Automation of style acquisition and implementation in 3D scene synthesis Dimitrios Makris 1 , Georgios Bardis 1 , Georgios Miaoulis 1 , Dimitri Plemenos 2 1 Technological Education Institute of Athens Department of Informatics Ag.Spyridonos St., 122 10 Egaleo, GREECE Email: demak@teiath.gr, gbardis@teiath.gr, gmiaoul@teiath.gr 2 XLIM Laboratory, University of Limoges, 83 rue d'Isle, Limoges, 87000, France Email: plemenos@unilim.fr Abstract In this paper we illustrate an integrated Computational Intelligence technique using machine learning and evolutionary search methods for the automation of qualitative aspects acquisition and implementation in conceptual design of 3D scene synthesis. The aim is to alleviate the task of adapting a scene to the particularities of a certain morphological or other intuitive demands while obeying a given set of functional, financial or other requirements. The approach comprises a knowledge obtainment stage, where the machine learning mechanism acquires the style implications of the morphology of a series of examples, and a generation stage, where the evolutionary mechanism conducts a partial exploration of the solution space of scenes compliant to the given set of requirements. The paradigm domain where the study of such a mechanism is placed is that of building design defined through declarative modeling by hierarchical decomposition. Keywords Computational Intelligence, Machine Learning, Evolutionary Search, Declarative Modeling, Qualitative Aspects, Style, Conceptual Design.