Sports Data Management, Mining, and Visualization

Most of the extant studies in social recommender system are based on explicit social relationships, while the potential of implicit relationships in the heterogeneous social networks remains largely unexplored. This study proposes a new approach to designing a recommender system by employing grey relational analysis on the heterogeneous social networks. It starts with the establishment of heterogeneous social networks through the user-item bipartite graph, user social network graph and user-attribute bipartite graph; and then uses grey relational analysis to identify implicit social relationships, which are then incorporated into the matrix factorization model. Five experiments were conducted to test the performance of our approach against four state-of-the-art baseline methods. The results show that compared with the baseline methods, our approach can effectively alleviate the sparsity problem, because the heterogeneous social network provides richer information. In addition, the grey relational analysis method has the advantage of low requirements for data size and efficiently relieves the cold start problem. Furthermore, our approach saves processing time, thus increases recommendation efficiency. Overall, the proposed approach can effectively improve the accuracy of rating prediction in social recommendations and provide accurate and efficient recommendation service for users.

Harnessing heterogeneous social networks for better recommendations: A grey relational analysis approach

Mining for Fake News

In food diet communication domain, images convey important information to capture users' attention beyond the traditional ingredient content, making it crucial to influence user-decision about the relevancy of a given diet. By using a deep learning-based image clustering method, this paper proposes an Explainable Food Recommendation system that uses the visual content of food to justify their recommendations. n the recommendation system. Especially, a new similarity score based on a tendency measure that quantifies the extent to which user community prefers a given food category is introduced and incorporated in the recommendation. Finally, a rule-based explainability is introduced to enhance transparency and interpretability of the recommendation outcome. Our experiments on a crawled dataset showed that the proposed method enhances recommendation quality in terms of precision, recall, F1, and Normalized Discounted Cumulative Gain (NDCG) by 7.35%, 6.70%, 7.32% and 14.38%, respectively, when compared to other existing methodologies for food recommendation. Besides ablation study is performed to demonstrate the technical soundness of the various components of our recommendation system. 

An effective explainable food recommendation using deep image clustering and community detection

Massive Open Online Course (MOOC) has been criticized for low completion rates, and one of the major reasons is that it fails to offer personalized course recommendations for different users with different demands. To solve this problem, this paper proposes a personalized course recommendation model based on convolutional neural network combined with negative sequence pattern mining. The model first models the course-learning sequence as a negative sequence pattern according to the user’s course registration, degree of completion, and final grades, in which, the negative term means that students should not choose and misoperate the principle of courses. Then, it employs a convolutional neural network structure to extract the internal features of negative sequence patterns for representation learning. Finally, through the convolutional sequence-embedding model, each user is recommended with a course list that includes the user’s maximized needs in recent temporal terms and the courses that are easy to be misselected. Experiment results show that the recommended model achieves higher recommendation performance with lower course dropout rate compared to baselines, which provides a new insight for both online and offline course recommendation.

Online Course Recommendation Using Deep Convolutional Neural Network with Negative Sequence Mining

In recent years, the ever-increasing number of cloud services has led to the service targeting issue. It becomes challenging to provide users with services that fit their needs. Recommender systems address the service targeting problem by helping users to easily retrieve services matching their tastes. Collaborative filtering-based methods are spreadly used recommendation approaches since numerous methods are based on them. However, most of them assume the stationarity of user profiles over time, which is an unrealistic consideration. Indeed, in practice, users’ tastes are dynamic and therefore show a significant variability over time. Time-aware recommender systems effectively integrate the variability of users’ needs to improve the recommendation quality. However, the recommendation credibility is another crucial requirement that is not met by those time-aware recommender systems that moreover badly deal with the data sparsity issue. To address the service targeting problem while meeting requirements on the recommendation reliability and the users’ tastes variability, we propose in this paper a realistic temporal service recommendation approach based on implicit trust relationships inference. Our method integrates the time feature into the recommendation process in order to consider the instantaneity of users’ needs. Our proposal is based on a trusted network that infers implicit trust relationships to ensure the recommendation credibility even in case of data sparsity. Experiments are conducted on real-world service invocations datasets. Our proposal is then compared to existing methods and displays valuable performances in terms of mean absolute error, root mean square error, and normalized discounted cumulative.

A time-aware service recommendation based on implicit trust relationships and enhanced user similarities

Being one of the most solicited content (opinions) sharing platforms, Twitter is a granary of information serving as a base for our service discovery/exposure approach proposed in this paper. The growth of data caused by the internet growth has led to the birth of a growing number of services, making the task difficult for telecommunication operators in competition. In this paper, we propose a dual service discovery/exposure approach to reduce the gap between offered services and subscribers’ needs in an IMS context. This approach is based on opinion mining related to Twitter trending topics in order to estimate the sensitivity of the target user to a service or another. Compared to both the classic approach and the collaborative service discovery/exposure approach, our results show an improvement in the accuracy and error of the service targeting at the target user’s starting phase on the operator’s network.

An IP Multimedia Subsystem Service Discovery and Exposure Approach Based on Opinion Mining by Exploiting Twitter Trending Topics

The increasing number of services gives the telecommunication operators a great opportunity to fulfil the demand of their subscribers. Indeed, since various practically comparative services are promptly accessible, to offer a service that is relevant to the subscriber is at the centre of the operator's concerns. This is the case of IMS (IP Multimedia Subsystem), which is this platform that effectively allows access to a range of varied services regardless of the underlying access networks. By this way, to offer services fitting the needs of the subscriber within the IMS remains a challenging issue. The service discovery approach by exploiting only the information associated with the user's profile has poor performance when the usage history of the target subscriber is not provided. This restricts the relevance of the proposed service. Therefore, in addition to the information associated with the target subscriber's profile, we are interested in the information associated with the profiles of other subscribers similar to that of the target subscriber to improve the accuracy of the service targeting. In this paper we present a collaborative approach applied to service discovery that is based on taking into account the degree of satisfaction of the user. By the end we demonstrate through our results a clear improvement in service targeting accuracy.

IMS Predictive Collaborative Service Discovery Approach

The rise of high-quality cloud services has made service recommendation a crucial research question. Quality of Service (QoS) is widely adopted to characterize the performance of services invoked by users. For this purpose, the QoS prediction of services constitutes a decisive tool to allow end-users to optimally choose high-quality cloud services aligned with their needs. The fact is that users only consume a few of the broad range of existing services. Thereby, perform a high-accurate service recommendation becomes a challenging task. To tackle the aforementioned challenges, we propose a data sparsity resilient service recommendation approach that aims to predict relevant services in a sustainable manner for end-users. Indeed, our method performs both a QoS prediction of the current time interval using a flexible matrix factorization technique and a QoS prediction of the future time interval using a time series forecasting method based on an AutoRegressive Integrated Moving Average (ARIMA) model. The service recommendation in our approach is based on a couple of criteria ensuring in a lasting way, the appropriateness of the services returned to the active user. The experiments are conducted on a real-world dataset and demonstrate the effectiveness of our method compared to the competing recommendation methods.

/pdf/service-recommendation-driven-by-a-matrix-factorization-4hojayujco.pdf

Armielle Noulapeu Ngaffo

Papers

A time-aware service recommendation based on implicit trust relationships and enhanced user similarities

An IP Multimedia Subsystem Service Discovery and Exposure Approach Based on Opinion Mining by Exploiting Twitter Trending Topics

IMS Predictive Collaborative Service Discovery Approach

Service recommendation driven by a matrix factorization model and time series forecasting