Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
Probabilistic formal concepts with negation. / Vityaev, E. E.; Martinovich, V. V.
Perspectives of System Informatics - 9th International Ershov Informatics Conference, PSI 2014, Revised Selected Papers. ed. / Irina Virbitskaite; Andrei Voronkov; Irina Virbitskaite. Springer, 2015. p. 385-399 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8974).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
}
TY - GEN
T1 - Probabilistic formal concepts with negation
AU - Vityaev, E. E.
AU - Martinovich, V. V.
PY - 2015
Y1 - 2015
N2 - The probabilistic generalization of formal concept analysis, as well as it’s comparison to standard formal analysis is presented. Construction is resistant to noise in the data and give one an opportunity to consider contexts with negation (object-attribute relation which allows both attribute presence and it’s absence). This generalization is obtained from the notion of formal concepts with its definition as fixed points of implications, when implications, possibly with negations, are replaced by probabilistic laws. We prove such fixed points (based on the probabilistic implications) to be consistent and wherefore determine correct probabilistic formal concepts. In the end, the demonstration for the probabilistic formal concepts formation is given together with noise resistance example.
AB - The probabilistic generalization of formal concept analysis, as well as it’s comparison to standard formal analysis is presented. Construction is resistant to noise in the data and give one an opportunity to consider contexts with negation (object-attribute relation which allows both attribute presence and it’s absence). This generalization is obtained from the notion of formal concepts with its definition as fixed points of implications, when implications, possibly with negations, are replaced by probabilistic laws. We prove such fixed points (based on the probabilistic implications) to be consistent and wherefore determine correct probabilistic formal concepts. In the end, the demonstration for the probabilistic formal concepts formation is given together with noise resistance example.
KW - Association rules
KW - Data mining
KW - Formal concept analysis
KW - Noise
KW - Probability
UR - http://www.scopus.com/inward/record.url?scp=84942543651&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-46823-4_31
DO - 10.1007/978-3-662-46823-4_31
M3 - Conference contribution
AN - SCOPUS:84942543651
SN - 9783662468227
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 385
EP - 399
BT - Perspectives of System Informatics - 9th International Ershov Informatics Conference, PSI 2014, Revised Selected Papers
A2 - Virbitskaite, Irina
A2 - Voronkov, Andrei
A2 - Virbitskaite, Irina
PB - Springer
T2 - 9th International Ershov Informatics Conference on Perspectives of System Informatics, PSI 2014
Y2 - 24 June 2014 through 27 June 2014
ER -
ID: 25327916