Eighth International Workshop “What can FCA do for Artificial Intelligence?”
The seven preceding editions of the FCA4AI Workshop showed that many researchers working in Artificial Intelligence are deeply interested by a well-founded method for classification and data mining such as Formal Concept Analysis (see https://conceptanalysis. wordpress.com/fca/). FCA4AI was co-located with ECAI 2012 (Montpellier), IJCAI 2013 (Beijing), ECAI 2014 (Prague), IJCAI 2015 (Buenos Aires), ECAI 2016 (The Hague), IJCAI/ECAI 2018 (Stockholm), and IJCAI 2019 (Macao). The workshop has now a quite long history and all the proceedings are available as CEUR proceedings (see http://ceur-ws. org/, volumes 939, 1058, 1257, 1430, 1703, 2149, and 2529). This year, the workshop has again attracted many researchers from many countries working on actual and important topics related to FCA, showing the diversity and the richness of the relations between FCA and AI. Formal Concept Analysis (FCA) is a mathematically well-founded theory aimed at data analysis and classification. FCA allows one to build a concept lattice and a system of dependencies (implications and association rules) which can be used for many Artificial Intelligence needs, e.g. knowledge discovery, learning, knowledge representation, reasoning, ontology engineering, as well as information retrieval and text processing. Recent years have been witnessing increased scientific activity around FCA, in particular a strand of work emerged that is aimed at extending the possibilities of FCA w.r.t. knowledge processing, such as work on pattern structures, relational context analysis, and triadic analysis. These extensions are aimed at allowing FCA to deal with more complex data, both from the data analysis and knowledge discovery points of view. Actually these investigations provide new possibilities for AI practitioners within the framework of FCA. Accordingly, we are interested in the following issues: • How can FCA support AI activities such as knowledge processing, i.e. knowledge discovery, knowledge representation and reasoning, learning, i.e. clustering, pattern and data mining, natural language processing, and information retrieval (non exhaustive list). • How can FCA be extended in order to help Artificial Intelligence researchers to solve new and complex problems in their domains. The workshop is dedicated to discussion of such issues. First of all we would like to thank all the authors for their contributions and all the PC members for their reviews and precious collaboration. This year, 24 papers were submitted and 14 were accepted for presentation at the workshop, out of which 6 short papers. The papers submitted to the workshop were carefully peer-reviewed by three members of the program committee. Finally, the order of the papers in the proceedings (see page 5) follows the program order (see http://fca4ai. hse.ru/2020/).
The goal of subgroup discovery is to find groups of objectsthat are significantly different than “average” object w.r.t. some super-vised information. It is a computational intensive procedure that tra-verses a large searching space corresponding to the set of formal con-cepts. It was recently found that a part of formal concepts, called stableconcepts, can be found in polynomial time. Accordingly, in this paper anew algorithm, calledSD-SOFIA, is presented.SD-SOFIAfits subgroup dis-covery process in the framework of stable concept search. The proposedalgorithm is evaluated on a dataset from UCI repository. It is shown thatits practical computational complexity is polynomial.