Search results

In this “Info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activity and e-commerce. The purpose of this paper is to model users' preference evolution to recommend potential items which users may be interested in.

A novel recommendation system, namely evolution-learning recommendation (ELR), is developed to precisely predict user interest for making recommendations. Differing from prior related methods, the authors integrate the matrix factorization (MF) and recurrent neural network (RNN) to effectively describe the variation of user preferences over time.

A novel cumulative factorization technique is proposed to efficiently decompose a rating matrix for discovering latent user preferences. Compared to traditional MF-based methods, the cumulative MF could reduce the utilization of computation resources. Furthermore, the authors depict the significance of long- and short-term effects in the memory cell of RNN for evolution patterns. With the context awareness, a learning model, V-LSTM, is developed to dynamically capture the evolution pattern of user interests. By using a well-trained learning model, the authors predict future user preferences and recommend related items.

Based on the relations among users and items for recommendation, the authors introduce a novel concept, virtual communication, to effectively learn and estimate the correlation among users and items. By incorporating the discovered latent features of users and items in an evolved manner, the proposed ELR model could promote “right” things to “right” users at the “right” time. In addition, several extensive experiments are performed on real datasets and are discussed. Empirical results show that ELR significantly outperforms the prior recommendation models. The proposed ELR exhibits great generalization and robustness in real datasets, including e-commerce, industrial retail and streaming service, with all discussed metrics.

Textual information about restaurants, such as online reviews and food categories, is essential for consumer purchase decisions. However, previous restaurant recommendation studies have failed to use textual information containing essential information for predicting consumer preferences effectively. This study aims to propose a novel restaurant recommendation model to effectively estimate the assessment behaviors of consumers for multiple restaurant attributes.

The authors collected 1,206,587 reviews from 25,369 consumers of 46,613 restaurants from Yelp.com. Using these data, the authors generated a consumer preference vector by combining consumer identity and online consumer reviews. Thereafter, the authors combined the restaurant identity and food categories to generate a restaurant information vector. Finally, the nonlinear interaction between the consumer preference and restaurant information vectors was learned by considering the restaurant attribute vector.

This study found that the proposed recommendation model exhibited excellent performance compared with state-of-the-art models, suggesting that combining various textual information on consumers and restaurants is a fundamental factor in determining consumer preference predictions.

To the best of the authors’ knowledge, this is the first study to develop a personalized restaurant recommendation model using textual information from real-world online restaurant platforms. This study also presents deep learning mechanisms that outperform the recommendation performance of state-of-the-art models. The results of this study can reduce the cost of exploring consumers and support effective purchasing decisions.

关于餐厅的文本信息, 如在线评论和食品分类, 对于消费者的购买决策产生至关重要。然而, 先前的餐厅推荐研究未能有效利这些文本信息去预测消费者喜好。本研究提出了一种新颖的餐厅推荐模型, 以有效估计消费者对多个餐厅属性的评估行为。

我们从 Yelp.com 收集了来自25,369名消费者对 46,613 家餐厅的 1,206,587 条评论。利用这些数据, 我们通过结合消费者身份和在线消费者评论生成了消费者偏好向量。然后, 我们结合了餐厅身份和食品分类来生成餐厅信息向量。最后, 考虑到餐厅属性向量, 本研究调查了消费者偏好和餐厅信息向量之间的非线性交互关系。

我们发现, 所提出的推荐模型相比于之前最先进的模型表现出更优秀的性能, 这表明结合消费者和餐厅的各种文本信息是预测消费者喜好的基本因素。

据我们所知, 这是第一项利用来自真实在线餐厅平台的文本信息开发个性化餐厅推荐模型的研究。本研究还提出了胜过最先进模型的深度学习机制。本研究的结果可以降低探索消费者行为的成本并支持有效的购买决策。

Improving the diversity of recommendation information has become one of the latest research hotspots to solve information cocoons. Aiming to achieve both high accuracy and diversity of recommender system, a hybrid method has been proposed in this paper. This study aims to discuss the aforementioned method.

This paper integrates latent Dirichlet allocation (LDA) model and locality-sensitive hashing (LSH) algorithm to design topic recommendation system. To measure the effectiveness of the method, this paper builds three-level categories of journal paper abstracts on the Web of Science platform as experimental data.

(1) The results illustrate that the diversity of recommended items has been significantly enhanced by leveraging hashing function to overcome information cocoons. (2) Integrating topic model and hashing algorithm, the diversity of recommender systems could be achieved without losing the accuracy of recommender systems in a certain degree of refined topic levels.

The hybrid recommendation algorithm developed in this paper can overcome the dilemma of high accuracy and low diversity. The method could ameliorate the recommendation in business and service industries to address the problems of information overload and information cocoons.

Given the critical role of user-generated content (UGC) in e-commerce, exploring various aspects of UGC can aid in understanding user purchase intention and commodity recommendation. Therefore, this study investigates the impact of UGC on purchase decisions and proposes new recommendation models based on sentiment analysis, which are verified in Douban, one of the most popular UGC websites in China.

After verifying the relationship between various factors and product sales, this study proposes two models, collaborative filtering recommendation model based on sentiment (SCF) and hidden factors topics recommendation model based on sentiment (SHFT), by combining traditional collaborative filtering model (CF) and hidden factors topics model (HFT) with sentiment analysis.

The results indicate that sentiment significantly influences purchase intention. Furthermore, the proposed sentiment-based recommendation models outperform traditional CF and HFT in terms of mean absolute error (MAE) and root mean square error (RMSE). Moreover, the two models yield different outcomes for various product categories, providing actionable insights for organizers to implement more precise recommendation strategies.

The findings of this study advocate the incorporation of UGC sentimental factors into websites to heighten recommendation accuracy. Additionally, different recommendation strategies can be employed for different products types.

This study introduces a novel perspective to the recommendation algorithm field. It not only validates the impact of UGC sentiment on purchase intention but also evaluates the proposed models with real-world data. The study provides valuable insights for managerial decision-making aimed at enhancing recommendation systems.

Traditional library book recommendation methods are mainly based on association rules and user profiles. They may help to learn about students' interest in different types of books, e.g., students majoring in science and engineering tend to pay more attention to computer books. Nevertheless, most of them still need to identify users' interests accurately. To solve the problem, the authors propose a novel embedding-driven model called InFo, which refers to users' intrinsic interests and academic preferences to provide personalized library book recommendations.

The authors analyze the characteristics and challenges in real library book recommendations and then propose a method considering feature interactions. Specifically, the authors leverage the attention unit to extract students' preferences for different categories of books from their borrowing history, after which we feed the unit into the Factorization Machine with other context-aware features to learn students' hybrid interests. The authors employ a convolution neural network to extract high-order correlations among feature maps which are obtained by the outer product between feature embeddings.

The authors evaluate the model by conducting experiments on a real-world dataset in one university. The results show that the model outperforms other state-of-the-art methods in terms of two metrics called Recall and NDCG.

It requires a specific data size to prevent overfitting during model training, and the proposed method may face the user/item cold-start challenge.

The embedding-driven book recommendation model could be applied in real libraries to provide valuable recommendations based on readers' preferences.

The proposed method is a practical embedding-driven model that accurately captures diverse user preferences.

The purpose of this study is to investigate how book group recommendations can be used as a meaningful way to suggest suitable books to users, given the increasing number of individuals engaging in sharing and discussing books on the web.

The authors propose reviews fine-grained classification (CFGC) and its related models such as CFGC1 for book group recommendation. These models can categorize reviews successively by function and role. Constructing the BERT-BiLSTM model to classify the reviews by function. The frequency characteristics of the reviews are mined by word frequency analysis, and the relationship between reviews and total book score is mined by correlation analysis. Then, the reviews are classified into three roles: celebrity, general and passerby. Finally, the authors can form user groups, mine group features and combine group features with book fine-grained ratings to make book group recommendations.

Overall, the best recommendations are made by Synopsis comments, with the accuracy, recall, F-value and Hellinger distance of 52.9%, 60.0%, 56.3% and 0.163, respectively. The F1 index of the recommendations based on the author and the writing comments is improved by 2.5% and 0.4%, respectively, compared to the Synopsis comments.

Previous studies on book recommendation often recommend relevant books for users by mining the similarity between books, so the set of book recommendations recommended to users, especially to groups, always focuses on the few types. The proposed method can effectively ensure the diversity of recommendations by mining users’ tendency to different review attributes of books and recommending books for the groups. In addition, this study also investigates which types of reviews should be used to make book recommendations when targeting groups with specific tendencies.

Session-based recommendation aims to predict the user's next preference based on the user's recent activities. Although most existing studies consider the global characteristics of items, they only learn the global characteristics of items based on a single connection relationship, which cannot fully capture the complex transformation relationship between items. We believe that multiple relationships between items in learning sessions can improve the performance of session recommendation tasks and the scalability of recommendation models. At the same time, high-quality global features of the item help to explore the potential common preferences of users.

This work proposes a session-based recommendation method with a multi-relation global context–enhanced network to capture this global transition relationship. Specifically, we construct a multi-relation global item graph based on a group of sessions, use a graded attention mechanism to learn different types of connection relations independently and obtain the global feature of the item according to the multi-relation weight.

We did related experiments on three benchmark datasets. The experimental results show that our proposed model is superior to the existing state-of-the-art methods, which verifies the effectiveness of our model.

First, we construct a multi-relation global item graph to learn the complex transition relations of the global context of the item and effectively mine the potential association of items between different sessions. Second, our model effectively improves the scalability of the model by obtaining high-quality item global features and enables some previously unconsidered items to make it onto the candidate list.

This article proposes an architecture of an intelligent and autonomous recommendation system to be applied to any virtual learning environment, with the objective of efficiently recommending digital resources. The paper presents the architectural details of the intelligent and autonomous dimensions of the recommendation system. The paper describes a hybrid recommendation model that orchestrates and manages the available information and the specific recommendation needs, in order to determine the recommendation algorithms to be used. The hybrid model allows the integration of the approaches based on collaborative filter, content or knowledge. In the architecture, information is extracted from four sources: the context, the students, the course and the digital resources, identifying variables, such as individual learning styles, socioeconomic information, connection characteristics, location, etc. Tests were carried out for the creation of an academic course, in order to analyse the intelligent and autonomous capabilities of the architecture.

The explosion of data due to the sophistication of information and communication technology makes it simple for prospective tourists to learn about previous hotel guests' experiences. They prioritize the rating score when selecting a hotel. However, rating scores are less reliable for suggesting a personalized preference for each aspect, especially when they are in a limited number. This study aims to recommend ratings and personalized preference hotels using cross-domain and aspect-based features.

We propose an aspect-based cross-domain personalized recommendation (AsCDPR), a novel framework for rating prediction and personalized customer preference recommendations. We incorporate a cross-domain personalized approach and aspect-based features of items from the review text. We extracted aspect-based feature vectors from two domains using bidirectional long short-term memory and then mapped them by a multilayer perceptron (MLP). The cross-domain recommendation module trains MLP to analyze sentiment and predict item ratings and the polarities of the aspect based on user preferences.

Expanded by its synonyms, aspect-based features significantly improve the performance of sentiment analysis on accuracy and the F1-score matrix. With relatively low mean absolute error and root mean square error values, AsCDPR outperforms matrix factorization, collaborative matrix factorization, EMCDPR and Personalized transfer of user preferences for cross-domain recommendation. These values are 1.3657 and 1.6682, respectively.

This study assists users in recommending hotels based on their priority preferences. Users do not need to read other people's reviews to capture the key aspects of items. This model could enhance system reliability in the hospitality industry by providing personalized recommendations.

This study introduces a new approach that embeds aspect-based features of items in a cross-domain personalized recommendation. AsCDPR predicts ratings and provides recommendations based on priority aspects of each user's preferences.

The purpose of this study is to extensively explore the vehicular network paradigm, challenges faced by them and provide a reasonable solution for securing these vulnerable networks. Vehicle-to-everything (V2X) communication has brought the long-anticipated goal of safe, convenient and sustainable transportation closer to reality. The connected vehicle (CV) paradigm is critical to the intelligent transportation systems vision. It imagines a society free of a troublesome transportation system burdened by gridlock, fatal accidents and a polluted environment. The authors cannot overstate the importance of CVs in solving long-standing mobility issues and making travel safer and more convenient. It is high time to explore vehicular networks in detail to suggest solutions to the challenges encountered by these highly dynamic networks.

This paper compiles research on various V2X topics, from a comprehensive overview of V2X networks to their unique characteristics and challenges. In doing so, the authors identify multiple issues encountered by V2X communication networks due to their open communication nature and high mobility, especially from a security perspective. Thus, this paper proposes a trust-based model to secure vehicular networks. The proposed approach uses the communicating nodes’ behavior to establish trustworthy relationships. The proposed model only allows trusted nodes to communicate among themselves while isolating malicious nodes to achieve secure communication.

Despite the benefits offered by V2X networks, they have associated challenges. As the number of CVs on the roads increase, so does the attack surface. Connected cars provide numerous safety-critical applications that, if compromised, can result in fatal consequences. While cryptographic mechanisms effectively prevent external attacks, various studies propose trust-based models to complement cryptographic solutions for dealing with internal attacks. While numerous trust-based models have been proposed, there is room for improvement in malicious node detection and complexity. Optimizing the number of nodes considered in trust calculation can reduce the complexity of state-of-the-art solutions. The theoretical analysis of the proposed model exhibits an improvement in trust calculation, better malicious node detection and fewer computations.

The proposed model is the first to add another dimension to trust calculation by incorporating opinions about recommender nodes. The added dimension improves the trust calculation resulting in better performance in thwarting attacks and enhancing security while also reducing the trust calculation complexity.