Learning Hypotheses from Triadic Labeled Data

Author

Ignatov, Dmitry I. ; Zhuk, Roman ; Konstantinova, Natalia

Author_Institution

Sch. of Appl. Math. & Inf. Sci., Lab. of Intell. Syst. & Struct. Anal., Moscow, Russia

Volume

2

fYear

2014

fDate

11-14 Aug. 2014

Firstpage

474

Lastpage

480

Abstract

We propose extensions of the classical JSM-method and the Naive Bayesian classifier for the case of triadic relational data. We performed a series of experiments on various types of data (both real and synthetic) to estimate quality of classification techniques and compare them with other classification algorithms that generate hypotheses, e.g. ID3 and Random Forest. In addition to classification precision and recall we also evaluated the time performance of the proposed methods.

Keywords

Bayes methods; learning (artificial intelligence); pattern classification; relational databases; JSM-method; classification algorithms; classification precision; classification recall; classification techniques; learning hypotheses; naive Bayesian classifier; real data; synthetic data; time performance; triadic labeled data; triadic relational data; Cats; Context; Dogs; Educational institutions; Formal concept analysis; Manganese; Noise; Classification; Formal Concept Analysis; JSM-method; Triadic data;

fLanguage

English

Publisher

ieee

Conference_Titel

Web Intelligence (WI) and Intelligent Agent Technologies (IAT), 2014 IEEE/WIC/ACM International Joint Conferences on

Conference_Location

Warsaw

Type

conf

DOI

10.1109/WI-IAT.2014.136

Filename

6927663