Discriminative Markov Logic Network Structure Learning based on Propositionalization and chi 2-test

Abstract : In this paper we present a bottom-up discriminative algorithm to learn automatically Markov Logic Network structures. Our approach relies on a new propositionalization method that transforms a learning dataset into an approximative representation in the form of boolean tables, from which to construct a set of candidate clauses according to a chi2-test. To compute and choose clauses, we successively use two different optimization criteria, namely pseudo-log-likelihood (PLL) and conditional log-likelihood (CLL), in order to combine the efficiency of PLL optimization algorithms together with the accuracy of CLL ones. First experiments show that our approach outperforms the existing discriminative MLN structure learning algorithms.
Document type :
Conference papers
Longbing Cao, Yong Feng and Jiang Zhong. Advanced Data Mining and Applications - 6th International Conference, ADMA 2010, Nov 2010, Chongqing, China. Springer, 6440, pp.24-35, 2010, LNCS. 〈10.1007/978-3-642-17316-5〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00512439
Contributor : Matthieu Exbrayat <>
Submitted on : Monday, August 30, 2010 - 2:35:05 PM
Last modification on : Thursday, January 17, 2019 - 3:06:04 PM

Identifiers

Collections

Citation

Quang-Thang Dinh, Matthieu Exbrayat, Christel Vrain. Discriminative Markov Logic Network Structure Learning based on Propositionalization and chi 2-test. Longbing Cao, Yong Feng and Jiang Zhong. Advanced Data Mining and Applications - 6th International Conference, ADMA 2010, Nov 2010, Chongqing, China. Springer, 6440, pp.24-35, 2010, LNCS. 〈10.1007/978-3-642-17316-5〉. 〈hal-00512439〉

Share

Metrics

Record views

68