The proliferation of multimedia content demands advanced video recommendation systems capable of capturing complex multimodal, heterogeneity, cold-start scenarios and dynamic user-item interactions. In this paper, we introduce a framework that integrates graph-based modeling, multimodal representation learning, and sequential modeling to address these challenges. Our approach begins with the transformation of a Heterogeneous Information Network (HIN) into a homogeneous video graph, preserving high-order semantic relationships through meta-path-guided transformations. We integrate multimodal features aligned into a shared embedding space via geometric encoding. To capture the temporal dynamics of user interactions, gated recurrent units are employed, providing a nuanced understanding of evolving preferences. The framework is optimized with a multi-objective loss to balance ranking accuracy and structural regularization. Experiments on two datasets demonstrate our framework superiority over state-of-the-art approaches. Additionally, our approach demonstrates exceptional resilience in addressing cold-start scenarios and provides explainable recommendations by revealing semantic connections between user history and suggested content.

The promotion of e-commerce platforms has changed the lifestyle of several people from traditional marketing to digital marketing where businesses are made online and the concurrence reached high levels. These platforms have helped the ease of purchases while providing more advantages to the customers such as benefiting from a wide range of high-quality products, low prices, buying at any time, and more importantly supplying information and reviews about the products, and so on. Unfortunately, a plethora of companies mislead the customers to buy their products or demote the competitors' by using deceptive opinion spams which has a negative impact on the decision and the behavior of the purchasers. Deceptive opinion spams are written deliberately to seem legitimate and authentic so that to misguide or delude the customer's purchases. Consequently, the detection of these opinions is a hard task due to their nature for both humans and machines. Most of the studies are based on traditional machine learning and sparse feature engineering. However, these models do not capture the semantic aspect of reviews. According to many researchers, it is the key to the detection of deceptive opinion spam. Besides, only a few studies consider using contextual information by adopting neural networks in comparison with plenty of traditional machine learning classifiers. These models face numerous shortcomings as long as their representations are obtained while mining each review considering only words, sentences, reviews, or a combination of them, thereby classifying them based on their representations. In fact, deceptive opinions are written by the same deceivers belonging to the same companies with similar aims to promote or demolish a product. In other words, Deceptive opinion spams tend to be semantically coherent with each other. To the best of our knowledge, no model tries to obtain a representation based on the contextual relationships between opinions. This article proposes to use a capsule neural network, bidirectional long short-term memory, attention mechanism, and paragraph vector distributed bag of words to detect deceptive opinion spam. Our model provides a powerful representation of the opinions since it centers on the preservation of their contexts and the relationships between them. The results show that our model significantly outperforms the existing state-of-the-art models.

Real-world datasets often comprise multiple views or representations, such as user profiles, comments, and preferences. Integrating information from these multi-view datasets can significantly enhance recommendation performance. Consequently, several multi-view clustering (MVC) approaches have been proposed and extensively applied to exploit the compatibility of diverse data sources. Among these methods, Non-negative Matrix Factorization (NMF) has garnered significant attention due to its ability to process high-dimensional data. However, most NMF-based MVC methods suffer from two critical drawbacks: the similarity matrices constructed using traditional methods fail to fully exploit neighborhood information among data points, and they are highly sensitive to noise, missing values, and outliers. Additionally, existing NMF-based clustering methods often do not generate sparse coefficient matrices, leading to suboptimal data representation based on the basis matrix. To address these challenges and enhance the accuracy and efficiency of recommendation algorithms, we propose a novel approach called Multi-View Clustering based on Sparse Non-Negative Matrix Factorization (MVC-SNMF). This method leverages the manifold structure of multi-view data and incorporates a sparsity constraint to effectively handle noise while preserving the geometric structure. Furthermore, we introduce a sparsity constraint on the rows of the coefficient matrix to enhance the visibility of clustering structures. To efficiently solve the proposed non-convex optimization problem, we devise an iterative update scheme. Our experimental results on real datasets from Last.fm and Yelp demonstrate the effectiveness of our approach in improving clustering performance, surpassing most existing clustering algorithms. The proposed method not only enhances the clustering accuracy but also exhibits robustness to noise and missing values, making it a promising solution for practical recommendation systems.

Abstract In this article, we address a critical concern of the growth of computer servers number and the resulting power expenditures in data centers by analyzing the statistical metrics related to the workload executed in each physical node. The aim is to build a stochastic model for power consumption estimation based on historical data. Relying on in‐depth investigation and experimental testing of the power consumption features and performance of the various workload datasets, we propose a model that considers the workload and the power consumed to be executed by a server as random variables. Based on the properties of the probabilistic distribution function of each random variable, we establish the correlation relationship between the workload and the power consumption using a non‐parametric approach. Our use of a non‐parametric method to learn a given probability model is challenging because it requires estimating the full distribution from the available data samples. The accuracy of our approach is demonstrated by estimating the energy consumption of various workloads. The experimental and simulation results show that our model outperforms many existing approaches in terms of accuracy, and it can be applied to a wide variety of workloads.