative framework from the heterogeneous hypergraph in an unsupervised manner. Moreover, a hyperedge attention module is designed to learn the importance of different types of nodes in each hyperedge. The major merit of HeteHG-VAE lies in its ability of modeling multi-level relations in heterogeneous settings. Extensive experiments on real-world datasets demonstrate the effectiveness and efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

Automatic music recommendation is an open research problem that has seen much work in recent years. A common and successful music recommendation approach is collaborative filtering, which has worked well in this domain. One major drawback of this method is that it suffers from a cold‐start problem, and it requires a lot of user‐personalized information. It is an ineffective mechanism for recommending new and unpopular songs as well as for new users. In this article, we report a hybrid methodology that uses the song's content information. We use MIDI (Musical Instrument Digital Interface) content data, a compressed version of an audio song that contains digital information about a song and is machine‐readable. We describe a model called MSA‐SRec (MIDI Based Self Attentive Sequential Music Recommendation), a latent factor‐based self‐attentive deep learning model that uses a substantial amount of sequential information as content information of the song for recommendation generation. We use MIDI data of a song that is under‐explored content information for music recommendation. We show that using MIDI as content data with user and item latent vector produces reasonable recommendations. We also demonstrate that using MIDI over other music metadata performs better with various state‐of‐the‐art models of recommendation systems. [ABSTRACT FROM AUTHOR]

The expansion of the population that wants to learn online is growing due to several e-learning platforms, which help innovate and suggest courses to learners. Several techniques are devised for determining optimal courses for the learner. In recent days, researchers began to utilize recommendation systems in e-learning. This paper devises a novel technique for course recommendation to students in an e-learning platform, which helps learners select the best course. Here, the Butterfly Weed Optimization (BWO) is newly devised by combining Invasive Weed Optimization (IWO) and Butterfly Optimization Algorithm (BOA). At first, the process is performed by inputting the data to the Course subscription matrix for constructing the matrix based on learner interest and courses. Here, course grouping is performed using Interval type-2 Fuzzy Local Enhancement Based Rough K-means Clustering. Furthermore, the course is matched with input data based on entropy and angular distance. Finally, the sentiment classification is performed using the Ontology-based approach SentiWordNet and Deep Neural Network (DNN). Here, the DNN is trained with the proposed BWO algorithm, and thus the course recommendation is attained by offering a suitable course recommendation to learners. [ABSTRACT FROM AUTHOR]

With the advent of the Internet era, the frequency and proportion of candidates obtaining recruitment information through the Internet is getting higher and higher, and the amount of human resource information, such as talent information and job information, has also increased unprecedentedly, which makes human resource services face information overload. Especially with the gradual increase of the amount of information, this method is not enough for the acquisition and classification of massive data. After that, experts developed search engines to deal with the retrieval problem, and the first ones were Google and Baidu. As long as the search engine is clear about the direction of the search, it is indeed very convenient for the retrieval of massive data. However, in many cases, most users cannot clearly recognize the content they need or how to accurately express their needs. Faced with this problem, people propose recommender systems to solve the problem of obtaining preference information, which can better increase the user's experience and meet their own needs more easily. Based on the main workflow of the recommender system, this paper designs the overall architecture of the human resources recommendation system and implements a human resources recommendation prototype system based on deep learning. The system can better overcome the cold start problem and provide real-time recommendation results, improving the quality of HR personalized recommendation results. [ABSTRACT FROM AUTHOR]

With the popularity of location-aware devices (e.g., smart phones), large amounts of location-based social media data such as check-ins are generated. This stimulates plenty of studies for POI recommendations by applying machine learning techniques. However, most of the existing studies focus on POI recommendations in the same city or region, and fail to recommend POIs for users when they travel to a new city. In this paper, we propose a novel deep neural network, named as ST-TransRec, for crossing-city POI recommendations. It integrates the deep neural network, transfer learning technique, and density-based resampling method into a unified framework. In this model, the deep neural network is used to capture users’ preferences for POIs and learn the embeddings of POIs. Besides, the transfer learning technique is employed to bridge the gap between cities that results from the city-dependent features. As the distributions over POIs are imbalanced, we design a density-based spatial resampling model which enables POIs to be well matched across cities. We conduct extensive experiments on two real-world datasets. The experimental results show the advantages of ST-TransRec over the state-of-the-art methods for crossing-city POI recommendations. [ABSTRACT FROM AUTHOR]

Rating prediction is a classic problem underlying recommender systems. It is traditionally tackled with matrix factorization. Recently, deep learning based methods, especially graph neural networks, have made impressive progress on this problem. Despite their effectiveness, existing methods focus on modeling the user-item interaction graph. The inherent drawback of such methods is that their performance is bound to the density of the interactions, which is however usually of high sparsity. More importantly, for a strict cold start user/item that neither appears in the training data nor has any interactions in the test stage, such methods are unable to learn the preference embedding of the user/item since there is no link to this user/item in the graph. In this work, we develop a novel framework Attribute Graph Neural Networks (AGNN) by exploiting the attribute graph rather than the commonly used interaction graph. This leads to the capability of learning embeddings for the strict cold start users/items. Our AGNN can produce the preference embedding for a strict cold user/item by learning on the distribution of attributes with an extended variational auto-encoder (eVAE) structure. Moreover, we propose a new graph neural network variant, i.e., gated-GNN, to effectively aggregate various attributes of different modalities in a neighborhood. Empirical results on three real-world datasets demonstrate that our model yields significant improvements for strict cold start recommendations and outperforms or matches the state-of-the-art performance in the warm start scenario. [ABSTRACT FROM AUTHOR]

Easy internet access and technological advancements have resulted in information overload and a plethora of options, making decision-making extremely difficult. Recommender System (RS) is a potential solution for assisting users in making decisions by recommending or predicting product ratings. Three fundamental forms of RS that use implicit or explicit feedback for recommendation are collaborative, content-based, and hybrid filtering. Ratings are the most common form of feedback, but product descriptions, reviews, images, audios, and videos are also important and can help improve the performance of the traditional RS. These additional variables can have a significant impact on RS's performance. Traditional RSs used approaches based on the nearest neighbor or other machine learning models, but thanks to recent advances in artificial intelligence and deep learning, RSs are now being developed using Convolutional Neural Networks (CNN), which can efficiently exploit auxiliary information. In addition to comparing CNN-based RSs on common grounds, this article provides a full examination of CNN-based RSs and how they might use various types of auxiliary information. The study also discusses data characteristics, data statistics, and auxiliary information in a variety of publicly available datasets. Different evaluation measures for RSs are also discussed, and readers are provided with interesting challenges and open research issues. [ABSTRACT FROM AUTHOR]

The tremendous growth the pharmaceutical field witnessed especially concerning drug-development, caused drug-related experiments such as those concerned with estimating their interactions with targets harder to conduct. This issue led researchers to transition from vivo and vitro to selico experimenting methods, which deploy artificial intelligence (AI) techniques such as recommender systems (RS) to solve problems at hand. However, most RSs depend on global features such as indices, which render them incapable of fully understanding the behaviors of individuals. In this paper, we introduce our Convolutional Recurrent Deep Learning model-based RS approach, or 'RSCRDL', which tackles the interaction prediction problem from a natural language processing approach's (NLP) point of view. RSCDRL processes the textual representations of both protein sequences and drugs in order to learn the implicit features of the tokens forming them, rendering it able to process unseen and non-reactive individuals indirectly. We also present a greedier variant of RSCRDL that can exploit additional drugs explicit features for better predictive abilities. Experiments performed on two real-world datasets show significant outperformance achieved by our models over two general-purpose state-of-the-art RSs, while other experiments conducted on orphan protein sequences helped deduce drug candidates that are assumed to be capable of inhibiting their activities. [ABSTRACT FROM AUTHOR]

Vector database management systems have emerged as an important component in modern data management, driven by the growing importance for the need to computationally describe rich data such as texts, images and video in various domains such as recommender systems, similarity search, and chatbots. These data descriptions are captured as numerical vectors that are computationally inexpensive to store and compare. However, the unique characteristics of vectorized data, including high dimensionality and sparsity, demand specialized solutions for efficient storage, retrieval, and processing. This narrative literature review provides an accessible introduction to the fundamental concepts, use-cases, and current challenges associated with vector database management systems, offering an overview for researchers and practitioners seeking to facilitate effective vector data management. [ABSTRACT FROM AUTHOR]

The cross-domain recommendation (CDR) model addresses challenges such as data sparsity, the long tail distribution of user–item interactions, and the cold start of items or users. However, solely transferring domain-shared knowledge based on the co-occurrence patterns, without considering user preferences, leads to negative transfer in CDR. To overcome these limitations, we propose an advanced deep learning CDR model called the Domain Adversarial Deep Interest Network (DADIN) aims to facilitate smooth knowledge transfer from the source domain to the target domain and effectively alleviate negative transfer. Firstly, the joint distribution alignment of user preference in DADIN is realized by introducing a skip-connection-based domain agnostic layer, and then the domain classifier is artificially designed to distinguish between the information coming from the source domain or the target domain. Additionally, DADIN combines prediction loss, global domain confusion loss, and intra-class domain confusion losses through the Min-Max game and gradient reverse layer to achieve collaborative optimization. Two real-world experiments show the area under curve (AUC) of DADIN is 0.78 on the Huawei dataset, and it outperforms its competitors by 0.71% on the Amazon dataset, showcasing its state-of-the-art performance. Moreover, our ablation studies further demonstrate that domain adversarial technique increases the AUC by 2.34% on the Huawei dataset and 16.67% on the Amazon dataset, respectively. • DADIN first employing domain adversarial method in Cross-Domain Recommendation (CDR). • With label information in source domain, DADIN addresses negative transfer in CDR. • The Joint distribution alignment of user preferences is achieved in DADIN. • Hierarchical attention is employed to enhance feature representation in DADIN. • DADIN achieving competitive performance in comparison with popular baseline models. [ABSTRACT FROM AUTHOR]

With the advent of the Internet era, the frequency and proportion of candidates obtaining recruitment information through the Internet is getting higher and higher, and the amount of human resources information such as talent information and postinformation has also increased unprecedentedly, which makes human resources services face the problem of information overload. At the same time, deep learning has achieved great success in a series of fields such as computer vision, natural language processing, and semantic recognition in recent years. However, there are few related works in the field of deep learning applied to human resource management system at present. Therefore, this paper studies and improves the recommendation algorithm based on deep learning and applies it to the field of human resources recommendation. In order to improve the traditional and single algorithm of the existing recommendation system, and improve the performance of the human resource management recommendation system. [ABSTRACT FROM AUTHOR]

Representation learning has attracted growing attention in recommendation system. In addition, deep learning has been adopted to build a representation generator based on content data (e.g., reviews, descriptions), and has been verified to be an excellent method for recommendation system. However, the content data may not be sufficient to capture the hidden features of user behavior patterns. We argue that the underlying information in behavior patterns can characterize users, by generating specific representation from user activities. In this paper, we propose a deep transfer learning-based recommendation model (DeepTransferR), which conducts knowledge transfer from multiple user activities. We adopt attention network to migrate the behavior pattern from auxiliary activities, and to generate personalized representations for users. In DeepTransferR, we set up an independent predictor for each user activity. We then define a weighted loss function to model knowledge interaction by incorporating the independent loss in each activity predictor. Experiments have been conducted on real-world datasets, and the results show that the proposed model outperforms the state-of-the-art methods in not only recommendation performance, but also convergence and robustness in sparse-data and cold-start environments. [ABSTRACT FROM AUTHOR]

Advertising Click-Through Rate (CTR) prediction is a relatively successful application in the field of recommendation system. Improving the accuracy of advertising click-through rate can not only make the user experience better, but also give more benefits to advertising platforms and advertisers. It can be seen from the research status that the interaction between learning features has become a very important part of the advertising CTR prediction model. Although the existing CTR prediction models based on deep learning have achieved good results, some models only consider a single interaction mode and lack the diversity of feature interaction. To resolve this problem, this paper proposes a CTR prediction model based on multi-feature interaction, called EDIF, which aims to enhance the diversity of feature interaction. Firstly, the model learns multiple different embedding vectors for each feature in the embedding layer, which reflects the correlation between features; secondly, in the interaction of high-order features, the embedding vectors of each feature are added and pooled to form the aggregation vector of the whole feature as the input, which reflects the integrity of the feature; finally, after the feature embedding operation, the model introduces two layers in parallel: compressed excitation network layer and explicit high-order interaction layer, which improves the ability of feature interaction. We have done a lot of experiments on two public data sets, Avazu and Criteo. The results show that the model effect of this paper has great advantages over the latest model. [ABSTRACT FROM AUTHOR]

The autoencoder network has been proven to be one of the powerful techniques for recommender systems. Currently, the ways of utilizing autoencoder in recommender systems can be divided into two categories: modeling user-item interaction rely solely on autoencoder and integrating autoencoder with other models. Most existing methods based on autoencoder assume that all features of model's input are equally the same contributing to the final prediction, which can be regarded as attention weight vectors; however, this hypothesis is not reliable, especially when exploring users' interaction frequency with different items. Moreover, combining autoencoder with traditional methods, the usual strategy is to leverage a linear kernel of the inner product of user and item vectors to predict user preferences, which will lead to insufficient expression power and hurt the performance of recommendation when facing data sparsity and cold start problems. To tackle the above two problems, we propose a novel hybrid deep learning model for top-n recommendation, called attentive stacked sparse autoencoder (A-SAERec), which can capture attention weights vector of a user for items, and then combined with the neural matrix factorization to improve the performance of recommender model. Extensive experiments on four real-world datasets show that our A-SAERec algorithm has significant improvements over state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

With the rapid development of technologies such as cloud computing, big data, and the Internet of Things, the scale of data continues to grow. The recommendation system has become one of the important intelligent software to help users make decisions. The recommendation model based on user rating data is widely studied and applied, but the data sparsity problem and the cold start problem seriously affect the recommendation quality. In this paper, Multi-view Hybrid Recommendation Model (MHRM) based on deep learning is proposed. First, we use WLDA (an improved Latent Dirichlet Allocation method) to extract the vector representation of user review text, and then apply LSTM to contextual semantic level user review sentiment analysis. At the same time, the emotion fusion method based on user score embedding is proposed. The problems such as deviations between the user score and actual interest preference, and unbalanced distribution of the score level are solved. This paper has been tested on Amazon product data and compared with various classic recommendation algorithms, using Mean Absolute Error (MAE), hit rate and standardized discount cumulative return for performance evaluation. The experimental results show that the prediction of the MHRM proposed in this paper on the 7 recommendation data and the TopN recommendation index have been significantly improved. [ABSTRACT FROM AUTHOR]

Regular physical activity has a positive impact on our physical and mental health. Adhering to a fixed physical activity regimen is essential for good health and mental wellbeing. Today, fitness trackers and smartphone applications are used to promote physical activity. These applications use step counts recorded by accelerometers to estimate physical activity. In this research, we performed a two-level clustering on a dataset based on individuals' physical and physiological features, as well as past daily activity patterns. The proposed model exploits the user data with partial or complete features. To include the user with partial features, we trained the proposed model with the data of users who possess exclusive features. Additionally, we classified the users into several clusters to produce more accurate results for the users. This enables the proposed system to provide data-driven and personalized activity planning recommendations every day. A personalized physical activity plan is generated on the basis of hourly patterns for users according to their adherence and past recommended activity plans. Customization of activity plans can be achieved according to the user's historical activity habits and current activity objective, as well as the likelihood of sticking to the plan. The proposed physical activity recommendation system was evaluated in real time, and the results demonstrated the improved performance over existing baselines. [ABSTRACT FROM AUTHOR]

In the era of big data, recommender systems (RSs) have become growing essential tools. They represent important machine learning solutions that mainly contribute to keeping users engaged with personalized content in e-platforms. Several RSs have been proposed in the literature, and most of them have focused on English content. However, for content in other languages like Arabic, very restricted works have been done to develop RSs. In recent times, the Arabic content on the Web has increased significantly because of the growing number of Arabic web users. This highlights the need for building RSs for Arabic content. To better handle this challenge, we decided to provide the research community with a novel deep learning (DL)-based RS devoted to Arabic content. The main goal of the proposed RS is to predict user preferences from textual reviews written in the Arabic language. This is achieved by combining two independent DL techniques into one system: a convolutional neural network (CNN)-text processor for representing users and items; and a neural network, in particular, a multi-layer perceptron (MLP) to estimate interactions between user-item pairs. Extensive experiments on four large-scale Arabic datasets demonstrate that our proposed system can achieve better prediction accuracy than other state-of-the-art alternatives. Notably, it improves the MSE between 0.84% and 16.96%, and the MAE between 0.14% and 13.71%. This work is the first attempt designed to deal with a large volume of data in the Arabic context, opening up new research possibilities for future developments of Arabic RSs. [ABSTRACT FROM AUTHOR]

Recommendation efficiency and data sparsity problems have been regarded as two challenges of improving performance for online recommendation. Most of the previous related work focus on improving recommendation accuracy instead of efficiency. In this paper, we propose a Deep Pairwise Hashing (DPH) to map users and items to binary vectors in Hamming space, where a user's preference for an item can be efficiently calculated by Hamming distance, which significantly improves the efficiency of online recommendation. To alleviate data sparsity and cold-start problems, the user-item interactive information and item content information are unified to learn effective representations of items and users. Specifically, we first pre-train robust item representation from item content data by a Denoising Auto-encoder instead of other deterministic deep learning frameworks; then we finetune the entire framework by adding a pairwise loss objective with discrete constraints; moreover, DPH aims to minimize a pairwise ranking loss that is consistent with the ultimate goal of recommendation. Finally, we adopt the alternating optimization method to optimize the proposed model with discrete constraints. Extensive experiments on three different datasets show that DPH can significantly advance the state-of-the-art frameworks regarding data sparsity and item cold-start recommendation. [ABSTRACT FROM AUTHOR]

The massive number of research articles on the Web makes it troublesome for researchers to identify related works that could meet their preferences and interests. Consequently, various network representation learning-based models have been proposed to produce citation recommendations. Nevertheless, these models do not exploit semantic relations and contextual information between the objects of bibliographic papers' networks, which can result in inadequate citation recommendations. Moreover, existing citation recommendation methods face problems such as lack of personalization, cold-start, and network sparsity. To mitigate such problems and produce individualized citation recommendations, we propose a heterogeneous network embedding model that jointly learns node representations by exploiting semantics corresponding to the author, time, context, field of study, citations, and topics. Compared to baseline models, the results produced by the proposed model over the DBLP datasets prove 10% and 12% improvement on mean average precision (MAP) and normalized discounted cumulative gain (nDCG@10) metrics, respectively. Also, the effectiveness of our model is analyzed on the cold-start papers and network sparsity problems, where it gains 12% and 9% better MAP and recall@10 scores, respectively. [ABSTRACT FROM AUTHOR]

Descriptive method names have a great impact on improving program readability and facilitating software maintenance. Recently, due to high similarity between the task of method naming and text summarization, large amount of research based on natural language processing has been conducted to generate method names. However, method names are much shorter compared to long source code sequences. The salient information of the whole code snippet account for an relatively small part. Additionally, unlike natural language, source code has complicated structure information. Thus, modelling the salient information from highly structured input presents a great challenge. To tackle this problem, we propose a graph neural network (GNN)‐based model with a novel salient information selection layer. Specifically, to comprehensively encode the tokens of the source code, we employ a GNN‐based encoder, which can be directly applied to the code graph to ensure that the syntactic information of code structure and semantic information of code sequence can be modelled sufficiently. To effectively discriminate the salient information, we introduce an information selection layer which contains two parts: a global filter gate used to filter irrelevant information, and a semantic‐aware convolutional layer used to focus on the semantic information contained in code sequence. To improve the precision of the copy mechanism when decoding, we introduce a salient feature enhanced attention mechanism to facilitate the accuracy of copying tokens from input. Experimental results on an open source dataset indicate that our proposed model, equipped with the salient information selection layer, can effectively improve method naming performance compared to other state‐of‐the‐art models. [ABSTRACT FROM AUTHOR]

Several applications of deep learning, such as image classification and retrieval, recommendation systems, and especially image synthesis, are of great interest to the fashion industry. Recently, image generation of clothes gained lot of popularity as it is a very challenging task that is far from being solved. Additionally, it would open lots of possibilities for designers and stylists enhancing their creativity. For this reason, in this paper we propose to tackle the problem of style transfer between two different people wearing different clothes. We draw inspiration from the recent StarGANv2 architecture that reached impressive results in transferring a target domain to a source image and we adapted it to work with fashion images and to transfer clothes styles. In more detail, we modified the architecture to work without the need of a clear separation between multiple domains, added a perceptual loss between the target and the source clothes, and edited the style encoder to better represent the style information of target clothes. We performed both qualitative and quantitative experiments with the recent DeepFashion2 dataset and proved the efficacy and novelty of our method. [ABSTRACT FROM AUTHOR]

Accurate recommendation of tourist attractions is conducive to improving users' travel efficiency and tourism experience. However, the choice of tourism feature factors and the difference of recommendation algorithm will affect the accuracy of scenic spot recommendation. Aiming at the problems of sparse data, insufficient tourism factors, and low recommendation accuracy in the existing tourism recommendation research, this paper puts forward a scenic spot recommendation method based on microblog data and machine learning by using the characteristics of personalized expression and strong current situation of microblog data and the intelligent prediction function of machine learning, so as to realize accurate and personalized scenic spot recommendation. This paper extracts rich tourism characteristic factors. Typical tourism recommendation algorithms choose tourism characteristic factors from scenic spots, tourists, and other aspects, without considering the travel time, tourism season, and other contextual information of tourists' destination, which can help understand users' tourism preferences from different angles. Aiming at the problem of sparse data and cold start of collaborative filtering recommendation algorithm, this paper introduces deep learning algorithm and combines the proposed multifeature tourism factors to build dynamic scenic spot prediction models (random forest preferred attraction prediction (RFPAP) and neural networks preferred attraction prediction (NNPAP)). The experimental results show that RFPAP and NNPAP methods can overcome the problem of data sparsity and achieve 89.61% and 89.51% accuracy, respectively. RFPAP method is better than NNPAP method and has stronger generalization ability. [ABSTRACT FROM AUTHOR]

It is one of the main goals of personalized music recommendation system that how to accurately recommend the songs in line with users' interests in the huge music library. In view of the above problems, this study proposes a personalized music recommendation method based on convolutional neural network. First, this study defines a training set containing potential musical characteristics and, combined with the depth of the belief network, design a music information prediction model and the research in the music-type classification method with different dimensions. Based on selecting four different kinds of music information better describing the underlying characteristics of 40D feature vector to every song music composition, the music feature set is constructed. Then, the CNN (convolutional neural network), which is widely used in audio field, is used as the music information prediction model, and its structural parameters are redesigned to complete the multidimensional music information prediction, which solves the cold start problem to some extent. [ABSTRACT FROM AUTHOR]

Recommender systems are popular tools used in many applications, such as e-commerce, e-learning, and social networks to help users select their desired items. Collaborative filtering is a widely used recommendation technique that employs previous ratings of users to predict their future interests. Lack of sufficient ratings often reduces the performance of collaborative filtering recommendation methods. Additional side resources, such as trust relationships and tag information can be employed to enhance the recommendation accuracy. However, trust and tag data are often heavily sparse as users mainly provide insufficient information about these side resources. Moreover, such additional resources have generally large dimensions which result in increasing the computational complexity of recommendation models in calculating similarity values. To cope with these problems, a new recommendation model is proposed that utilizes deep neural networks to model the representation of trust relationships and tag information. To this end, a sparse autoencoder is used to extract latent features from user-user trust relationships and user-tag matrices. Then, the extracted latent features are utilized to calculate similarity values between users, which are then used to form the nearest neighbors of the target user and predict unseen items. The proposed method can tackle the data sparsity problem and reduce the computational complexity of recommender systems as the extracted latent features have smaller dimensions in comparison to the original data. Experimental results on two benchmark datasets reveal the effectiveness of the proposed method and show its outperformance over state-of-the-art recommender systems. [ABSTRACT FROM AUTHOR]

Although most list-ranking frameworks are based on multilayer perceptrons (MLP), they still face limitations within the method itself in the field of recommender systems in two respects: (1) MLP suffer from overfitting when dealing with sparse vectors. At the same time, the model itself tends to learn in-depth features of user–item interaction behavior but ignores some low-rank and shallow information present in the matrix. (2) Existing ranking methods cannot effectively deal with the problem of ranking between items with the same rating value and the problem of inconsistent independence in reality. We propose a list ranking framework based on linear and non-linear fusion for recommendation from implicit feedback, named RBLF. First, the model uses dense vectors to represent users and items through one-hot encoding and embedding. Second, to jointly learn shallow and deep user–item interaction, we use the interaction grabbing layer to capture the user–item interaction behavior through dense vectors of users and items. Finally, RBLF uses the Bayesian collaborative ranking to better fit the characteristics of implicit feedback. Eventually, the experiments show that the performance of RBLF obtains a significant improvement. [ABSTRACT FROM AUTHOR]

Feature combinations are essential for the success of many web applications, such as personalised recommendation and online advertising. State-of-the-art methods usually model explicit feature interactions to help neural networks reduce the number of parameters and achieve better performance. However, their explicit feature interactions are often restricted to the second-order due to computational complexity. In this work, we propose efficient ways to represent explicit high-order feature combinations as well as prune redundant features in the mean time. To begin with, we make novel use of the Count Sketch algorithm within a DNN classifier such that high-order feature combinations can be compactly represented. After that, to combat the problem of redundant features which degrade the prediction performance, we introduce an adaptive hashing algorithm, AutoHash, which can automatically select meaningful features to interact at high orders according to the specific dataset in question. This is an AutoML approach. Experiments on three well-known public datasets demonstrate that AutoHash is significantly superior to state-of-the-art methods. Meanwhile, due to its efficient scheme of automatically selecting useful high-order feature interactions, AutoHash has less model complexity and can be trained in an end-to-end manner with less training time than state-of-the-art methods. [ABSTRACT FROM AUTHOR]