1,720,975 research outputs found
Fouille de données et analyse de qualité des règles d’association dans les bases de données massives : Application dans le domaine de la sécurité routière
No full textKnowledge discovery in databases (KDD), often called Data Mining. « Is the non-trivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns in data ». Data mining is an active field of research aiming to exploit the vast amounts of data collected every day in various fields of computer science applications. This multidisciplinary field comes from artificial intelligence, statistics, and databases.In this thesis, we are interested in the problem of extracting association rules by introducing new algorithms and approaches. In general, an association rule is a conditional implication between sets of binary attributes called items. The extraction of such rules is composed of two main steps which are the extraction of frequent itemsets and the generation of association rules from them. The complexity of each of these steps is exponential: the number of frequent itemsets is exponential, and the number of association rules extracted can be very high, due to the quality measures used. In the literature, the extraction of the association rules is composed of two main difficulties, the response time and the memory space.To overcome these difficulties, we propose in this thesis three main contributions respectively allowing the extraction of relevant association rules, the integration of the spatial component into the extraction process, and mining relevant association rules from big data. In the first contribution, we propose an extraction approach of the relevant association rules based on multicriteria decision analysis. Then, in the second contribution, we propose an efficient algorithm for extracting spatial predicates from which frequent sets of items and spatial association rules can be generated based on the preparation of the spatial context and the fuzzy set theory. We also proposed in the third contribution a distributed algorithm for the extraction of association rules from Big Data. Using these contributions, we were able to extract the relevant association rules and reduce the execution time and memory space.Besides, to test concretely the contribution of the proposed solutions, we designed and developed a software prototype consisting of three interfaces. The first entitled ARM interface, is an interactive web interface dedicated to the extraction of association rules. The second interface, entitled MCDA interface, it is an interactive web interface dedicated to the evaluation and extraction of relevant association rules. For the last one, entitledTime Series Forecasting, is an interactive web interface dedicated to the prediction of road accidents. Moreover, interactive and user-friendly interfaces have been developed by using R language and rshiny. Finally, the experiments conducted on some databases on road accidents in Morocco show the significant feasibility of our contributions.L'extraction de connaissances dans les bases de données (ECD), également appelée fouille de données, « désigne le processus non trivial d'extraction d'information implicite, précédemment inconnue et potentiellement utile ». La fouille de données est un domaine de recherche en plein essor visant à exploiter les grandes quantités de données collectées chaque jour dans divers domaines d'application de l'informatique. Ce domaine pluridisciplinaire est issu de l'intelligence artificielle, des statistiques et des bases de données.Dans ce travail, nous nous intéressons au problème de l'extraction des règles d'association en introduisant de nouveaux algorithmes et approches d'aide à la décision multicritère. D'une manière générale, une règle d'association est une implication conditionnelle entre des ensembles d'attributs binaires appelés items. L'extraction de telles règles est décomposée en deux étapes principales, à savoir l'extraction des itemsets fréquents et la génération des règles d'association à partir de ceux-ci. Dans la majorité des approches existantes dans la littérature, l'extraction des règles d'association présente trois difficultés majeurs, à savoir; la qualité des règles extraites, l'aspect spatiale de données et le temps de réponse des algorithmes d'extraction.Pour surmonter ces difficultés, nous proposons dans cette thèse l'intégration de l'analyse multicritère au processus d'extraction des règles d'association pour l'analyse de la qualité. Ensuite, afin de prendre en considération l'aspect spatiale de données, et plus précisément l'estimation des distances métriques, nous avons proposé l'utilisation de la logique floue. Nous avons proposé également une intégration de l'algorithme FP-growth dans un environnement du Big Data pour l'extraction des règles d'association dans les bases de données massives.En plus, en vue de tester concrètement l'apport des solutions proposées, nous avons conçu et développé un prototype logiciel constitué de trois interfaces interactives. La première intitulée interface ARM, est une interface web dédiée à l'extraction des règles d'association. La deuxième interface, intitulée interface MCDA, est une interface web dédiée à l'analyse de qualité des règles d'association extraites. Quant à la dernière, intitulée Time Series Forcasting, est une interface web dédiée à la prédiction des accidents routières en termes du nombre de blessures et décès. Ces interfaces interactives d'exploration de données ont été développées en utilisant le langage R et rshiny. En fin, les expérimentations menées sur quelques bases de données relatives aux accidents routières au Maroc montrent la faisabilité notable de nos contributions
KBot : a Knowledge graph based chatBot for natural language understanding over linked data
No full textWith the rapid progress of the semantic web, a huge amount of structured data has become available on the web in the form of knowledge bases (KBs). Making these data accessible and useful for end-users is one of the main objectives of chatbots over linked data. Building a chatbot over linked data raises different challenges, including user queries understanding, multiple knowledge base support, and multilingual aspect. To address these challenges, we first design and develop an architecture to provide an interactive user interface. Secondly, we propose a machine learning approach based on intent classification and natural language understanding to understand user intents and generate SPARQL queries. We especially process a new social network dataset (i.e., myPersonality) and add it to the existing knowledge bases to extend the chatbot capabilities by understanding analytical queries. The system can be extended with a new domain on-demand, flexible, multiple knowledge base, multilingual, and allows intuitive creation and execution of different tasks for an extensive range of topics. Furthermore, evaluation and application cases in the chatbot are provided to show how it facilitates interactive semantic data towards different real application scenarios and showcase the proposed approach for a knowledge graph and data-driven chatbot
DM-MCDA: A web-based platform for data mining and multiple criteria decision analysis : A case study on road accident
No full textToday's ultra-connected world is generating a huge amount of data stored in databases and cloud environment especially in the era of transportation. These databases need to be processed and analyzed to extract useful information and present it as a valid element for transportation managers for further use, such as road safety, shipping delays, and shipping optimization. The potential of data mining algorithms is largely untapped, this paper shows large-scale techniques such as associations rule analysis, multiple criteria analysis, and time series to improve road safety by identifying hot-spots in advance and giving chance to drivers to avoid the dangers. Indeed, we proposed a framework DM-MCDA based on association rules mining as a preliminary task to extract relationships between variables related to a road accident, and then integrate multiple criteria analysis to help decision-makers to make their choice of the most relevant rules. The developed system is flexible and allows intuitive creation and execution of different algorithms for an extensive range of road traffic topics. DM-MCDA can be expanded with new topics on demand, rendering knowledge extraction more robust and provide meaningful information that could help in developing suitable policies for decision-makers
Comparative survey of association rule mining algorithms based on multiple-criteria decision analysis approach
No full textToward efficient resource utilization at edge nodes in federated learning
Get PDFFederated learning (FL) enables edge nodes to collaboratively contribute to
constructing a global model without sharing their data. This is accomplished by
devices computing local, private model updates that are then aggregated by a
server. However, computational resource constraints and network communication
can become a severe bottleneck for larger model sizes typical for deep learning
applications. Edge nodes tend to have limited hardware resources (RAM, CPU),
and the network bandwidth and reliability at the edge is a concern for scaling
federated fleet applications. In this paper, we propose and evaluate a FL
strategy inspired by transfer learning in order to reduce resource utilization
on devices, as well as the load on the server and network in each global
training round. For each local model update, we randomly select layers to
train, freezing the remaining part of the model. In doing so, we can reduce
both server load and communication costs per round by excluding all untrained
layer weights from being transferred to the server. The goal of this study is
to empirically explore the potential trade-off between resource utilization on
devices and global model convergence under the proposed strategy. We implement
the approach using the federated learning framework FEDn. A number of
experiments were carried out over different datasets (CIFAR-10, CASA, and
IMDB), performing different tasks using different deep-learning model
architectures. Our results show that training the model partially can
accelerate the training process, efficiently utilizes resources on-device, and
reduce the data transmission by around 75% and 53% when we train 25%, and 50%
of the model layers, respectively, without harming the resulting global model
accuracy.Comment: 16 pages, 5 tables, 8 figure
A Web-Based Platform for Mining and Ranking Association Rules
No full textIn this demo, we introduce an interactive system, which effectively applies multiple criteria analysis to rank association rules. We first use association rules techniques to explore the correlations between variables in given data (i.e., database and linked data (LD)), and secondly apply multiple criteria analysis (MCA) to select the most relevant rules according to user preferences. The developed system is flexible and allows intuitive creation and execution of different algorithms for an extensive range of advanced data analysis topics. Furthermore, we demonstrate a case study of association rule mining and ranking on road accident data.</p
Bridging AI and Privacy : Federated Learning for Leukemia Diagnosis
No full textLeukemia is a heterogeneous group of hematologic malignancies, with acute lymphoblastic leukemia (ALL) being one of the most harmful forms. Accurate and early diagnosis is crucial for effective treatment, potentially saving lives. Recent advances in machine learning (ML) and deep learning (DL) have significantly enhanced diagnostic capabilities. However, these advancements often compromise the confidentiality of sensitive medical data. In this paper, we propose a federated learning (FL) framework for the binary classification of ALL versus normal cases. This framework leverages decentralized data from multiple clients, where each client trains its model locally on its own data, transmitting only model updates to a central server. The central server then aggregates these updates using the FedAvg algorithm, creating a global model while ensuring that patient data remains at its source, thereby preserving confidentiality. Using an EfficientNetV2S-based model architecture and a dataset of 10,661 images containing normal cells and lymphoblasts, our experiments demonstrate that the proposed FL approach achieves an accuracy of 95.6% and a kappa coefficient of 0.89. This performance is competitive with centralized methods while maintaining data privacy. These results highlight the potential of FL to revolutionize the clinical detection of acute lymphoblastic leukemia, offering a scalable and privacy-preserving solution for medical applications.</p
Decentralized Diagnosis : Privacy-Preserving Brain Tumor Classification with Federated Learning
No full textBrain tumors pose a significant global health challenge, driving ongoing research advancements in early detection methods. Artificial intelligence (AI) and deep learning (DL) techniques have shown great potential in this field, enabling the creation of highly accurate models for brain tumor identification from medical images. However, centralized approaches to these methods often raise critical concerns regarding patient data privacy and security. This paper presents a novel federated learning (FL) framework for brain tumor identification that effectively addresses these privacy concerns. FL enables collaborative model training across multiple institutions without the need for raw data sharing. Each participating institution trains the model locally on their Magnetic Resonance Imaging (MRI) datasets and only transmits model updates to a central server for secure aggregation. This iterative process results in a robust global model trained on a distributed dataset while preserving patient data confidentiality. The proposed FL model is evaluated using a dataset of 3,000 MRI images. Experimental results demonstrate the effectiveness of our approach, achieving a high accuracy rate of 96.88% for brain tumor identification. These findings suggest that FL provides a viable solution for privacy-preserving brain tumor identification, maintaining comparable performance to centralized models while ensuring the security of patient data.KDD-AIDSH 2024 [email protected]</p
- …
