Search

The recommendation system has far-reaching significance and great practical value, which alleviates people's troubles about choosing from a huge amount of information. The existing recommendation system usually faces the selection bias problem due to the ignorance of samples with delayed feedback. To alleviate this problem, by modeling the recommendation as a batch contextual bandit problem, we propose a counterfactual reward estimation approach in this work. First, we formalize the counterfactual problem as "would the user be interested in the recommended item if the delayed time is before the collection time point?". The above counterfactual reward is estimated in a survival analysis framework, by fully exploring the causal generation process of user feedback on batch data. Second, based on the above estimated counterfactual rewards, the policy of batch contextual bandit is updated for online recommendation in the next episode. Third, new batch data are generated in the online recommendation for further counterfactual reward estimation. The above three steps are iteratively conducted until the optimal policy is learned. We also prove the sub-linear regret bound of the learned bandit policy theoretically. Our method achieved a 4 % improvement in average reward compared to the baseline methods in experiments conducted on synthetic and Criteo datasets, demonstrating the efficacy of our approach. [ABSTRACT FROM AUTHOR]

Due to the superior performance of graph convolutional networks (GCNs) in feature extraction and representation, researchers have introduced GCNs to collaborative filtering (CF) to improve the accuracy of recommendations. However, existing GCN-based CF models still have three shortcomings: degrading the model performance due to symmetric normalization in neighborhood aggregation, underutilizing the node (user or item) representations at lower layers, and failing to alleviate the over-smoothing effect in graph convolution operations. We present a Residual Graph Convolution Collaborative Filtering model with Asymmetric neighborhood aggregation (ARGCCF) to tackle the above shortcomings. Firstly, ARGCCF utilizes asymmetric normalization in neighborhood aggregation, enabling better preservation of high-degree node information for more precisely learning node representations. Secondly, the propagation of ARGCCF embeddings proceeds in two steps. During the initial propagation, a residual connection is established between the node representation of the current layer and the previous layer to utilize the low-level layer's node representation fully. The embeddings generated after the initial propagation are propagated again in the same graph convolutional layer, which obtains more refined embedding representations to alleviate the over-smoothing problem. Finally, experimentations on five truthful datasets exhibit that ARGCCF has better-improved performance over several mainstream methods. [ABSTRACT FROM AUTHOR]

Capturing dynamic preference features from user historical behavioral data is widely applied to improve the accuracy of recommendations in sequential recommendation tasks. However, existing deep neural network-based sequential recommendation methods often ignore the noise information in user behavioral data that influence recommendation effectiveness, making recommendation models sensitive to noisy data. Additionally, these deep learning-based recommendation models often require a large number of parameters to capture user behavior patterns, leading to higher time complexity and susceptibility to overfitting due to noise fluctuations. To address these issues, in this paper, we apply the moving average concept from the field of time-series analysis to sequential recommendation tasks. The approach smooths sequence data, removes noise, making the data more stable and reliable, increases the model's insensitivity to noise information, and better understands the evolution of user interests and behavioral patterns. Specifically, in our experiments, the moving average algorithm effectively denoises sequential data with low time complexity. Therefore, we propose a Sequential Recommendation with Exponential Moving Average with MLP architecture that makes the model more competitive in terms of time complexity. Experimental results conducted on two datasets demonstrate that EMARec outperforms state-of-the-art sequential recommendation methods across various common evaluation metrics. [ABSTRACT FROM AUTHOR]

The online content, including live streaming and short videos, provides abundant visual and textual product information to users, which offers insights into users' multiple and changeable preferences toward product style, brand, color, etc. These dynamic preferences are helpful to the product recommender system to extract users' purchasing patterns. However, existing sequential recommendation fails to combine with items' multimodal characteristics and dynamic preferences. This paper proposes the Online Content-Based Sequence Recommendation method (named OCBSR), where dynamic preferences and multimodal representation such as visual images, textual descriptions, and item-ID of products are leveraged with convolutional autoencoder to map multimodal data into a unified latent space. We devise a multimodal transformer mechanism to aggregate the multimodal representation. A novel contrastive infomax loss is constructed through contrastive prototype representations and noise-contrastive sequences. The loss enhances the correlation between single-modality and multimodality preferences and learns the contrastive representation. Then, users' dynamic preferences are extracted from the contrastive representation in each time step using the gated recurrent unit with skip connections. Experiments are conducted on two real-life multimodal datasets, H&M and KuaiRec. The hit ratio (HR) and normalized discounted cumulative gain (NDCG) results indicate that our approach noticeably outperforms the current advanced recommenders. The HR increases by above 20% and 25% on H&M and KuaiRec datasets, respectively; the NDCG rises about 12% and 24% on H&M and KuaiRec datasets, respectively. The ablation study verifies each module's effectiveness. This research contributes to online content literature and multimodal sequential recommendation in practice. [ABSTRACT FROM AUTHOR]

Knowledge graph (KG) shows great potential in improving recommendation systems. Recent studies have focused on developing end-to-end models based on graph neural networks. However, these previous approaches do not fully utilize collaborative signals, which are the collective information and feedback provided by multiple users in a recommendation system. Specifically, we think that collaborative signals could be better utilized in two points. Firstly, users' collective information leads to differences in node importance in KG, offering opportunities to improve item representation. Secondly, an item's collaborative filtering (CF) feature can reveal the multiple interests of a user, helping to gain fine-grained user representation. In this paper, we highlight the effect of node importance estimation and CF-based item clustering and propose the collaborative signals injection module. We integrate this module in our Collaborative Signals Enhanced Knowledge Graph recommendation (CSEKG) model. CSEKG outperforms state-of-the-art methods like KGIN (Wang et al. in Learning intents behind interactions with knowledge graph for recommendation. CoRR arXiv:2102.07057, 2021) , KGCL (Yang et al. in Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval, 2022) , and KGRec (Yang et al. in Proceedings of the 29th ACM SIGKDD conference on knowledge discovery and data mining, 2023) through improving recall@20 by 3.86 % , 5.15 % , and 2.88 % on Last.FM, Amazon-Book and MovieLens-1M, respectively. Furthermore, we flexibly integrate the collaborative signals injection module into existing methods, such as LightGCN, KGAT and KGIN, which improves their performance. Further visualizing recommendation cases show the interpretability benefits brought by collaborative signals. [ABSTRACT FROM AUTHOR]

Crop Recommendation Systems are invaluable tools for farmers, assisting them in making informed decisions about crop selection to optimize yields. These systems leverage a wealth of data, including soil characteristics, historical crop performance, and prevailing weather patterns, to provide personalized recommendations. In response to the growing demand for transparency and interpretability in agricultural decision-making, this study introduces XAI-CROP an innovative algorithm that harnesses eXplainable artificial intelligence (XAI) principles. The fundamental objective of XAI-CROP is to empower farmers with comprehensible insights into the recommendation process, surpassing the opaque nature of conventional machine learning models. The study rigorously compares XAI-CROP with prominent machine learning models, including Gradient Boosting (GB), Decision Tree (DT), Random Forest (RF), Gaussian Naïve Bayes (GNB), and Multimodal Naïve Bayes (MNB). Performance evaluation employs three essential metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R2). The empirical results unequivocally establish the superior performance of XAI-CROP. It achieves an impressively low MSE of 0.9412, indicating highly accurate crop yield predictions. Moreover, with an MAE of 0.9874, XAI-CROP consistently maintains errors below the critical threshold of 1, reinforcing its reliability. The robust R2 value of 0.94152 underscores XAI-CROP's ability to explain 94.15% of the data's variability, highlighting its interpretability and explanatory power. [ABSTRACT FROM AUTHOR]

This paper proposes a structure preserving generative adversarial network (S-P GAN) to solve the problem of small structure information loss during the modality translation from computed tomography (CT) to magnetic resonance (MR). A novel generator of the S-P GAN is designed to encode features from CT, where an original CT information branch and its corresponding high-frequency information branch form a dual-branch structure. The small details are highlighted by using a filter in the high-frequency information branch, which offers a complement to the integrity of structural information in CT. Meanwhile, a mixed attention mechanism is introduced to better fuse the dual-branch features for decoding features to MR, where small structure features get more attention in channel and space. Additionally, a new joint loss function is proposed to guide the adversarial training of S-P GAN, which contains structural consistency constrain, pixel translation constrain, and adversarial constrain, so that global similarity and local detail consistency are obtained at the same time. Experimental results show that the results of the S-P GAN are superior to the state-of-the-art models in mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM). In real clinical situations, the proposed method has shown good performance in the diagnosis of lumbar disk herniation, the new and old degree of compressibility fracture, and the Modic change of cartilage end plate. [ABSTRACT FROM AUTHOR]

Graph convolution networks (GCNs) play an increasingly vital role in recommender systems, due to their remarkable relation modeling and representation capabilities. Concretely, they can capture high-order semantic correlations within sparse bipartite interaction graphs, thereby enhancing user–item collaborative encodings. Despite the exciting prospects, the existing GCN-based models mainly focus on user–item interactions and seldom consider effectiveness of the side item co-occurrence information on user behavior guidance, resulting in limited performance improvement. Therefore, we propose a novel side item co-occurrence information-aware GCN model. Specifically, we first decouple the original heterogeneous relation graph into corresponding user–item and item–item subgraphs for user–item interaction and item co-occurrence relation modeling. Thereafter, we conduct adaptive iterative aggregation on these subgraphs for user intention understanding and co-occurring item correlation perception. Finally, we present two semantic fusion strategies for sufficient user–item semantic collaborative learning, thereby boosting the overall recommendation performance. Extensive comparison experiments are conducted on three benchmark datasets to justify the superiority of our model. [ABSTRACT FROM AUTHOR]

Nowadays, the volume of online information is growing and it is difficult to find the required information. Effective strategies such as recommender systems are required to overcome information overload. Collaborative filtering is a widely used type of recommender system in e-commerce environments and can simply provide suggestions for users. Recently, deep learning approaches were applied in collaborative filtering to tackle some drawbacks. This systematic review aims to provide a comprehensive review of recent research efforts on deep learning-based collaborative filtering recommender systems. We explain the research methodology and paper selection process and the search query. 102 papers are selected out of 719 papers that were published between 2019 and May 2023. Furthermore, the approaches in the selected papers are classified into two main categories: memory-based and model-based techniques. The main ideas, advantages, disadvantages, used tools, type of neural network, applications, and evaluation parameters of each selected paper are also discussed in detail. It was found that CNN (Convolutional Neural Network), AE (Autoencoder), DNN (Deep neural network), and Hybrid networks are the four mostly used neural networks in recommender systems. Also, Python, MATLAB, and Java are the most frequently used tools in the reviewed papers. Regarding the applications of the recommender systems in the reviewed papers, movies, products, and music recommendation are three most frequent applications. We point out the open issues and future research directions. Some key challenges such as cold start, data sparsity, scalability, and accuracy are still open to be addressed. [ABSTRACT FROM AUTHOR]

A recommendation system (RS) can help overcome information overload issues by offering personalized predictions for users. Typically, RS considers the overall ratings of users on items to generate recommendations for them. However, users may consider several aspects when evaluating items. Hence, a multi-criteria RS considers n-aspects of items to generate more accurate recommendations than a single-criteria RS. This research paper proposes two deep encoder–decoder models based on shared learning for a multi-criteria RS, multi-modal deep encoder–decoder-based shared learning (MMEDSL) and multi-criteria deep encoder–decoder-based shared learning (MCEDSL). MMEDSL employs the shared learning technique by concentrating on the multi-modality concept in deep learning, while MCEDSL focuses on the training process to apply the shared learning technique. The shared learning captures useful shared information during the learning process since the multi-criteria may have hidden inter-relationships. A set of experiments were conducted to compare the proposed models with recent baseline approaches. The Yahoo! Movies multi-criteria dataset was utilized. The results demonstrate that the proposed models outperform other algorithms. In addition, the results show that integrating the shared learning technique with the RS produces precise recommendation predictions. [ABSTRACT FROM AUTHOR]

Recommendation systems for TV programs play an important role in alleviating the information overload problem. Existing TV program recommendation methods either do not aggregate neighborhood information well to capture collaborative signals from interaction data, or fail to make good use of auxiliary information, because they ignore the heterogeneity of different entities and relationships. In this paper, we propose a multi-component graph collaborative filtering recommendation based on auxiliary information, which learns representations of user and program through heterogeneous data modeling and information propagation on graphs. We extract homogeneous subgraphs from the heterogeneous graph based on multiple symmetric meta-paths, learn the components of the node representation by performing graph convolution on the homogeneous subgraphs, and finally combine the components to obtain the complete user representation and program representation. Experiments on real-world datasets show that our approach can effectively improve the performance of TV program recommendations compared to the existing baselines. [ABSTRACT FROM AUTHOR]

Removing rain streaks from rainy images can improve the accuracy of computer vision applications such as object detection. In order to make full use of the frequency domain analysis characteristics of wavelet and combine the advantages of Convolutional Neural Network (CNN) and Transformer, a Multi-level Wavelet Network Based on CNN-Transformer Hybrid Attention (MWN-CTHA) for single image deraining is proposed. MWN-CTHA obtains multi-scale low-frequency and high-frequency images through multi-level non-separable lifting wavelet transform and uses CNN-Transformer Hybrid Attention Block (CTHAB) to learn global structure and detail information from low-frequency and high-frequency, respectively. CTHAB consists of CA-SA Layer (CSL) and Detail-enhanced Attention Feed-forward Layer (DAFL). CSL uses the non-local modeling ability of self-attention to capture long-range rain streaks and uses convolutional attention to enhance the search ability for local rain streaks, where convolution can assist self-attention to achieve better feature representation. DAFL utilizes Depth-wise Convolutional Layer to supplement detailed features and filters the information of feed-forward layer through Dual-branch Attention. The experimental results on the four synthetic datasets demonstrate that the proposed method achieves higher PSNR and SSIM than the state-of-the-art method DANet, with an improvement of 1.07 dB and 0.0098, respectively. The code is available at https://github.com/fashyon/MWN-CTHA. [ABSTRACT FROM AUTHOR]

Navigation through large volumes of images is a complex and tedious task that requires tools to facilitate the exploration and discovery of visual information. Photo summaries are one of these tools, which consist of selecting a reduced set of images that best represent the original data source. However, creating photo summaries in the context of recommender systems poses several challenges: How to select the most relevant images for each item? How to encode each image? How to evaluate the quality of the generated summary? In this manuscript, we propose a clustering-based method to create a visual summary in the context of a restaurant recommender system, which includes the photos taken by users who visited the restaurants (items) in a given city. These photos are encoded using a deep neural network that takes into account not only their content but also the relationships between users and restaurants. This encoding will allow us to create a visual summary that captures the essence of user tastes and illustrates the gastronomic offer of the city. We also propose a similarity measure between items based on the users who have visited them and an evaluation method that calculates to what extent the summary obtained represents the original data source. The experimentation carried out includes five datasets and the obtained results demonstrate the adequacy of our proposal for the construction of these summaries. [ABSTRACT FROM AUTHOR]

Exploiting user trust information for developing a recommendation system has gained increasing research interest in recent years. Due to the exchange of opinions about items over the social network, trust plays a crucial role for a user to like or dislike an item. Graph Neural Networks (GNNs), which have the intrinsic power of integrating node information and topological structure, have a high potential to advance the field of trust-aware social recommendation. However, as of now, this area is little explored, with most of the existing GNN-based models ignoring the trust propagation and trust composition properties. To address this issue, in this paper, we propose a novel GNN-based framework that can capture such trust propagation and trust composition aspects by incorporating the concept of 'user-reliability.' Our proposed user-reliability-aware social recommendation framework, termed as SoURA, generates the user-embedding and item-embedding with consideration to the user-reliability values, which, in turn, helps in better evaluation of the user trust. Experimental evaluations on the benchmark Ciao and Epinion datasets demonstrate the effectiveness of incorporating user-reliability for finding user-embedding and item embedding in a social recommendation system. The proposed SoURA is found to show a minimum of 25% improvement over the state-of-the-art GNN-based recommendation algorithms. [ABSTRACT FROM AUTHOR]

Recommender systems are widely adopted as an increasing research and development area, since they provide users with diverse and useful information tailored to their needs. Several strategies have been proposed, and in most of them some concept of similarity is used as a core part of the approach, either between items or between users. At the same time, Siamese Neural Networks are being used to capture the similarity of items in the image domain, as they are defined as a subtype of Artificial Neural Networks built with (at least two) identical networks that share their weights. In this review, we study the proposals done in the intersection of these two fields, that is, how Siamese Networks are being used for recommendation. We propose a classification that considers different recommendation problems and algorithmic approaches. Some research directions are pointed out to encourage future research. To the best of our knowledge, this paper is the first comprehensive survey that focuses on the usage of Siamese Neural Networks for Recommender Systems. [ABSTRACT FROM AUTHOR]

Recommendation to groups of users is a challenging subfield of recommendation systems. Its key concept is how and where to make the aggregation of each set of user information into an individual entity, such as a ranked recommendation list, a virtual user, or a multi-hot input vector encoding. This paper proposes an innovative strategy where aggregation is made in the multi-hot vector that feeds the neural network model. The aggregation provides a probabilistic semantic, and the resulting input vectors feed a model that is able to conveniently generalize the group recommendation from the individual predictions. Furthermore, using the proposed architecture, group recommendations can be obtained by simply feedforwarding the pre-trained model with individual ratings; that is, without the need to obtain datasets containing group of user information, and without the need of running two separate trainings (individual and group). This approach also avoids maintaining two different models to support both individual and group learning. Experiments have tested the proposed architecture using three representative collaborative filtering datasets and a series of baselines; results show suitable accuracy improvements compared to the state of the art. [ABSTRACT FROM AUTHOR]

Recommender systems face longstanding challenges in gaining users' trust due to the unreliable information caused by profile injection or human misbehavior. Traditional solutions to those challenges focus on leveraging users' social relationships for inferring the user preference, i.e., recommending items according to the preference by user's trusted friends; or adding random noise to the input to improve the robustness of the recommender systems. However, such approaches cannot defend the real-world noises like fake ratings. The recommender model is generally built upon all the user-item interactions, which incorporates the information from fake ratings or spammer groups, that neglects the reliability of the ratings. To address the above challenges, we propose an adversarial training approach in this work. In details, our approach includes two components: a predictor that infers the user preference; and a discriminator that enforces cohort rating patterns. In particular, the predictor applies an encoder-decoder structure to learn the shared latent information from sparse users' ratings and trust relationships; the discriminator enforces the predictor to provide ratings as coherent with the cohort rating patterns. Our extensive experiments on three real-world datasets show the advantages of our approach over several competitive baselines. [ABSTRACT FROM AUTHOR]

Recommendation system, or recommender system, is widely used in online Web applications like e-commerce Web sites and movie review Web sites. Sequential recommender put more emphasis upon user's short-term preference through exploiting information from its recent history. By incorporating the user short-term preference into the recommendation, the algorithm achieves a higher accuracy, which proves that a more accurate user portrait or representation boosts the performance to a great extent. Intuitionally, we seek to improve the current item representation modeling via incorporating the item trend information. Most of the recommendation models neglect the importance of the ever-changing item popularity. To this end, this paper introduces a novel sequential recommendation approach dubbed TRec. TRec learns the item trend information from the implicit user interaction history and incorporates the item trend information into the subsequent item recommendation tasks. After that, a self-attention mechanism is used for better representation. We also investigate alternative ways to model the proposed item trend representation; we evaluate two variant models which leverage the power of gated graph neural network upon the item trend representation modeling to boost the representation ability. We conduct extensive experiments with seven baseline methods on four benchmark datasets. The empirical results show that our proposed approach outperforms the state-of-the-art models as high as 18.2%. The experiment result displays the effectiveness in item trend information learning while with low computational complexity as well. Our study demonstrates the importance of item trend information in recommendation system. [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]

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]

Recommender systems are considered to be capable of predicting what the next product a customer should purchase is. It is crucial to identify which customers are more suitable than others to target a product for cross-selling in the retail industry. Using recurrent neural networks with self-attention mechanisms, this study proposes a hybrid model. Furthermore, the proposed design is capable of handling both sequential and non-sequential features, which correspond to purchase behavior and non-behavioral customer specific information, respectively. This study represents an alternative solution to a well-known business problem: improving cross-selling effectiveness by estimating customers' likelihood for which products or services to buy next time. A recommender system which works on additional data configurations is the core concept of the framework. With an online shopping data set, this study shows that concatenation of relevant features adds additional information to the model, and it is found that evaluation metrics are improved by approximately 12%. [ABSTRACT FROM AUTHOR]

As the information industry and the Internet develop rapidly, the use of big data enters people's vision and attracts attention. It makes the recommendation system come into being how to quickly extract the desired information from the excessive information. In the recommendation system, user-based collaborative filtering algorithm has become a research hotspot. Existing researches focus on improving collaborative filtering recommendation algorithm by using the kernel method, but still face the cold start problem, the diversity problem, the data sparsity problem, the concept drift problem and more others. To solve these problems, this paper proposes the user-based collaborative filtering based on kernel method and multi-objective optimization (MO-KUCF) which introduces kernel density estimation and multi-objective optimization. It can be increasing diversity of the recommendation systems, improving concept drift in dynamic data and the accuracy and diversity of the recommendation system. The dataset used in this article is the Netflix dataset. It analyzes the MO-KUCF algorithm with the user-based collaborative filtering (UCF) and user-based collaborative filtering based on kernel method (KUCF) by the mean absolute error (MAE). The MAE is compared with the internal user diversity I u index, and the pre-processed data set is divided into the training set and the test set, which are provided to the recommendation system for recommendation and evaluation. The results show that the accuracy of MO-KUCF improves by 5.6%, and the diversity also increases with decreasing values. Combining multi-objective optimization techniques with kernel density estimation methods can improve the diversity of recommendation systems effectively and solve the concept drift problem to achieve the purpose of improving system accuracy. [ABSTRACT FROM AUTHOR]

Previous works mainly employ triple structural information in learning representations for knowledge graph, which results in poor performance of link prediction especially when it predicts new or few-fact entities. It is intuitive to introduce text information to supplement the missing semantic information for knowledge representation. However, existing methods only make alignment on the level of word or score function, and have not yet considered textual and structural information. Moreover, since the textual information is potentially redundant to represent the entities, how to extract relevant information and simultaneously alleviate the irrelevant information contained in the text is a challenging task. To tackle the above problems, this paper proposes a novel knowledge representation learning framework of joint semantic embedding using structural knowledge and entity description (JointSE). Firstly, we design a mutual attention mechanism to filter the effective information of fact triples and entity descriptions with respect to specific relationships. Secondly, we project the triples into the text semantic space using dot product to connect the triple with the relevant text description. In addition, we enhance triple-based entity representation and text-based entity representation using graph neural network to capture more useful graph structure information. Finally, extensive experiments on benchmark datasets and Chinese legal provisions dataset demonstrate that JointSE realizes the effective fusion of triple information, text semantic information, and graph structure information. We observe that JointSE is superior to previous methods in entity prediction and relationship prediction tasks. [ABSTRACT FROM AUTHOR]

In recent years, graph convolution networks and hypergraph convolution networks have become a research hotspot in collaborative filtering (CF) because of their information extraction ability in dealing with the user-item interaction information. In particular, hypergraph can model high-order correlation of users and items to achieve better performance. However, the existing graph-based CF methods for mining interactive information remain incomplete and limit the expressiveness of the model. Moreover, they directly use low-order Chebyshev polynomials to fit the convolution kernel of graph and hypergraph without experimental proof or analysis, lacking interpretability. We propose an effective hybrid graph and hypergraph convolutional network (EHGCN) for CF to obtain a capable and interpretable framework. In EHGCN, the graph and the hypergraph are used to model the correlation among nodes in the interaction graph for multilevel learning. EHGCN also optimizes the information flow framework to match the improved convolution strategy of the graph and hypergraph we proposed. Extensive experiments on four real-world datasets show the considerable improvements of EHGCN over other state-of-the-art methods. Moreover, we analyze the graph and hypergraph convolution kernel in terms of the spectral domain to reveal the core of the graph-based CF, which has a heuristic effect on future work. [ABSTRACT FROM AUTHOR]

In this paper, we propose a knowledge-aware attentional neural network (KANN) for dealing with movie recommendation tasks by extracting knowledge entities from movie reviews and capturing understandable interactions between users and movies at the knowledge level. In most recommendation systems, review information is already widely utilized to uncover the explicit preferences of users for items, especially for domains including movie recommendations, music recommendations, and book recommendations, as reviews are full of knowledge entities relevant to the domain. When processing review information, current methods usually use word embeddings to represent reviews for modeling users and items. As a result, they may split the meaning of a phrase, and thereby induce erroneous predictions. Moreover, most methods capture high-order interactions between users and items after obtaining latent low-dimensional representations, which means they cannot discover understandable interactions or provide knowledge-level explanations. By incorporating knowledge graph representation into movie recommendation tasks, the proposed KANN can not only capture the inner attention among user (movie) reviews but also compute the outer attention values between users and movies before generating corresponding latent vector representations. These characteristics enable the explicit preferences of users for movies to be learned and understood. We test our model on two datasets (IMDb and Amazon) for the movie rating prediction task and the click-through rate prediction task and show that it outperforms some of the existing state-of-the-art models and gains outstanding prediction performances in cases with a very small amount of reviews. Furthermore, we demonstrate the high explainability of the proposed KANN by visualizing the interaction between users and movies through a case study. Our results and analyses highlight the relatively high effectiveness and reliability of KANN for movie recommendation tasks. [ABSTRACT FROM AUTHOR]

There is a growing interest in the offering of novel alternative choices to users of recommender systems. These recommendations should match the target query while at the same time they should be diverse with each other in order to provide useful alternatives to the user, i.e., novel recommendations. In this paper, the problem of extracting novel recommendations, under the similarity–diversity trade-off, is modeled as a facility location problem. The results from tests in the benchmark Travel Case Base were satisfactory when compared to well-known recommender techniques, in terms of both similarity and diversity. It is shown that the proposed method is flexible enough, since a parameter of the adopted facility location model constitutes a regulator for the trade-off between similarity and diversity. Also, our work can broaden the perspectives of the interaction and combination of different scientific fields in order to achieve the best possible results. [ABSTRACT FROM AUTHOR]

Collaborative filtering algorithms take into account users' tastes and interests, expressed as ratings, in order to formulate personalized recommendations. These algorithms initially identify each user's "near neighbors," i.e., users having highly similar tastes and likings. Then, their already entered ratings are used, in order to formulate rating predictions, and predictions are typically used thereafter to drive the recommendation formulation process, e.g., by selecting the items with the top-K rating predictions; henceforth, the quality of the rating predictions significantly affects the quality of the generated recommendations. However, certain types of users prefer to experience (purchase, listen to, watch, play) items the moment they become available in the stores, or even preorder, while other types of users prefer to wait for a period of time before experiencing, until a satisfactory amount of feedback (reviews and/or evaluations) becomes available for the item of interest. Notably, a user may apply varying practices on different item categories, i.e., be keen to experience new items in some categories while being uneager in other categories. To formulate successful recommendations, a recommender system should align with users' patterns of practice and avoid recommending a newly released item to users that delay to experience new items in the particular category, and vice versa. Insofar, however, no algorithm that takes into account this aspect has been proposed. In this work, we (1) present the Experiencing Period Criterion rating prediction algorithm (CFEPC) which modifies the rating prediction value based on the combination of the users' experiencing wait period in a certain item category and the period the rating to be predicted belongs to, so as to enhance the prediction accuracy of recommender systems and (2) evaluate the accuracy of the proposed algorithm using seven widely used datasets, considering two widely employed user similarity metrics, as well as four accuracy metrics. The results show that the CFEPC algorithm, presented in this paper, achieves a considerable rating prediction quality improvement, in all the datasets tested, indicating that the CFEPC algorithm can provide a basis for formulating more successful recommendations. [ABSTRACT FROM AUTHOR]

Heterogeneous information network (HIN) has recently been receiving increasing attention in recommender systems due to its practicability in depicting data heterogeneity. The rich structural and semantic information embodied in the HIN can help mining latent features of users and items for recommendations. However, almost all existing HIN-based recommendation methods focus on the design of complicated learning architecture while using simply initialized features. In this paper, we propose a novel feature-enhanced embedding learning model which combines informative feature initialization strategy with simple learning architecture for heterogeneous collaborative filtering. We first build multiple homogeneous sub-networks by extracting different relations guided by meta-paths from the HIN. We then design a comprehensive feature initialization strategy that contains semantic and spatial encoding module to characterize the node feature. After that, a simple learning architecture based on multi-layer perceptron is applied to learn the latent representation of users and items. Next, a novel convolutional neural network-based fusion mechanism is used to determine the attention weight of semantic relations and compress multiple embedding vectors into a compact representation to apply for final recommendation. Finally, we conduct extensive experiments on two classic datasets to demonstrate the effectiveness and feasibility of the proposed FHetCF method in solving HIN-based recommendation tasks. Results show that the proposed method soundly outperforms the competitive baselines by 1.71 to 10.46% on hit ratio and 3.17 to 13.75% on normalized discounted cumulative gain, respectively. The proposed method opens up a new avenue to effectively utilize heterogeneous information to improve recommendation performance. [ABSTRACT FROM AUTHOR]

Fairness is an important research problem for recommendation systems, and unfair recommendation methods can lead to discrimination against users. Gender is a kind of sensitive feature, exposure sensitive feature can lead to unfair treatment of males and females. However, gender discrimination is still a significant challenge that cannot be addressed well with current recommender methods. To alleviate this problem, we present a novel Unbias Gender Recommendation (UGRec) to balance performance between females and males. We propose a multihop graph aggregation mechanism to improve the representation of users and items, and we design a gender debias component with adversarial learning to eliminate gender bias by filtering out users' representations. With this design, UGRec can effectively filter out gender bias information and balance fairness between females and males. The experimental results based on three datasets demonstrate the effectiveness and advancement of our proposed UGRec model on fair recommendation. [ABSTRACT FROM AUTHOR]

Healthcare decision support systems aid physicians in disease classification by analyzing patients’ medical histories to suggest preliminary diagnoses. As physicians largely base their analysis on anamnesis, integrating this process into an automated recommendation system can expedite decision-making and transition to relevant clinical investigations, thus enhancing efficiency in diagnosing potential pathologies. In this research, an innovative method for feature construction is introduced, drawing on the concepts of Situation Awareness and Granular Computing. The aim of this method is to enhance the performance of out-of-the-box classification algorithms used in machine learning. The approach is specifically tailored to mimic physicians’ cognitive processes when analyzing a patient’s medical history, resulting in the generation of new, information-dense features that can be used for classification tasks. By employing this strategy, a deeper comprehension of the data can be achieved, as well as a more precise categorization of anamneses in relation to possible medical conditions. To authenticate the efficacy of the proposed technique, three major disease categories, namely cardiac, gastrointestinal, and thyroid, were considered. The dataset comprised 1213 medical histories. The experimental results indicate that the study’s six classifiers attained a balanced accuracy exceeding 90%. Among these, the SVM classifier demonstrated the highest balanced accuracy at 93%. Overall, the proposed approach resulted in an average increase of 16 percentage points in balanced accuracy, representing an improvement over the traditional methods commonly employed in machine learning. This approach could be integrated into a clinical decision support system, aiding physicians in accurately identifying necessary investigations and expediting diagnosis. [ABSTRACT FROM AUTHOR]

Most recommendation systems are implemented using numerical or categorical data, that is, traditional data. This type of data can be a limiting factor when used to model complex concepts where there is internal variability or internal structure in the data. To overcome these limitations, symbolic data is used, where data can be represented by different types of values, such as intervals, lists, or histograms. This work introduces a single approach to constructing recommendation systems based on content or based on collaborative filtering using modal variables for users and items. In the content-based system, user profiles and item profiles are created from modal representations of their features, and a list of items is matched against a user profile. For collaborative filtering, user profiles are built, and users are grouped to form a neighborhood, products rated by users of this neighborhood are recommended based on the similarity between the neighbor and the user who will receive the recommendation. Experiments are carried out, using a movie domain dataset, to evaluate the effectiveness of the proposed approach. The outcomes suggest our ability to generate ranked lists of superior quality compared to previous methods utilizing symbolic data. Specifically, the lists created through the proposed method exhibit higher normalized discounted cumulative gain and, in qualitative terms, showcase more diverse content. [ABSTRACT FROM AUTHOR]

Personalized recommendation plays a crucial role in Internet platforms, providing users with tailored content based on their user models and enhancing user satisfaction and experience. To address the challenge of information overload, it is essential to analyze user needs comprehensively, considering historical behavior and interests and the user's intentions and profiles. Previous user modeling approaches for personalized recommendations have exhibited certain limitations, relying primarily on historical behavior data to infer user preferences, which results in challenges such as the cold-start problem, incomplete modeling, and limited explanation. Motivated by recent advancements in large language models (LLMs), we present a novel approach to user modeling by embracing generative user modeling using LLMs. We propose generative user modeling with chain-of-thought prompting for personalized recommendation, which utilizes LLMs to generate comprehensive and accurate user models expressed in natural language and then employs these user models to empower LLMs for personalized recommendation. Specifically, we adopt the chain-of-thought prompting method to reason about user attributes, subjective preferences, and intentions, integrating them into a holistic user model. Subsequently, we utilize the generated user models as input to LLMs and design a collection of prompts to align the LLMs with various recommendation tasks, encompassing rating prediction, sequential recommendation, direct recommendation, and explanation generation. Extensive experiments conducted on real-world datasets demonstrate the immense potential of large language models in generating natural language user models, and the adoption of generative user modeling significantly enhances the performance of LLMs across the four recommendation tasks. Our code and dataset can be found at https://github.com/kwyyangfan/GUMRec. [ABSTRACT FROM AUTHOR]

As internet information technology continues to advance, individuals are increasingly encountering and managing a vast volume of data and information. A large and complex amount of information hinders the effective transmission of valuable information, making it difficult to find multimedia content of interest in the vastness of the internet. As the volume of multimedia content rapidly grows, personalized recommendation algorithms play a crucial role in matching relevant content to users. Knowledge graphs, due to their powerful organizational and relationship processing capabilities, are commonly applied in intelligent search engines and recommendation systems. This article focuses on the effective utilization of semantic association information in knowledge graphs for multimedia content recommendation. Two main areas of research are conducted. In this article, two novel approaches are presented. To begin with, the primary objective is to improve the learning of knowledge feature representation. This is achieved by introducing a model based on self-attention, which effectively captures the diverse significance of triplets in determining the semantics of entities. This leads to improved quality of knowledge feature representation, thereby serving as valuable auxiliary information for multimedia content recommendation systems. Secondly, the article addresses the integration of knowledge graphs in multimedia content recommendation applications. This paper proposes a content recommendation algorithm that integrates a combined embedding of behavior and knowledge features. By leveraging past preferences and utilizing the semantic structure of knowledge graphs, this algorithm provides a comprehensive exploration of user interests and hobbies. Ultimately, this conducts extensive experiments to assess the effectiveness and performance of the proposed algorithms. The results validate the feasibility and efficacy of these algorithms in enhancing multimedia content recommendation systems. [ABSTRACT FROM AUTHOR]

A primary strategy for constructing click-through rate models based on deep learning involves combining a multi-layer perceptron (MLP) with custom networks that can effectively capture the interactions between different features. This is due to the widespread recognition that relying solely on a vanilla MLP network is not effective in acquiring knowledge about multiplicative feature interactions. These custom networks often employ product methods, such as inner, Hadamard, and outer products, to construct dedicated architectures for this purpose. Among these methods, the outer product has shown superiority in capturing feature interactions. However, the resulting quadratic form from the outer product operation limits the conveyance of informative higher-order interactions to the MLP. Efforts to address this limitation have led to models attempting to increase interaction degrees to higher orders. However, utilizing matrix factorization techniques to reduce learning parameters has resulted in information loss and decreased performance. Furthermore, previous studies have constrained the MLP’s potential by providing it with inputs consisting of homogeneous outer products, thus limiting available information diversity. To overcome these challenges, we introduce DONN, a model that leverages a composite-wise bilinear module incorporating factorized bilinear pooling to mitigate information loss and facilitate higher-order interaction development. Additionally, DONN utilizes a feature-wise bilinear module for outer product computations between feature pairs, augmenting the MLP with combined information. By employing heterogeneous outer products, DONN enhances the MLP’s prediction capabilities, enabling the recognition of additional nonlinear interdependencies. Our evaluation on two benchmark datasets demonstrates that DONN surpasses state-of-the-art models in terms of performance. [ABSTRACT FROM AUTHOR]

The main objective of this paper is to clarify the importance of explainability in the crop recommendation process and provide insights on how Explainable Artificial Intelligence (XAI) can be incorporated into existing models successfully. The objective is to increase the definition and transparency of the recommendations implemented by AI in smart agriculture, leading to a detailed analysis of the synchronization between crop recommendation systems and XAI that informs decisions as it has sustainable knowledge and practices in modern agriculture. It reviews state-of-the-art XAI techniques such as local interpretable model-agnostic interpretation (LIME), SHapley interpretation additive approach (SHAP), integrated gradients (IG), and level-wise relevance propagation (LRP). It focuses on interpretable models and critical features analysis, and XAI methods are discussed in terms of their applications, critical features, and definitions. The paper found that XAI methods such as LIME and SHAP can make AI-driven crop recommendation systems more transparent and reliable. Graphical techniques such as dependency plots, summary plots, waterfall graphs, and decision plots effectively analyze feature importance. The paper includes counterfactual explanations using dice ml and hearing with advanced techniques combining IG and LRP to provide in-depth narrative model behavior. The novelty of this study lies in a detailed investigation of how XAI can be incorporated into crop recommendation systems to address the “black box” nature of AI models. It uses a unique XAI technique and model approach to make AI-driven recommendations more meaningful and practical for farmers. The proposed systems and techniques are designed to consume agriculture, addressing the specific needs of intelligent systems, making this research a significant contribution to agricultural AI. [ABSTRACT FROM AUTHOR]

Graph convolutional networks are capable of handling non-Euclidean data with sparse features, and some research has begun to apply them to the field of recommendation systems. Graph convolutional network’s aggregation and propagation mechanism can learn features well and improve the embedding quality. However, simply applying GCN to the recommendation domain can only show some of its advantages, and the complex structure makes it difficult for the model to handle the massive amount of data in the industrial domain. Some work has been done to integrate GCN with recommendation systems better. However, most related work pursues the model’s simplicity and ignores the large amount of hidden auxiliary information. In this paper, we propose a GCN-SA model, which adds a multi-head self-attention mechanism to the aggregation and propagation process to learn the weights of neighboring nodes and analyze the importance of the relationships between nodes; we also design a new embedding splicing layer for the graph convolutional network, which dynamically adjusts the embedding of different layers to achieve adaptive layer smoothing and mitigate the over-smoothing phenomenon. After experimental results on five benchmark datasets, we show that the GCN-SA model outperforms previous related work, captures a large amount of auxiliary information, and enhances the expressive ability of the model. [ABSTRACT FROM AUTHOR]

Identifying user requirements and preferences on the basis of the current context, is one of main challenges of the Internet of Things (IoT) paradigm. Users, services and applications interact maintaining, often unreliable, relations which need of smart management systems to satisfy their demands. Traditional information handling approaches in distributed systems are most often unsuitable for modern Smart Environments due to the huge amount and the extreme dynamism of the entities involved. This paper proposes NARIoT platform that allows building recommendation systems in IoT environment. The approach relies on vector representations of IoT resources obtained by using of a word embedding tool, the Doc2Vec neural model, which, starting from text documents describing the resources, provides real-valued vectors mapping them. The vectors are handled through intelligent agents, which self-organize themselves creating an ordered virtual structure, so enabling informed mechanisms of information filtering. In particular, an ordered overlay network emerges from the autonomous, parallel and decentralized work of intelligent agents, thus enabling efficient recommendation operations. The experimental validation confirms the effectiveness of the approach and provides very encouraging results. [ABSTRACT FROM AUTHOR]

For the recommendation domain, most of the existing integrated graph neural network (GNN)-based architectures have still much focused on encoding the associated extra side information in forms of heterogeneous information network (HIN). Then, it is simultaneously utilized in multiple fine-tuning processes to effectively learn preferences from user–item interaction data which might require tremendous computational efforts. In addition, these approaches have also failed to incorporate with previous learnt and transferable knowledge from pre-trained models to better fine-tune for recommendation task. To meet these challenges, in this paper we propose a novel heterogeneous graph neural architecture, named: PreHIN4Rec. The proposed PreHIN4Rec is considered as the graph pre-training approach for leveraging the performance of recommendation task in both accuracy and scalability aspects. In general, our proposed PreHIN4Rec is designed to efficiently preserve both heterogeneous schematic and local structural latent features of user–item interactions in forms of HINs. It can effectively support to better fine-tune for achieving remarkable improvements in recommendation tasks through integrating with existing recommendation frameworks. The extensive experiments in benchmark datasets demonstrate the effectiveness of our proposed PreHIN4Rec model in comparing with recent state-of-the-art GNN-based recommendation baselines. [ABSTRACT FROM AUTHOR]

Personalization systems have proved to be one of the most powerful tools for e-commerce sites, assisting users in discovering the most relevant products across enormous product catalogues. The formulation of product suggestions in the most widely used collaborative filtering is dependent on ratings contributed by the customer base. Though numerous domains consider allowing users to give an overall rating to products, a burgeoning number of online platforms are allowing users to rate products on a variety of dimensions. According to previous research, these multidimensional ratings offer valuable perceptions that can be used in generating a personalization list for users. Within the personalization systems research domain, multi-criteria systems have garnered significant attention since they use multiple criteria to predict rating scores. New strategies for leveraging information produced from multi-criteria scores to increase the prediction precision of multi-criteria (MC) systems are presented in this paper. In particular, we propose to fuse deep neural networks (DNN), matrix factorization (MF), and social spider optimization (SSO) to exploit nonlinear, non-trivial, and concealed interactions between users in terms of MC preferences. Experimenting on Yahoo! and TripAdvisor datasets reveals that our proposed approach outperforms both modern single-rating recommender systems based on MF and traditional multi-criteria systems. As a result, we believe that using multi-criteria customer evaluations can help e-commerce companies enhance the quality and specificity of their recommended services. [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]

The goal of session-based recommendation is to make item predictions at the next timestamp based on the anonymous ongoing session. Previous work mainly models user's preference by exploring the transition pattern between the interacted items in the session. However, they generally fail to pay enough attention to the item importance in terms of the relevance of the items to user's main purpose. This paper proposes a Session-based Recommendation approach with an Importance Extraction Module, i.e., SR-IEM improved , which can simultaneously consider user's long-term interactions and recent behavior in current session. Specifically, we modify the self-attention mechanism to avoid introducing bias for estimating the importance of each item in the session. Then, the item importances are utilized to produce user's long-term preference, and the sequential signals are incorporated in the long-term interest modeling. Next, the long-term preference and user's current interest which is conveyed by the last interacted item in the session are combined to obtain user's final preference representation. Finally, item predictions are generated using the user preference, where a normalization layer is adopted to solve the long-tail problem. Extensive experiments are conducted on three public benchmark datasets, i.e., Yoochoose 1/64, Yoochoose 1/4 and Diginetica. The experimental results show that SR-IEM improved can outperform the start-of-the-art baselines in terms of Recall and MRR for session-based recommendation. In addition, compared to the state-of-the-art neural methods, SR-IEM improved can obviously reduce the computational complexity. [ABSTRACT FROM AUTHOR]

Link prediction is a statistical relational learning problem that has a variety of applications in recommender systems, expert systems, and knowledge bases. Numerous approaches have already been devised to solve the problem. Tensor factorization is one of the ways to solve the link prediction problem. Many tensor factorization techniques have been devised in the last few decades, including Tucker, CANDECOMP/PARAFAC, and DEDICOM. RESCAL is one of the famous tensor factorization technique that can solve large scale problems with relatively less time and space complexity. The time complexity of RESCAL can further be reduced by making it parallel. This variant can also be applied to large scale datasets. This article focuses on devising a parallel version for RESCAL. A decent decrease in execution time has been observed in the execution of parallel RESCAL. [ABSTRACT FROM AUTHOR]

Currently, graph convolutional networks (GCN) have achieved significant progress in recommender systems, due to its remarkable capability on representation learning and the ability to integrate complex auxiliary information. However, the graph convolution operation is prone to cause over-smoothing due to the use of the graph Laplacian operator, so that the node embeddings become very similar after the multi-layer graph convolution, which leads to a decrease in recommendation performance. The recently proposed model based on simplified GCN can relieve this issue to a certain extent; however, they still only design the model from the viewpoint of GCN. Inspired by the recent developments of label propagation algorithms (LPA), in this paper, we propose a new recommender model that unifies graph convolutional networks and label propagation algorithms. Specifically, we utilize the GCN to build a basic recommendation prediction model, and unify the LPA to provide regularization of training edge weights, which has been proven to effectively alleviate the over-smoothing problem. In addition, we introduce an attention network to capture the attention weight of each user-item pair, which takes into account the fact that users attach different degrees of importance to various relationships of items. Extensive experiments on three real-world datasets demonstrate that the proposed algorithm has a significant improvement over other state-of-the-art recommendation algorithms. [ABSTRACT FROM AUTHOR]

We propose a tourist attraction IoT-enabled deep learning-based recommendation system to enhance tourist experience in a smart city. Travelers will enter details about their travels (traveling alone or with a companion, type of companion such as partner or family with kids, traveling for business or leisure, etc.) as well as user side information (age of the traveler/s, hobbies, etc.) into the smart city app/website. Our proposed deep learning-based recommendation system will process this personal set of input features to recommend the tourist activities/attractions that best fit his/her profile. Furthermore, when the tourists are in the smart city, content-based information (already visited attractions) and context-related information (location, weather, time of day, etc.) are obtained in real time using IoT devices; this information will allow our proposed deep learning-based tourist attraction recommendation system to suggest additional activities and/or attractions in real time. Our proposed multi-label deep learning classifier outperforms other models (decision tree, extra tree, k-nearest neighbor and random forest) and can successfully recommend tourist attractions for the first case [(a) searching for and planning activities before traveling] with the loss, accuracy, precision, recall and F1-score of 0.5%, 99.7%, 99.9%, 99.9% and 99.8%, respectively. It can also successfully recommend tourist attractions for the second case [(b) looking for activities within the smart city] with the loss, accuracy, precision, recall and F1-score of 3.7%, 99.5%, 99.8%, 99.7% and 99.8%, respectively. [ABSTRACT FROM AUTHOR]

In recent years, the ever-growing contents (movies, clothes, books, etc.) accessible and buyable via the Internet have led to the information overload issue and therefore the item targeting problem. Indeed, the huge mass of contents complexifies the identification of items fitting users' expectations. As powerful filtering tools, recommender systems efficiently alleviate the item targeting issue. Collaborative filtering-based methods are among the most influential algorithms adopted in recommender systems. Among collaborative filtering-based methods, model-based approaches are widely used in recent powerful recommendation methods. Due to its efficiency, the matrix factorization technique is spreadly employed in model-based approaches. However, those methods badly deal with issues such as data sparseness and cold-start problems that severely affect the recommendation quality. To overcome these limitations shown by state-of-the-art methods, we propose in this paper a recommender approach that couples the effectiveness of an enhanced matrix factorization technique to the power of a deep neural network model. In the first step, the user's latent factors and item latent factors are extracted from a doubly-regularized matrix factorization process. Thereafter, those latent factors are used to feed a deep learning structure in a forward-propagation process, and a normalized cross-entropy method is used to increase the precision of the deep neural network through a backpropagation process. The end prediction is made by combining results from the matrix factorization step and the deep neural structure. Extensive experiments are conducted on real-world datasets and show that our proposal outperforms other methods in terms of prediction accuracy and recommendation quality. [ABSTRACT FROM AUTHOR]

E-commerce platforms apply recommendation technology to a wide variety of commercial websites to help users quickly find the product they need from a large number of products. At present, recommendation systems have been widely used in large-scale e-commerce websites, and most e-commerce shopping websites attract customers through the image information of goods. In this paper, the research status of content-based image retrieval algorithms is analysed, and how to use image content information to recommend goods is studied. A hierarchical commodity classification and retrieval system are designed. A class decision layer is used to determine the category of commodity images and then precisely retrieve the corresponding category of commodity image features. The corresponding relationship between different commodity visual features is used to recommend commodities, and a matching recommendation algorithm for commodities is proposed. Experiments show that the proposed image content-based recommendation method can coordinate with features, and its recommendation results have high accuracy in practical applications. [ABSTRACT FROM AUTHOR]

Recently, the reciprocal recommendation, especially for online dating applications, has attracted increasing research attention. Different from the conventional recommendation problems, the reciprocal recommendation aims to simultaneously best match users' mutual interests to make the recommendations. However, the most existing RRS algorithms seldom model users' interest and attractiveness simultaneously under the high-dimensional feature space. Furthermore, the sparsity of reciprocal relations seriously deteriorates the recommendation performance. Thus, we propose a novel Deep Contrast Reciprocal Recommender System (DCRS) to address the aforementioned research issues. Particularly, we resolve the sparsity issue by introducing the reciprocal neighbors to increase the number of possible reciprocal relations. Then, a novel deep contrast neural network is then proposed to model the mutual interest by contrasting between the reciprocal and non-reciprocal relations. As a result, it was able to better identify the reciprocal relations for the latter recommendation. Extensive experiments have been evaluated on two real-world datasets, and the promising results demonstrate that the proposed DCRS is superior to both baseline and the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Recently, various cross-domain recommendation (CDR) models are proposed to overcome the sparsity problem, which leverage relatively abundant rating data from the auxiliary domain to improve recommendation performance of target domain. Though matrix factorization-based collaborative filtering algorithms gain dominance in single-domain recommendation systems, they cannot be used directly in cross-domain cases as the obtained latent factors of the target and auxiliary domains may not be aligned, which will lead to inaccurate knowledge transfer from the auxiliary domain to the target one. A CDR model named CDCFLFA is presented in this paper to solve this problem. In CDCFLFA, firstly latent factors between the two domains are aligned based on pattern matching. Then, user preferences of the auxiliary domain are transferred to update the original user latent vectors of target domain. Finally, a linear least square problem is solved to compute the item latent vectors of target domain and thus unknown ratings are obtained according to the updated user and item latent vectors. CDCFLFA does not require the same user or item sets between the two domains. Extensive experiments are conducted, and the results show that CDCFLFA achieves smaller MAE and RMSE values and larger precision and recall than the previous single- and cross-domain recommendation methods. Hence, CDCFLFA can be regarded as an effective cross-domain extension of single-domain matrix factorization algorithm. [ABSTRACT FROM AUTHOR]

In recent years, studies have revealed that introducing knowledge graphs (KGs) into recommendation systems as auxiliary information can improve recommendation accuracy. However, KGs are usually based on third-party data that may be manipulated by malicious individuals. In this study, we developed a poisoning attack strategy applied on a KG-based recommendation system to analyze the influence of fake links. The aim of an attacker is to recommend specific products to improve their visibility. Most related studies have focused on adversarial attacks on graph data; KG-based recommendation systems have rarely been discussed. We propose an attack model corresponding to recommendations. In the model, the current recommended status and a specified item are analyzed to estimate the effects of different attack decisions (addition or deletion of facts), thereby generating the optimal attack combination. Finally, the KG is contaminated by the attack combination so that the trained recommendation model recommends a specific item to as many people as possible. We formulated the process into a deep reinforcement learning method. Conducting experiments on the movie and the fund data sets enabled us to systematically analyze our poisoning attack strategy. The experimental results proved that the proposed strategy can effectively improve an item's ranking in a recommendation list. [ABSTRACT FROM AUTHOR]

In the context of recommender systems based on collaborative filtering (CF), obtaining accurate neighborhoods of the items of the datasets is relevant. Beyond particular individual recommendations, knowing these neighbors is fundamental for adding differentiating factors to recommendations, such as explainability, detecting shilling attacks, visualizing item relations, clustering, and providing reliabilities. This paper proposes a deep learning architecture to efficiently and accurately obtain CF neighborhoods. The proposed design makes use of a classification neural network to encode the dataset patterns of the items, followed by a generative process that obtains the neighborhood of each item by means of an iterative gradient localization algorithm. Experiments have been conducted using five popular open datasets and five representative baselines. The results show that the proposed method improves the quality of the neighborhoods compared to the K-Nearest Neighbors (KNN) algorithm for the five selected similarity measure baselines. The efficiency of the proposed method is also shown by comparing its computational requirements with that of KNN. [ABSTRACT FROM AUTHOR]