Search

Adding noise to user history data helps to protect user privacy in recommendation systems but affects the recommendation performance. To solve this problem, a matrix factorization tourism point of interest recommendation model based on interest offset and differential privacy is proposed in this paper. The recommendation performance of the model is improved by analyzing user interest preferences. Specifically, user interest offsets are extracted from user tags and user ratings under time-series factors to calculate user interest drift. Then, similar neighbors are found to train user feature preferences which are integrated into the matrix model in the form of regular terms. Meanwhile, based on the differential privacy theory, a privacy neighbor selection algorithm combining the K-Medoides clustering algorithm and index mechanism is designed to effectively protect the identity of neighbors and prevent KNN attacks. Besides, the Laplace mechanism is used to implement differential privacy protection for the model's gradient descent process. Finally, the feasibility of the proposed recommendation model is verified through experiments, and the experimental results indicate that this model has advantages in recommendation accuracy and privacy protection. [ABSTRACT FROM AUTHOR]

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

Knowledge graph (KG) has been widely used in the field of recommender systems. There are some nodes in KG that guide the occurrence of interaction behaviors. We call them guided nodes. However, the current application doesn't take into account the guided nodes in KG. We explore the utility of guided nodes in KG. It is applied in repository recommendations. In this paper, we propose an end-to-end framework, namely Guided Node Graph Convolutional Network (GNGCN), which effectively captures the connections between entities by mining the influence of related nodes. We extract samples of each entity in KG as their guided nodes and then combine the information and bias of the guided nodes when computing the representation of a given entity. The guided nodes can be extended to multiple hops. We evaluate our model on a real-world Github dataset named Github-SKG and music recommendation dataset, and the experimental results show that the method outperforms the recommendation baselines and our model is much lighter than others. [ABSTRACT FROM AUTHOR]

Among the primary aims of Artificial Intelligence (AI) is the enhancement of User Experience (UX) by providing deep understanding, profound empathy, tailored assistance, useful recommendations, and natural communication with human interactants while they are achieving their goals through computer use. To this end, AI is used in varying techniques to automate sophisticated functions in UX and thereby changing what UX is apprehended by the users. This is achieved through the development of intelligent interactive systems such as virtual assistants, recommender systems, and intelligent tutoring systems. The changes are well received, as technological achievements but create new challenges of trust, explainability and usability to humans, which in turn need to be amended by further advancements of AI in reciprocity. AI can be utilised to enhance the UX of a system while the quality of the UX can influence the effectiveness of AI. The state of the art in AI for UX is constantly evolving, with a growing focus on designing transparent, explainable, and fair AI systems that prioritise user control and autonomy, protect user data privacy and security, and promote diversity and inclusivity in the design process. Staying up to date with the latest advancements and best practices in this field is crucial. This paper conducts a critical analysis of published academic works and research studies related to AI and UX, exploring their interrelationship and the cause-effect cycle between the two. Ultimately, best practices for achieving a successful interrelationship of AI in UX are identified and listed based on established methods or techniques that have been proven to be effective in previous research reviewed. [ABSTRACT FROM AUTHOR]

Knowledge graphs have been introduced into recommender systems due to the rich connectivity information. Many knowledge-aware recommendation methods use graph neural networks (GNNs) to capture the high-order structural and semantic information of knowledge graphs. However, previous GNN-based methods have the following limitations: (1) they fail to make full use of the neighborhood information of entities and (2) they ignore the importance of user interaction sequences on reflecting user preferences. As such, these models are insufficient for generating accurate representations of users and items. In this study, we propose a Knowledge-aware Hierarchical Attention Network (KHAN) to provide better recommendation. Specifically, the proposed model mainly consists of an item encoder and a user encoder. The item encoder is equipped with a hierarchical attention network, which is used to generate entity (item) representations by carefully aggregating neighborhood information of entities. The user encoder is also designed to learn more informative user representations from user interaction sequences using multi-head self-attention. The learned user representations are then combined with user representations introduced in the item encoder through a gating mechanism to generate the final user representations. Extensive experiments on two real-world datasets about movie and restaurant recommendation demonstrate the effectiveness of our model. [ABSTRACT FROM AUTHOR]

Background: Mobile devices and their capabilities (e.g., device sensors and human-device interactions) are increasingly being considered for use in clinical assessments and disease monitoring due to their ability to provide objective, repeatable, and more accurate measures of neurocognitive performance. These mobile-based assessments also provide a foundation for the design of intervention recommendations. Objective: The purpose of this work was to assess the benefits of various physical intervention programs as they relate to Parkinson's disease (PD), its symptoms, and stages (Hoehn and Yahr (H&Y) Stages 1–5). Methods: Ninety-five participants (n = 70 PD; n = 25 control) completed 14 tablet-based neurocognitive functional tests (e.g., motor, memory, speech, executive, and multi-function) and standardized health questionnaires. 208 symptom-specific digital features were normalized to assess the benefits of various physical intervention programs (e.g., aerobic activity, non-contact boxing, functional strength, and yoga) for individuals with PD. While previous studies have shown that physical interventions improve both motor and non-motor PD symptoms, this paper expands on previous works by mapping symptom-specific neurocognitive functionalities to specific physical intervention programs across stages of PD. Results: For early-stage PD (e.g., H&Y Stages 1 & 2), functional strength activities provided the largest overall significant delta improvement (Δ= 0.1883; p = 0.0265), whereas aerobic activity provided the largest overall significant delta improvement (Δ= 0.2700; p = 0.0364) for advanced stages of PD (e.g., H&Y Stages 3–5). Conclusions: As mobile-based digital health technology allows for the collection of larger, labeled, objective datasets, new ways to analyze and interpret patterns in this data emerge which can ultimately lead to new personalized medicine programs. [ABSTRACT FROM AUTHOR]

With the rapid development of social networks, studying and analyzing their structures and behaviors has become one of the most important requirements of businesses. Social network analysis can be used for many different purposes such as product ads, market orientation detection, influential members detection, predicting user behaviors, recommender systems improvements, etc. One of the newest research topics in social network analysis is the enhancement of the information propagation performance in different aspects based on application. In this paper, a new method is proposed to improve few metrics such as distribution time and precision on social networks. In this method, the local attributes of nodes and also the structural information of the network is used to forward data across the network and reduce the propagation time. First of all, the centrality and Assortativity are calculated for all nodes separately to select two sets of nodes with the highest values for both criteria. Then, the initial active nodes of the network are selected by calculating the intersection of the two sets. Next, the distribution paths are detected based on the initial active nodes to calculate the propagation time. The performance analysis results show that the proposed method has better outcomes in comparison to other state-of-the-art methods in terms of distribution time, precision, recall, and AUPR criteria. [ABSTRACT FROM AUTHOR]

Semantic alignment is a key component in Cross-Language Text Matching (CLTM) to facilitate matching (e.g., query-document matching) between two languages. The current solutions for semantic alignment mainly perform word-level translation directly, without considering the contextual information for the whole query and documents. To this end, we propose a Dual-Level Collaborative Rough-to-Fine Filter Alignment Network (DLCCFA) to achieve better cross-language semantic alignment and document matching. DLCCFA is devised with both a coarse-grained filter in word-level and a fine-grained filter in sentence-level. Concretely, for the query in word-level, we firstly extract top-k translation candidates for each token in the query through a probabilistic bilingual lexicon. Then, a Translation Probability Attention (TPA) mechanism is proposed to obtain coarse-grained word alignment, which generates the corresponding query auxiliary sentence. Afterwards, we further propose a Bilingual Cross Attention and utilize Self-Attention to achieve fine-grained sentence-level filtering, resulting in the cross-language representation of the query. The idea is that each token in the query works as an anchor to filter the semantic noise in the query auxiliary sentence and accurately align semantics of different languages. Extensive experiments on four real-world datasets of six languages demostrate that our method can outperform the mainstream alternatives of CLTM. [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]

With the rapid growth of data scale, the problems of collaborative filtering recommendation algorithm are more and more obvious, such as data sparsity, cold start, scalability, and the change of user interest over time. About the existing problems, we introduce the fuzzy clustering and propose a collaborative filtering algorithm based on fuzzy C-means clustering. The algorithm performs fuzzy clustering on the item attribute information to make items belonging to different categories in different membership degree, increases the data density, effectively reduces the data sparsity, and solves the issue that the inaccuracy of similarity leads to the low recommendation accuracy. Meanwhile, the algorithm introduces the time weight function. Different evaluation times give different time weight values, and recently evaluated items are more representative of the user current interest, so we give a higher weight value, and early evaluated items have less effect on the user current interest, thus the weight value are relatively lower. The experimental results show that our algorithm can effectively alleviate the data sparsity problem and time migration of users preferences, thus achieve better performance. [ABSTRACT FROM AUTHOR]

Nowadays, online services, like e-commerce or streaming services, provide a personalized user experience through recommender systems. Recommender systems are built upon a vast amount of data about users/items acquired by the services. Such knowledge represents an invaluable resource. However, commonly, part of this knowledge is public and can be easily accessed via the Internet. Unfortunately, that same knowledge can be leveraged by competitors or malicious users. The literature offers a large number of works concerning attacks on recommender systems, but most of them assume that the attacker can easily access the full rating matrix. In practice, this is never the case. The only way to access the rating matrix is by gathering the ratings (e.g., reviews) by crawling the service's website. Crawling a website has a cost in terms of time and resources. What is more, the targeted website can employ defensive measures to detect automatic scraping. In this paper, we assess the impact of a series of attacks on recommender systems. Our analysis aims to set up the most realistic scenarios considering both the possibilities and the potential attacker's limitations. In particular, we assess the impact of different crawling approaches when attacking a recommendation service. From the collected information, we mount various profile injection attacks. We measure the value of the collected knowledge through the identification of the most similar user/item. Our empirical results show that while crawling can indeed bring knowledge to the attacker (up to 65% of neighborhood reconstruction on a mid-size dataset and up to 90% on a small-size dataset), this will not be enough to mount a successful shilling attack in practice. [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]

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]

Nowadays, online services, like e-commerce or streaming services, provide a personalized user experience through recommender systems. Recommender systems are built upon a vast amount of data about users/items acquired by the services. Such knowledge represents an invaluable resource. However, commonly, part of this knowledge is public and can be easily accessed via the Internet. Unfortunately, that same knowledge can be leveraged by competitors or malicious users. The literature offers a large number of works concerning attacks on recommender systems, but most of them assume that the attacker can easily access the full rating matrix. In practice, this is never the case. The only way to access the rating matrix is by gathering the ratings (e.g., reviews) by crawling the service's website. Crawling a website has a cost in terms of time and resources. What is more, the targeted website can employ defensive measures to detect automatic scraping. In this paper, we assess the impact of a series of attacks on recommender systems. Our analysis aims to set up the most realistic scenarios considering both the possibilities and the potential attacker's limitations. In particular, we assess the impact of different crawling approaches when attacking a recommendation service. From the collected information, we mount various profile injection attacks. We measure the value of the collected knowledge through the identification of the most similar user/item. Our empirical results show that while crawling can indeed bring knowledge to the attacker (up to 65% of neighborhood reconstruction on a mid-size dataset and up to 90% on a small-size dataset), this will not be enough to mount a successful shilling attack in practice. [ABSTRACT FROM AUTHOR]

The major challenge of recommendation system (RS) based on implict feedback is to accurately model users' preferences from their historical feedback. Nowadays, researchers has tried to apply adversarial technique in RS, which had presented successful results in various domains. To a certain extent, the use of adversarial technique improves the modeling of users' preferences. Nonetheless, there are still many problems to be solved, such as insufficient representation and low-level interaction. In this paper, we propose a recommendation algorithm NCGAN which combines neural collaborative filtering and generative adversarial network (GAN). We use the neural networks to extract users' non-linear characteristics. At the same time, we integrate the GAN framework to guide the recommendation model training. Among them, the generator aims to make user recommendations and the discriminator is equivalent to a measurement tool which could measure the distance between the generated distribution and users' ground distribution. Through comparison with other existing recommendation algorithms, our algorithm show better experimental performance in all indicators. [ABSTRACT FROM AUTHOR]

Collaborative filtering (CF), a representative algorithm of recommendation systems, is a method of using information of the neighbors of active user. The main idea of CF is that users who agreed in the ratings of certain items are likely to agree again in new items. The degree to which the two users' tendencies in the ratings of the co-rated items are consistent is measured using a similarity measure. Therefore, the similarity measure in CF plays a key role in the extraction of the representative neighbors. Studies on the improvement of similarity indicators for selecting representative neighbors are still ongoing. Recently, a new similarity measure, named OS, was proposed to enhance the recommendation performance by utilizing mathematical equations, such as the integral equation, system of linear differential equations, and non-linear systems. This study aims to understand the limitations of OS and overcome these limitations using the proposed method. In the proposed method, a sigmoid function was used to reflect preferences, such as the positive or negative sentiment of user ratings. In addition, to consider the absolute score difference, some of the formulas were modified, and finally, the performance improvement of the recommendation system was proved through experiments. [ABSTRACT FROM AUTHOR]

The purpose of aspect-level sentiment analysis is to identify the contextual sentence expressions given by sentiment for some aspects. For previous works, many scholars have proved the importance of the interaction between aspects and contexts. However, most existing methods ignore or do not specifically capture the position information of the aspect targets in the sentence. Thus, we propose an aspect-level sentiment analysis based on joint aspect and position hierarchy attention mechanism network. At the same time, the model adopts a joint approach to make the model of the aspect features and the position features. On the one hand, this method clearly captures the interaction between aspect words and context when inputting word vector information. On the other hand, this method can enhance the importance of position information in the sentence and boost the information retrieval ability of the model. Additionally, the model utilizes a hierarchical attention mechanism to extract feature information and to differentiate sentiment towards, which is similar to filtering useless information again. Experiment on the SemEval 2014 dataset represent that our model achieves better performance on aspect-level sentiment classification. [ABSTRACT FROM AUTHOR]

Recommender systems apply machine learning and data mining techniques for filtering unseen information, and they can provide an opportunity to predict whether a user would be interested in a given item. The main types of recommender systems are collaborative filtering (CF) and content-based filtering, which suffer from scalability and data sparsity resulting in poor quality recommendations and reduced coverage. There are two incremental algorithms based on Singular Value Decomposition (SVD) with high scalability for recommender systems which are named the incremental SVD algorithm and incremental Approximating the Singular Value Decomposition (ApproSVD) algorithm. In both mentioned methods, the estimated value of rank for approximating the recommender systems' data matrix is chosen experimentally in the related literature. In this paper, we investigate the role of singular values for estimating a more reliable amount of rank in the mentioned dimensionality reduction techniques to improve the recommender systems' performance. In other words, we offered a strategy for choosing the optimal rank that approximates the data matrix more accurately in incremental algorithms with the help of singular values. The numerical results illustrate that the suggested strategy improves the accuracy of the recommendations and run times of both algorithms when employs for Movielens, Netflix, and Jester dataset. [ABSTRACT FROM AUTHOR]

Ordinal regression has been widely used in applications, such as credit portfolio management, recommendation systems, and ecology, where the core task is to predict the labels on ordinal scales. Due to its learning efficiency, online ordinal regression using passive aggressive (PA) algorithms has gained a much attention for solving large-scale ranking problems. However, the PA method is sensitive to noise especially in the scenario of streaming data, where the ranking of data samples may change dramatically. In this paper, we propose a noise-resilient online learning algorithm using the Ramp loss function, called PA-RAMP, to improve the performance of PA method for noisy data streams. Also, we validate the order preservation of thresholds of the proposed algorithm. Experiments on real-world data sets demonstrate that the proposed noise-resilient online ordinal regression algorithm is more robust and efficient than state-of-the-art online ordinal regression algorithms. [ABSTRACT FROM AUTHOR]

Explainable recommendations enable users to understand why certain items are suggested and, ultimately, nurture system transparency, trustworthiness, and confidence. Large crowdsourcing recommendation systems ought to crucially promote authenticity and transparency of recommendations. To address such challenge, this paper proposes the use of stream-based explainable recommendations via blockchain profiling. Our contribution relies on chained historical data to improve the quality and transparency of online collaborative recommendation filters – Memory-based and Model-based – using, as use cases, data streamed from two large tourism crowdsourcing platforms, namely Expedia and TripAdvisor. Building historical trust-based models of raters, our method is implemented as an external module and integrated with the collaborative filter through a post-recommendation component. The inter-user trust profiling history, traceability and authenticity are ensured by blockchain, since these profiles are stored as a smart contract in a private Ethereum network. Our empirical evaluation with HotelExpedia and Tripadvisor has consistently shown the positive impact of blockchain-based profiling on the quality (measured as recall) and transparency (determined via explanations) of recommendations. [ABSTRACT FROM AUTHOR]

At present, the mankind of the entire world is under serious threat due to the unexpected COVID-19 pandemic. The advent of this pandemic exposes many drawbacks in the medical and healthcare system. As per the guidelines of WHO, the spread of the virus must be controlled through proper measures that help cease the virus. Tracing infected subjects (people/patients) is exceedingly difficult across the globe. The testing process in many countries is hampered by the unavailability of COVID-19 Test kits. Therefore, a testing process needs a robust mechanism to identify the infected subject to reduce the infection rate. To address this issue, a Symptom-based COVID-19 Test Recommendation System using Machine Learning methods is proposed and tested on real data set. It is found that the results of the system are promising and accurate up to 99%. The proposed piece of work undergoes four steps. First, it creates synthesized data set by using inputs of the Superintendent of Physical Health Centre (Rajam). Second, the synthesized data set is balanced by using Random under-sampling (RUS) followed by Synthetic minority oversampling (SMOTE). Third, different machine learning techniques such as K-Nearest Neighbor (KNN), Decision Tree (DT), Naïve Bayes, Random Forest (RF), Stochastic Gradient Descent (SGD), and Support vector machine (SVM) are applied on both the Synthesized and balanced data sets to classify subjects into different classes based on age, comorbidity-chronic disease- and other symptoms (cold, cough, fever, and breathlessness). Finally, the COVID-19 Test Recommended System is created and integrated with the best classification model. From the experimental results, it is observed that the training and testing accuracy of all the classification models is more than 99% consequently, the COVID-19 Testing recommended system also gives 100% accuracy in predicting the category of a subject based on input symptoms. [ABSTRACT FROM AUTHOR]

Tensors have been explored to share latent user-item relations and have been shown to be effective for recommendation. Tensors suffer from sparsity and cold start problems in real recommendation scenarios; therefore, researchers and engineers usually use matrix factorization to address these issues and improve the performance of recommender systems. In this paper, we propose matrix factorization completed multicontext data for tensor-enhanced algorithm a using matrix factorization combined with a multicontext data method for tensor-enhanced recommendation. To take advantage of existing user-item data, we add the context time and trust to enrich the interactive data via matrix factorization. In addition, Our approach is a high-dimensional tensor framework that further mines the latent relations from the user-item-trust-time tensor to improve recommendation performance. Through extensive experiments on real-world datasets, we demonstrated the superiority of our approach in predicting user preferences. This method is also shown to be able to maintain satisfactory performance even if user-item interactions are sparse. [ABSTRACT FROM AUTHOR]

Recently, the e-learners are drastically increased from the last two decades. Everything is learnt through internet without help of the tutor as well. For this purpose, the e-learners are required more e-learning applications that are able to supply optimal and satisfied data based on their capability. No content recommendation system is available for recommending suitable contents to the learners. For this purpose, this paper proposes a new semantic and fuzzy aware content recommendation system for retrieving the suitable content for the users. In this content recommendation system, we propose two content pre-processing algorithms namely Target Keyword based Data Pre-processing Algorithm (TKDPA) and Intelligent Anova-T Residual Algorithm (IAATRA) for selecting the more relevant features from the document. Moreover, a new Fuzzy rule based Similarity Matching algorithm (FRSMA) is proposed and used in this system for finding the similarity between the two terms and also rank them by using the newly proposed Similarity and Temporal aware Weighted Document Ranking Algorithm (STWDRA). In addition, a content clustering process is also incorporated for gathering relevant content. Finally, a new Fuzzy, Target Keyword and Similarity Score based Content Recommendation Algorithm (FTKSCRA) is also proposed for recommending the more relevant content to the learners accurately. The experiments have been conducted for evaluating the proposed content recommendation system and proved as better than the existing recommendation systems in terms of precision, recall, f-measure and prediction accuracy. [ABSTRACT FROM AUTHOR]

Recommender System (RS) is an information filtering approach that helps the overburdened user with information in his decision making process and suggests items which might be interesting to him. While presenting recommendation to the user, accuracy of the presented list is always a concern for the researchers. However, in recent years, the focus has now shifted to include the unexpectedness and novel items in the list along with accuracy of the recommended items. To increase the user acceptance, it is important to provide potentially interesting items which are not so obvious and different from the items that the end user has rated. In this work, we have proposed a model that generates serendipitous item recommendation and also takes care of accuracy as well as the sparsity issues. Literature suggests that there are various components that help to achieve the objective of serendipitous recommendations. In this paper, fuzzy inference based approach is used for the serendipity computation because the definitions of the components overlap. Moreover, to improve the accuracy and sparsity issues in the recommendation process, cross domain and trust based approaches are incorporated. A prototype of the system is developed for the tourism domain and the performance is measured using mean absolute error (MAE), root mean square error (RMSE), unexpectedness, precision, recall and F-measure. [ABSTRACT FROM AUTHOR]

Legal practitioners analyze relevant previous judgments to prepare favorable and advantageous arguments for an ongoing case. In Legal domain, recommender systems (RS) effectively identify and recommend referentially and/or semantically relevant judgments. Due to the availability of enormous amounts of judgments, RS needs to compute pairwise similarity scores for all unique judgment pairs in advance, aiming to minimize the recommendation response time. This practice introduces the scalability issue as the number of pairs to be computed increases quadratically with the number of judgments i.e., O (n2). However, there is a limited number of pairs consisting of strong relevance among the judgments. Therefore, it is insignificant to compute similarities for pairs consisting of trivial relevance between judgments. To address the scalability issue, this research proposes a graph clustering based novel Legal Document Recommendation System (LDRS) that forms clusters of referentially similar judgments and within those clusters find semantically relevant judgments. Hence, pairwise similarity scores are computed for each cluster to restrict search space within-cluster only instead of the entire corpus. Thus, the proposed LDRS severely reduces the number of similarity computations that enable large numbers of judgments to be handled. It exploits a highly scalable Louvain approach to cluster judgment citation network, and Doc2Vec to capture the semantic relevance among judgments within a cluster. The efficacy and efficiency of the proposed LDRS are evaluated and analyzed using the large real-life judgments of the Supreme Court of India. The experimental results demonstrate the encouraging performance of proposed LDRS in terms of Accuracy, F1-Scores, MCC Scores, and computational complexity, which validates the applicability for scalable recommender systems. [ABSTRACT FROM AUTHOR]

Volume 27(6) of the Intelligent Data Analysis (IDA) Journal contains fifteen articles covering theoretical contributions and applied Intelligent Data Analysis. The theoretical papers discuss topics such as new methods for data preparation and preprocessing, graph-based recommendation systems, and Bayesian network classifiers. The application papers focus on computer vision and image processing, including the detection of small objects in unmanned aerial vehicle images and the detection and classification of brain tumors using Magnetic Resonance Images. Other application papers cover topics such as classifying smart contracts, identifying important nodes in social networks, extracting failure evidence from complex hybrid data in QA systems, and producing note-level transcriptions of music melodies. [Extracted from the article]

A large number of practical applications of the recommendation system found that the novelty of the recommendation results and the user satisfaction are more closely related, making the novelty recommendation recently widely concerned and studied. Many novelty recommendation algorithms used the popularity of the item to measure novelty, but this method is too simple, and the change of item popularity is more reflective of its novelty. According to the product life cycle theory (PLC), this study proposed a novelty recommendation algorithm that recommends item that be not popular now and may be popular in the future to improve the novelty of the recommendation results, The time change of the popularity of the items to be recommended is analyzed, and the future popularity of the items are predicted by analogy. Two strategies for selecting recommended selection are selecting future popular items (the predicting popularity-based filtering Algorithm, PP algorithm) and excluding future recession items (the Excluding Recession-based filtering algorithm, ER algorithm), according to the definition of novelty of the item, recommended the novelty items to the target user. The effectiveness of the proposed algorithm was verified through an offline experiment. Results indicate that PP algorithm can significantly improve the accuracy and novelty, but seriously sacrifice the coverage and reduce the ability of the recommendation system to mine the long tail items when the number of alternative items N is small, the novelty of the recommendation list of the ER algorithm is remarkably higher than that of traditional algorithms, the novelty is high when the quantity of alternative sets reaches around 350, where the average popularity of the recommendation list declines by 40%, and the coverage is elevated by 150%, thereby improving the ability of the proposed system to extract all kinds of items. This study serves as reference for the improvement of user satisfaction with recommendation systems. [ABSTRACT FROM AUTHOR]

The new user cold-start problem is a grand challenge in content-based music recommender systems. This happens when the systems do not have sufficient information regarding the user's preferences. Towards solving this problem, in this study, a rating prediction framework is proposed. The proposed framework allows the systems to predict the user's rating scores for unrated musical pieces, by which good recommendations can be generated. The core idea here is to leverage the so-called MUSIC model, i.e., a five-factor musical preference model, which is characterized by Mellow, Unpretentious, Sophisticated, Intense, and Contemporary as the user's musical preference profiles. When a user newly joins the systems, the first five-factor musical preference profile is established based on the user's age and brain type information which is extracted from questionnaires. When the user experiences the systems for a certain period, his/her rating scores for experienced musical pieces are utilized for generating the second five-factor musical preference profile. The recommendations are then provided based on the rating scores predicted from a non-linear combination of these two five-factor musical preference profiles. The results demonstrated the effectiveness of the five-factor musical preference in alleviating the new user cold-start problem. In addition, the proposed method can potentially provide high-quality recommendations. [ABSTRACT FROM AUTHOR]

Matrix factorization (MF) models are effective and easy to expand and are widely used in industry, such as rating prediction and item recommendation. The basic MF model is relatively simple. In practical applications, side information such as attributes or implicit feedback is often combined to improve accuracy by modifying the model and optimizing the algorithm. In this paper, we propose an attribute interaction-aware matrix factorization (AIMF) method for recommendation tasks. We partition the original rating matrix into different sub-matrices according to the attribute interactions, train each sub-matrix independently, and merge all the latent vectors to generate the final score. Since the generated sub-matrices vary in size, an adaptive regularization coefficient optimization strategy and an adaptive latent vector dimension optimization strategy are proposed for sub-matrix training, and a variety of latent vector merging methods are put forward. The method AIMF has two advantages. When the original rating matrix is particularly large, the training time complexity of the MF-based model becomes higher and the update cost of the model is also higher. In AIMF, because each sub-matrix is usually much smaller than the original rating matrix, the training time complexity is greatly reduced after using parallel computing technology. Secondly, in AIMF, it is not necessary to modify the matrix factorization model to incorporate attributes and their interactive information into the model to improve the performance. The experimental results on the two classic public datasets MovieLens 1M and MovieLens 100k show that AIMF can not only effectively improve the accuracy of recommendation, but also make full use of parallel computing technology to improve training efficiency without modifying the matrix factorization model. [ABSTRACT FROM AUTHOR]

Identifying potential social media influencers (SMIs) accurately can achieve a long-time and effective concept marketing at a lower cost, and then promote the development of the corporate brand in online communities. However, potential SMIs discrimination often faces the problem of insufficient available information of the long-term evolution of the network, and the existing discriminant methods based on link analysis fail to obtain more accurate results. To fill this gap, a consensus smart discriminant algorithm (CSDA) is proposed to identify the potential SMIs with the aid of attention concentration (AC) between users in a closed triadic structure. CSDA enriches and expands the users' AC information by fusing multiple attention concentration indexes (ACIs) as well as filters the noise information caused by multi-index fusion through consensus among the indexes. Specifically, to begin with, to enrich the available long-term network evolution information, the unidirectional attention concentration indexes (UACIs) and the bidirectional attention concentration indexes (BACIs) are defined; next, the consensus attention concentration index (CACI) is selected according to the principle of minimum upper and lower bounds of link prediction bias to filter noise information; the potential SMI is determined by adaptively calculating CACI among the user to be identified, unconnected user group and their common neighbor. The validity and reliability of the proposed method are verified by the actual data of Twitter. [ABSTRACT FROM AUTHOR]

With the increase of online businesses, recommendation algorithms are being researched a lot to facilitate the process of using the existing information. Such multi-criteria recommendation (MCRS) helps a lot the end-users to attain the required results of interest having different selective criteria – such as combinations of implicit and explicit interest indicators in the form of ranking or rankings on different matched dimensions. Current approaches typically use label correlation, by assuming that the label correlations are shared by all objects. In real-world tasks, however, different sources of information have different features. Recommendation systems are more effective if being used for making a recommendation using multiple criteria of decisions by using the correlation between the features and items content (content-based approach) or finding a similar user rating to get targeted results (Collaborative filtering). To combine these two filterings in the multicriteria model, we proposed a features-based fb-knn multi-criteria hybrid recommendation algorithm approach for getting the recommendation of the items by using multicriteria features of items and integrating those with the correlated items found in similar datasets. Ranks were assigned to each decision and then weights were computed for each decision by using the standard deviation of items to get the nearest result. For evaluation, we tested the proposed algorithm on different datasets having multiple features of information. The results demonstrate that proposed fb-knn is efficient in different types of datasets. [ABSTRACT FROM AUTHOR]

Knowledge Graphs (KGs) are composed of structured information about a particular domain in the form of entities and relations. In addition to the structured information KGs help in facilitating interconnectivity and interoperability between different resources represented in the Linked Data Cloud. KGs have been used in a variety of applications such as entity linking, question answering, recommender systems, etc. However, KG applications suffer from high computational and storage costs. Hence, there arises the necessity for a representation able to map the high dimensional KGs into low dimensional spaces, i.e., embedding space, preserving structural as well as relational information. This paper conducts a survey of KG embedding models which not only consider the structured information contained in the form of entities and relations in a KG but also its unstructured information represented as literals such as text, numerical values, images, etc. Along with a theoretical analysis and comparison of the methods proposed so far for generating KG embeddings with literals, an empirical evaluation of the different methods under identical settings has been performed for the general task of link prediction. [ABSTRACT FROM AUTHOR]

Recommender Systems is a highly applicable subclass of information filtering systems, aiming to provide users with personalized item suggestions. These systems build on collaborative filtering and content-based methods to overcome the information overload issue. Hybrid recommender systems combine the abovementioned methods and are generally proved to be more efficient than the classical approaches. In this paper, we propose a novel approach for the development of a hybrid recommender system that is able to make recommendations under the limitation of processing small amounts of data with strong intercorrelation. The proposed hybrid solution integrates Machine Learning and Multi-Criteria Decision Analysis algorithms. The experimental evaluation of the proposed solution indicates that it performs better than widely used Machine Learning algorithms such as the k -Nearest Neighbors and Decision Trees. [ABSTRACT FROM AUTHOR]

The recommendation system is affected with attacks when the users are given liberty to rate the items based on their impression about the product or service. Some malicious user or other competitors' try to inject fake rating to degrade the item's graces that are mostly adored by several users. Attacks in the rating matrix are not executed just by a single profile. A group of users profile is injected into rating matrix to decrease the performance. It is highly complex to extract the fake ratings from the mixture of genuine profile as it resides the same pattern. Identifying the attacked profile and the target item of the fake rating is a challenging task in the big data environment. This paper proposes a unique method to identify the attacks in collaborating filtering method. The process of extracting fake rating is carried out in two phases. During the initial phase, doubtful user profile is identified from the rating matrix. In the following phase, the target item is analysed using push attack count to reduce the false positive rates from the doubtful user profile. The proposed model is evaluated with detection rate and false positive rates by considering the filler size and attacks size. The experiment was conducted with 6%, 8% and 10% filler sizes and with different attack sizes that ranges from 0%–100%. Various classification techniques such as decision tree, logistic regression, SVM and random forest methods are used to classify the fake ratings. From the results, it is witnessed that SVM model works better with random and bandwagon attack models at an average of 4% higher accuracy. Similarly the decision tree method performance better at an average of 3% on average attack model. [ABSTRACT FROM AUTHOR]

Context-Aware Recommender System (CARS) suggests more relevant services by adapting them to the user's specific context situation. Nevertheless, the use of many contextual factors can increase data sparsity while few context parameters fail to introduce the contextual effects in recommendations. Moreover, several CARSs are based on similarity algorithms, such as cosine and Pearson correlation coefficients. These methods are not very effective in the sparse datasets. This paper presents a context-aware model to integrate contextual factors into prediction process when there are insufficient co-rated items. The proposed algorithm uses Latent Dirichlet Allocation (LDA) to learn the latent interests of users from the textual descriptions of items. Then, it integrates both the explicit contextual factors and their degree of importance in the prediction process by introducing a weighting function. Indeed, the PSO algorithm is employed to learn and optimize weights of these features. The results on the Movielens 1 M dataset show that the proposed model can achieve an F-measure of 45.51% with precision as 68.64%. Furthermore, the enhancement in MAE and RMSE can respectively reach 41.63% and 39.69% compared with the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

The existing recommender system provides personalized recommendation service for users in online shopping, entertainment, and other activities. In order to improve the probability of users accepting the system's recommendation service, compared with the traditional recommender system, the interpretable recommender system will give the recommendation reasons and results at the same time. In this paper, an interpretable recommendation model based on XGBoost tree is proposed to obtain comprehensible and effective cross features from side information. The results are input into the embedded model based on attention mechanism to capture the invisible interaction among user IDs, item IDs and cross features. The captured interactions are used to predict the match score between the user and the recommended item. Cross-feature attention score is used to generate different recommendation reasons for different user-items.Experimental results show that the proposed algorithm can guarantee the quality of recommendation. The transparency and readability of the recommendation process has been improved by providing reference reasons. This method can help users better understand the recommendation behavior of the system and has certain enlightenment to help the recommender system become more personalized and intelligent. [ABSTRACT FROM AUTHOR]

The sustainable economic learning course recommendation can quickly find the knowledge information that the user really needs from the massive information space and realize the personalized recommendation to the user. However, the occurrence of trust attacks seriously affects the normal recommendation function of the recommendation system, resulting in its failure to provide users with reliable and reliable recommendation results. In order to solve the vulnerability of the recommendation system to the support attack, based on text vector model and support vector machine, this paper makes a comprehensive analysis of the current research status of the robust recommendation technology. Moreover, based on the idea of suspicious user metrics, this paper has conducts in-depth research on how to design highly robust recommendation algorithms, and constructs a highly reliable sustainable economic learning course recommendation model. In addition to this, this research tests the performance of the system from two perspectives of course recommendation satisfaction and system retrieval accuracy. The experiment proves that the model constructed in this paper performs well in the recommendation of sustainable economic learning courses. [ABSTRACT FROM AUTHOR]

User rating information on multiple predefined aspects gathered by hotel recommendation systems generally shows a deviation between the overall rating and detailed criteria ratings. In this study, to address this deviation, we proposed a novel hotel recommendation method that clusters users with different preferences into different groups using the K-means algorithm. Moreover, we allocated weights to different criteria and obtained a comprehensive score. A case study on actual data from Tripadvisor.com showed that compared with three other models, our proposed model demonstrated a more impressive performance. This research can offer advantages to hotel service providers and customers in terms of decision making. [ABSTRACT FROM AUTHOR]

Recommender systems can recommend products by analyzing the interests and habits of users. To make more efficient recommendation, the contextual information should be collected in recommendation algorithms. In the restaurant recommendation, the location and the current time of customers should also be considered to facilitate restaurants to find potential customers and give accurate and timely recommendations. However, the existing recommendation approaches often lack the consideration of the influence of time and location. Besides, the data sparsity is an inherent problem in the collaborative filtering algorithm. To address these problems, this paper proposes a recommendation approach which combines the contextual information including time, price and location. Instead of constructing the user-restaurant scoring matrix, the proposed approach clusters price tags and generates the user-price scoring matrix to alleviate the sparsity of data. The experiment on Foursquare dataset shows that the proposed approach has a better performance than traditional ones. [ABSTRACT FROM AUTHOR]

Recommender systems have recently become a significant part of e-commerce applications. Through the different types of recommender systems, collaborative filtering is the most popular and successful recommender system for providing recommendations. Recent studies have shown that using multi-criteria ratings helps the system to know the customers better. However, bringing multi aspects to collaborative filtering causes new challenges such as scalability and sparsity. Additionally, revealing the relation between criteria is yet another optimization problem. Hence, increasing the accuracy in prediction is a challenge. In this paper, an aggregation-function based multi-criteria collaborative filtering system using Rough Sets Theory is proposed as a novel approach. Rough Sets Theory is used to uncover the relationship between the overall criterion and the individual criteria. Experimental results show that the proposed model (RoughMCCF) successfully improves the predictive accuracy without compromising on online performance. [ABSTRACT FROM AUTHOR]

Recommendation systems provide reliable and relevant recommendations to users and also enable users' trust on the website. This is achieved by the opinions derived from reviews, feedbacks and preferences provided by the users when the product is purchased or viewed through social networks. This integrates interactions of social networks with recommendation systems which results in the behavior of users and user's friends. The techniques used so far for recommendation systems are traditional, based on collaborative filtering and content based filtering. This paper provides a novel approach called User-Opinion-Rating (UOR) for building recommendation systems by taking user generated opinions over social networks as a dimension. Two tripartite graphs namely User-Item-Rating and User-Item-Opinion are constructed based on users' opinion on items along with their ratings. Proposed approach quantifies the opinions of users and results obtained reveal the feasibility. [ABSTRACT FROM AUTHOR]

The recent privacy incidents reported in major media about global social networks raised real public concerns about centralized architectures. P2P social networks constitute an interesting paradigm to give back users control over their data and relations. While basic social network functionalities such as commenting, following, sharing, and publishing content are widely available, more advanced features related to information retrieval and recommendation are still challenging. This is due to the absence of a central server that has a complete view of the network. In this paper, we propose a new recommender system called P2PCF. We use collaborative filtering approach to recommend content in P2P social networks. P2PCF enables privacy preserving and tackles the cold start problem for both users and content. Our proposed approach assumes that the rating matrix is distributed within peers, in such a way that each peer only sees interactions made by her friends on her timeline. Recommendations are then computed locally within each peer before they are sent back to the requester. Our evaluations prove the effectiveness of our proposal compared to a centralized scheme in terms of recall and coverage. [ABSTRACT FROM AUTHOR]

A recommendation system is based on the user and the items, providing appropriate items to the user and effectively helping the user to find items that may be of interest. The most commonly used recommendation method is collaborative filtering. However, in this case, the recommendation system will be injected with false data to create false ratings to push or nuke specific items. This will affect the user's trust in the recommendation system. After all, it is important that the recommendation system provides a trusted recommendation item. Therefore, there are many algorithms for detecting attacks. In this article, it proposes a method to detect attacks based on the beta distribution. Different researchers in the past assumed that the attacker only attacked one target item in the user data. This research simulated an attacker attacking multiple target items in the experiment. The result showed a detection rate of more than 80%, and the false rate was within 16%. [ABSTRACT FROM AUTHOR]

In recent years, with the development of advanced mobile applications, people's various daily behavior data, such as geographic location, social information, hobbies, are more easily collected. To process these data, data cross-boundary fusion has become a key technology, and there are some challenges, such as solving the problems of the cross-boundary business integrity, cross-boundary value complementarity and so on. Mobile Services Recommendation requires improved recommendation accuracy. User trust is an effective measure of information similarity between users. Using trust can effectively improve the accuracy of recommendations. The existing methods have low utilization of general trust data, sparseness of trust data, and lack of user trust characteristics. Therefore, a method needs to be proposed to make up for the shortcomings of explicit trust relationships and improve the accuracy of user interest feature completion. In this paper, a recommendation model is proposed to mine the implicit trust relationships from user data and integrate the explicit social information of users. First, the rating prediction model was improved using the traditional Singular Value Decomposition (SVD) model, and the implicit trust relationships were mined from the user's historical data. Then, they were fused with the explicit social trust relationships to obtain a crossover data fusion model. We tested the model using three different orders of magnitude. We compared the user preference prediction accuracies of two models: one that does not integrate social information and one that integrates social information. The results show that our model improves the user preference prediction accuracy and has higher accuracy for cold start users. On the three data sets, the average error is reduced by 2.29%, 5.44% and 4.42%, suggesting that it is an effective data crossover fusion technology. [ABSTRACT FROM AUTHOR]

With the pervasiveness of GPS-enabled devices, a considerable number of GPS traces are accumulating continuously and unobtrusively in online communities. However, almost all current applications directly use raw GPS data, such as coordinates and time stamps, without interpreting these data. Thus far, online communities cannot offer much support to users in terms of recommending geospatial locations. Furthermore, because the data sets involved are large, users cannot browse each GPS trajectory individually. Therefore, users' GPS trajectories must be mined and then classified as positive or negative. When the number of ratings for a place exceeds a certain threshold, the place is considered suitable for the user. By contrast, when the ratings for a place are mostly negative, this place is considered unsuitable for the user. When a user searches for the best place, the recommender system determines the user's location (latitude, longitude) and then sends the best-rated destinations and the shortest routes between the user's location and the destination to the user's mobile device. Experiments were conducted in this study to determine the requisite similarity for GPS data points, the user's information, and the best route for the user. [ABSTRACT FROM AUTHOR]

Although spatial and temporal information has often been considered to improve recommendation performances, existing multi-criteria recommender systems often neglect to leverage spatial and temporal information. Also, it is a non-trivial task to simultaneously apply such information to recommendation services since the factors have interrelations to each other. In this paper, we propose a multi-criteria tensor model combining spatial and temporal information. The auxiliary information is categorized by several features and applied into the model. In particular, the spatial information of users' countries is grouped into seven continents to reduce response times for learning the model. The single model enables to us keep the inherent structure of and the interrelations between multi-criteria and spatial/temporal information. To predict user preferences, tensor factorization based on Higher Order Singular Value Decomposition is exploited. Experimental results with a TripAdvisor dataset show that the proposed method outperforms other baseline methods based on a 2-dimensional rating matrix, tensor model, and other multi-criteria recommendation, in terms of RMSE and MAE. Furthermore, several experiments reveal the influences of the individual factors (i.e., multi-criteria, spatial and temporal information) and their consolidations, on restaurant recommendation. A comparative analysis of the multi-criteria elements shows that their influences relate to their correlations. [ABSTRACT FROM AUTHOR]

Customer-based practices enable benefits to organizations in a contentious business. Offering individualized proposals increase customer loyalty to be able to afloat. Understanding customers is a vital difficulty to perform personalized recommendations. As a demographic feature, gender information essentially cannot be captured by human tracking technologies. Hence, several procedures are improved to predict undiscovered gender information. In the research, the followed indoor paths in a shopping mall are used to predict customer genders using fuzzy c-medoids, one of the soft clustering techniques. A Levenshtein-based fuzzy classification methodology is proposed the followed paths as string data. Although some studies focused on gender prediction, no research has centered on path-oriented. The novelty of the investigation is to analyze customer path data for the gender classes. [ABSTRACT FROM AUTHOR]

The main goal of a recommendation system is to recommend items of interest to users by analyzing their historical data. Content-based and collaborative filtering are the traditional recommendation strategies, each with its own strengths and weaknesses. Some of their weaknesses can be overcome by combining the two strategies. The resulting hybrid system performs qualitatively better than the traditional recommendation systems. However, historical data of some users may consist largely of only likes or only dislikes. Those users are termed as optimistic or pessimistic users respectively. On an average there are around 10 to 20% of pessimistic users present in a given dataset. For pessimistic users, whose profiles have mostly dislikes and very few likes, content-based filtering can hardly recommend any items of interest. In content-based filtering technique pessimistic users get poor recommendations of either uninteresting movies or no recommendations at all. This can be alleviated by boosting the content profiles of pessimistic users using the top- n recommendations of collaborative filtering. This content boosted hybrid filtering system provides a novel list of recommendations even for pessimistic users, with predictive accuracy better than that of a traditional content-based filtering system. [ABSTRACT FROM AUTHOR]

In this research, we combine relational learning with multi-domain to develop a formal framework for a recommendation system. The design of our framework aims at: (i) constructing general rules for recommendations, (ii) providing suggested items with clear and understandable explanations, (iii) delivering a broad range of recommendations including novel and unexpected items. We use relational learning to find all possible relations, including novel relations, and to form the general rules for recommendations. Each rule is represented in relational logic, a formal language, associating with probability. The rules are used to suggest the items, in any domain, to the user whose preferences or other properties satisfy the conditions of the rule. The information described by the rule serves as an explanation for the suggested item. It states clearly why the items are chosen for the users. The explanation is in if-then logical format which is unambiguous, less redundant and more concise compared to a natural language used in other explanation recommendation systems. The explanation itself can help persuade the user to try out the suggested items, and the associated probability can drive the user to make a decision easier and faster with more confidence. Incorporating information or knowledge from multiple domains allows us to broaden our search space and provides us with more opportunities to discover items which are previously unseen or surprised to a user resulting in a wide range of recommendations. The experiment results show that our proposed algorithm is very promising. Although the quality of recommendations provided by our framework is moderate, our framework does produce interesting recommendations not found in the primitive single-domain based system and with simple and understandable explanations. [ABSTRACT FROM AUTHOR]

Reviews text reflects user interests and product characteristics, which provide rich useful semantic textual information for modeling user and product. Some existing works improve the performance of rating prediction by distinguishing the usefulness of each review. However, most of them ignore the fact that the usefulness of each review should be dynamic and dependent on the target user-product pair. For example, when we predict the user's ratings of a restaurant, the user's previous reviews about restaurants should be more useful. To be more specific, when we model the target user-product pair, the usefulness of each review is associated with the target user and item. To address the above issue, we use a Review-level Dynamic Topic Co-Attention, which combines all reviews written by the user and all reviews that were written for the product to assign attention score for each review collaboratively. Besides, we believe that product category and user co-purchase information can further improve the rating prediction performance. In this paper, we propose a neural joint model called NMRP based on reviews, products category, and users' co-purchase information for rating prediction recommendation. First, the Review Extraction Module learns user and product information from reviews. Then, HIN Extraction Module extracts the association feature for a given target user-product pair from a heterogeneous information network (HIN). Finally, the two parts of data are connected and input into the feature interaction method Attentional Factorization Machines (AFM), to achieve rating prediction. The experimental study on three datasets from Amazon shows that our model outperforms recently proposed baselines such as DeepCoNN, TransNets, and NARRE. [ABSTRACT FROM AUTHOR]