Quality of Experience (QoE) modeling is crucial for optimizing video streaming services to capture the complex relationships between different features and user experience. We propose a novel approach to QoE modeling in video streaming applications using interpretable Machine Learning (ML) techniques over raw time series data. Unlike traditional black-box approaches, our method combines Kolmogorov-Arnold Networks (KANs) as an interpretable readout on top of compact frequency-domain features, allowing us to capture temporal information while retaining a transparent and explainable model. We evaluate our method on popular datasets and demonstrate its enhanced accuracy in QoE prediction, while offering transparency and interpretability.

In recent years, there has been a significant surge in machine learning techniques, particularly in the domain of deep learning, tailored for handling attributed graphs. Nevertheless, to work, these methods assume that the attributes values are fully known, which is not realistic in numerous real-world applications. This paper explores the potential of Optimal Transport (OT) to impute missing attributes on graphs. To proceed, we design a novel multi-view OT loss function that can encompass both node feature data and the underlying topological structure of the graph by utilizing multiple graph representations. We then utilize this novel loss to train efficiently a Graph Convolutional Neural Network (GCN) architecture capable of imputing all missing values over the graph at once. We evaluate the interest of our approach with experiments both on synthetic data and real-world graphs, including different missingness mechanisms and a wide range of missing data. These experiments demonstrate that our method is competitive with the state-of-the-art in all cases and of particular interest on weakly homophilic graphs.

Efficient and responsive traffic engineering is crucial for maintaining the robustness and reliability of Wide Area Networks (WANs). Traditional traffic engineering approaches often struggle to adapt to the dynamic and complex demands of today's network environments. To address these challenges, this paper enhances the Traffic Engineering algorithms by integrating an attention mechanism within the Edge-Path Embedding component. This significantly improves the model's adaptability and decision-making accuracy. Our comprehensive experimental evaluations demonstrate substantial improvements in terms of satisfied traffic demands and computational efficiency, highlightina the effectiveness of our approach.

Graph-based fraud detection is an important task in many real-world domains such as insurance, finance, and cybersecurity. Even if existing semi-supervised models have proven to be efficient in identifying anomalous nodes, they assume that a labeled sample of the nodes is available to train the model, without taking into account the real-world problem of the unreliability of such a sample. In practice, the labeling is often done manually and contains many errors. In this paper, we study fraud detection in attributed networks, and we propose a new framework, based on two graph auto-encoders trained following a suspicion mechanism: the first auto-encoder is trained to better reconstruct the normal nodes while the second one, the fraudulent ones. The final classification is done by coupling the result of both auto-encoders. We demonstrate that our approach obtains at least equivalent performances to state of the art methods in the case of a perfectly labeled sample while being more resilient to the introduction of mistakes in this sample.

Social networks are usually analyzed and mined without taking into account the presence of missing values. In this article, we consider dynamic networks represented by sequences of graphs that change over time and we study the robustness and the accuracy of the community detection algorithms in presence of missing edges. We assume that the network evolution can provide a complementary information allowing to neutralize the missing data. To confirm our hypothesis, we designed an experimental framework to simulate the missing data and compare the communities identified by the methods, with or without missing links. We explore two types of methods. The first ones, based on tensor decomposition, are adapted for dynamic networks. The second ones correspond to conventional community detection algorithms able to handle simple graphs. In our framework, the latter ones are adapted to dynamic graphs, either by merging the data during the preprocessing step or by merging the partitions during a post-processing step. The experimentation was conducted on synthetic and real dynamic networks for which the ground truth is available. The results confirm the best performances of the methods suited for dynamic networks when they present a complex community structure.

This paper shows how data mining and in particular graph mining and clustering can help to tackle difficult tracking problems such as tracking possibly multiple objects in a video with a moving camera and without any contextual information on the objects to track. Starting from different segmentations of the video frames (dynamic and non dynamic ones), we extract frequent sub graph patterns to create spatio-temporal patterns that may correspond to interesting objects to track. We then cluster the obtained spatio-temporal patterns to get longer and more robust tracks along the video. We compare our tracking method called TRAP to two state-of-the-art tracking ones and show on four synthetic and real videos that our method is effective in this difficult context.

Dynamic graph mining is the task of searching for subgraph patterns that capture the evolution of a dynamic graph. In this paper, we are interested in mining dynamic graphs in videos. A video can be regarded as a dynamic graph, whose evolution over t

A problem usually encountered in probabilistic automata learning is the difficulty to deal with large training samples and/or wide alphabets. This is partially due to the size of the resulting Probabilistic Prefix Tree (PPT) from which state merging-based learning algorithms are generally applied. In this paper, we propose a novel method to prune PPTs by making use of the H-divergence dH, recently introduced in the field of domain adaptation. dHis based on the classification error made by an hypothesis learned from unlabeled examples drawn according to two distributions to compare. Through a thorough comparison with state-of-the-art divergence measures, we provide experimental evidences that demonstrate the efficiency of our method based on this simple and intuitive criterion.

Computing frequent itemsets and their frequencies from large Boolean matrices (e.g., to derive association rules) has been one of the hot topics in data mining. Levelwise algorithms (e.g., the a priori algorithm) have been proved effective for frequent itemset mining from sparse data. However, in many practical applications, the computation turns out to be intractable for the user-given frequency threshold and the lack of focus leads to huge collections of frequent itemsets. In the last three years, two promising issues have been investigated: the use of user defined constraints and closed set mining. To the best of our knowledge, combining these two frameworks has not been studied yet. The authors show that the benefit of these two approaches can be combined into levelwise algorithms. An experimental validation related to the discovery of association rules with negations is reported.

Frequent association rules (e.g., A∧B⇒C to say that when properties A and B are true in a record then, C tends to be also true) have become a popular way to summarize huge datasets. The last 5 years, there has been a lot of research on association rule mining and more precisely, the tractable discovery of interesting rules among the frequent ones. We consider now the problem of mining association rules that may involve negations e.g., A∧B⇒⌝C or ⌝A∧B⇒C. Mining such rules is difficult and remains an open problem. We identify several possibilities for a tractable approach in practical cases. Among others, we discuss the active use of constraints. We propose a generic algorithm and discuss the use of constraints to mine the generalized sets from which rules with negations can be derived.