Cervantes-Alvarez, José F.Gutiérrez-Preciado, Luis F.Ortega-Guzmán, Víctor H.2024-07-312024-07-312024-07Ortega-Guzmán, V. H. (2024). Metodología para el descubrimiento de conocimiento en grafos. Tesis de doctorado, Doctorado en Ciencias de la Ingeniería. Tlaquepaque, Jalisco: ITESO.https://hdl.handle.net/11117/11021Graph mining has experienced remarkable growth in recent years, fueled by the proliferation of data amenable to graph representation and its diverse applications. This surge has been facilitated by the emergence of robust graph databases like Neo4J, AllegroGraph, and OrientDB, enabling researchers and practitioners to harness advanced graph mining algorithms for tasks ranging from community analysis to pathfinding. However, relying solely on graph databases proves inadequate, necessitating the development of a comprehensive methodology to guide practical information analysis and insight extraction. This doctoral thesis introduces the Knowledge Discovery in Graphs (KDG) methodology in response to this need. KDG aims to facilitate the extraction, transformation, loading, processing, modeling, visualization, and analysis of complex, interrelated information encapsulated within labeled and heterogeneous graphs. By leveraging mining algorithms, KDG seeks to unearth hidden structural attributes within these graphs, empowering stakeholders with actionable insights for decision-making processes. The methodology outlined in this thesis provides a holistic overview of the knowledge discovery process, modeled using graphs. It encompasses a broad spectrum of tasks, including information exploration, pattern detection, recommendation generation, visualization tool utilization, and inquiry resolution, which are essential for effective decision-making. The document is structured gradually, beginning with the presentation of fundamental concepts in the first chapter, followed by an exhaustive review of the literature on graph mining in the second chapter. Subsequently, the innovative KDG methodology is introduced in the third chapter, detailing its application to analyze interconnected information in graphs. The practical effectiveness of KDG is evidenced through three case studies in the fourth chapter, providing a concrete view of its implementation in contexts. The document concludes with a chapter dedicated to conclusions. Finally, this thesis synthesizes findings, outlines future research avenues, and includes appendices comprising a comprehensive reference list and a compilation of published works, offering a robust foundation for further exploration in graph mining and knowledge discovery.engMethodologyGraph MiningKnowledge DiscoveryLabeled Graphsinfo:eu-repo/semantics/doctoralThesis