Special Session on Formal Concept Analysis, Rough Set Theory and Their Applications


Objectives
Formal Concept Analysis (FCA) is a mathematically well-founded theory aimed at applications in data analysis and knowledge discovery. FCA allows one to build a concept lattice and dependencies of different kinds (rules, implications, etc.) for input data in both supervised and unsupervised forms. This machinery can be be used for many purposes, e.g. learning from data, biclustering, knowledge representation, reasoning, ontology engineering, information retrieval, recommendation, and text processing.

On the one hand FCA provides operationalisation of objects similarity in terms of their common attributes (or common descriptions), on the other hand Rough Set Theory studies their discernibility and provides convenient means for approximation, which are complementary, in a certain sense, to fuzzy concepts. Accordingly, there are many links between FCA and Rough Set Theory that can be found in data analysis, pattern mining, fuzzy sets theory, database theory, algebraic representation, logic systems, reasoning systems, etc.

Recent surveys on connections between Rough and Fuzzy formalisms in FCA domain, and on models and techniques can be found in:

1. Jonas Poelmans, Dmitry I. Ignatov, Sergei O. Kuznetsov, Guido Dedene:
Fuzzy and rough formal concept analysis: a survey. Int. J. General Systems 43(2): 105-134 (2014)
http://dx.doi.org/10.1080/03081079.2013.862377

2. Jonas Poelmans, Sergei O. Kuznetsov, Dmitry I. Ignatov, Guido Dedene:
Formal Concept Analysis in knowledge processing: A survey on models and techniques. Expert Syst. Appl. 40(16): 6601-6623
(2013)
http://dx.doi.org/10.1016/j.eswa.2013.05.007

Recent years have shown an increased activity in FCA, in particular in extending FCA to different formalisms: pattern
structures, database representation, fuzzy FCA, relational FCA, polyadic concepts, etc. These extensions allow FCA to deal with more complex than just binary object-attribute data (e.g. RDF data, rating matrices, sequences and graphs), for data analysis, knowledge discovery and knowledge engineering.
Therefore, this special session will be interested in issues such as:

- Relationship between FCA and Rough Set Theory, fuzzy closure systems, fuzzy orders.
- RS/FCA algorithms for Big Data
- RS/FCA algorithms for complex data
- How can existing FCA algorithms help RS and vice versa? - RS/FCA for Knowledge Discovery and Pattern Mining
- RS/FCA for Knowledge Engineering and Representation
- Bireducts and biclusters
- Rule-based approaches in RS/FCA
- Multi-way extensions of RS/FCA

Topics of interest.

The main topics include, but not limited to:

- Foundations of FCA and RS
- Scalable algorithms in FCA and RS for Big Data.
- Applications of FCA and RS in complex data: text mining, classification and mining in web of data, information retrieval, recommendation, visualization and navigation.

Organizers & Session Chairs:
• Jaume Baixeries
• Dmitry Ignatov dignatov {at} hse.ru
• Mehdi Kaytoue
• Sergei Kuznetsov

Important Dates

Deadline for submitting 10-20 page regular conference papers: March 28 midnight Greenwich time
Deadline for submitting 6+ page short papers: March 28 midnight Greenwich time
Notification of acceptance for 10-20 page regular conference papers and 6+ page long short papers: to be announced
Deadline for submitting camera-ready accepted conference papers: April 15, 2017
Conference: July 3-7, 2017


The main deadlines are in line with the conference schedule: http://ijcrs2017.uwm.edu.pl/