Formal Concept Analysis for Knowledge discovery: get ready for surprising practical applications

On Tuesday, the 16th of March, the International Workshop “Formal Concept Analysis for Knowledge Discovery” was held at the Faculty of Computer Science.

The event brought together   scientists and specialists of Data Analysis from St. Catherines (Canada), St. Petersburg, Novosibirsk, Tula, Kazan, Perm and other cities. Prof. Ivo Duentsch from Brock University (St. Catherines) made keynote talk Knowledge structures and skill assignments: Structural tools for diagnostic assessment.
The participants shared their successful experience of using FCA-based applications that allows solving the following tasks:

-         creating tools for knowledge assessment;
-         obtaining well-interpretable credit scoring models in accordance with strict bank standards and outperforming the models used in industry;

-         development of the individualized schemes of lymphoblastic leukemia treatment that substantially  improve life forecast;
-         improving mobile services by identifying the user’s needs in a smart way;
-         etc.
              Apart from the impressive practical contribution of FCA, the participants have got to know recent theoretical results in the following areas:
-         quality analysis of ontologies;
-         theoretical estimates of the probability of accidental formal concepts and the overfitting probability of classifiers;
-         development of approaches to improving neural network interpretability;
-         studies on n-ary concepts as the powerful tools for representation of data with complex structure.

Lengthy discussions, interesting questions, ideas and views on possible developments in these areas will undoubtedly encourage and inspire all participants to new discoveries.
The organizers thank all the participants for interesting talks, active and fruitful discussions and look forward to similar workshops in future.