Supervision : Christophe MARSALA

Co-supervision : CASTANIE Laurent

Knowledge extraction from complex 3D structures

This PhD work proposes to valorize a certain type of data that are the objects (structures) 3D constructed from mesh, by empirically justifying the undeniable contributions of an extraction of subparts coming from these latest. This objective is achieved by solving a forecast problem by a new supervised classification approach for the information recommendation. Beyond the expected result, a justification of the latter is also provided in the form of the visualization of sub-parts extracted discriminant, thus allowing interpretation by the specialist.
Thanks to an adaptation of Time series Shapelets and features selection methods, it is possible to select only the most relevant parts for the desired classification.
In addition to slightly better forecast results than those provided by the state of the art, it offers a visualization of the sub-parts of the most discriminating 3D objects within the framework of the classification model implemented, and therefore the areas that will have most allowed to classify the data.
Moreover, we propose an improvement of this method by two paths: the first one is the contribution of an adaptation of the transfer of knowledge (or transfer learning) applied to the previously proposed algorithm; the second one is an innovative method of feature selection, based on tools derived from fuzzy subset theory, is introduced, which proves to be potentially applicable to any type of attribute selection in supervised classification.
Results confirm the general potential of random selection of candidate attributes, especially in the context of large amounts of data.

2015-2018 Publications

• 2018
  • F. Meunier : “ Prédiction de phénomènes géologiques pour l’aide à la décision lors de la prise de permis d’exploitation”, thesis, phd defence 12/13/2018, supervision Marsala, Christophe, co-supervision : Castanie, Laurent (2018)
  • F. Meunier, Ch. Marsala, L. Castanie : “Mining 3D-Structures: Subparts Extraction and Transfer Learning”, IEEE International Conference on Systems, Man, and Cybernetics (SMC'2018), Miyazaki, Japan (2018)
  • F. Meunier, Ch. Marsala, L. Castanie : “Classification of 3D objects by random extraction of discriminant sub-parts for the study of the sub-soil in oil prospecting”, chapter in Advances in Knowledge Discovery and Management (2018)
• 2017
  • F. Meunier, Ch. Marsala, C. De Runz, L. Castanie : “Sélection d’attributs pour la classification d’objets 3D”, Rencontres Francophones sur la Logique Floue et ses Applications, Amiens, France, pp. 83-90, (Cépaduès Editions) (2017)
  • F. Meunier, Ch. Marsala, L. Castanie : “3DRESC-TF : Apprentissage par transfert pour la réutilisation de connaissances en classification d’objets 3D”, Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle, Caen, France (2017)
  • F. Meunier, Ch. Marsala, L. Castanie, B. Conche : “Classification d’objets 3D par extraction de sous-parties discriminantes pour l’étude du sous-sol en prospection pétrolière”, EGC 2017, vol. RNTI-E-33, Revue des Nouvelles Technologies de l'Information, Grenoble, France, pp. 225-236 (2017)
• 2016
  • F. Meunier, Ch. Marsala, L. Castanie : “Heat Kernel Signature pour la sélection et la classification d’objets - application aux géo-modèles 3D”, La 5^e édition des journées « Big Data Mining and Visualization », Metz, France (2016)
• 2015
  • F. Meunier, Ch. Marsala, L. Castanie : “Development of a recommendation system for decision support in geosciences”, MATHIAS – TOTAL Symposium on Mathematics, Paris, France (2015)