bibtype	C - Conference Paper (international conference)
ARLID	0507111
utime	20240103222342.6
mtime	20190731235959.9
SCOPUS	85061163500
DOI	10.1007/978-3-030-10925-7_16
title (primary) (eng)	Multiple Instance Learning with Bag-Level Randomized Trees
specification	page_count 14 s. media_type P
serial	ARLID cav_un_epca*0507110 ISBN 978-3-030-10925-7 title Machine Learning and Knowledge Discovery in Databases page_num 259-272 publisher place Cham name Springer International Publishing year 2019 editor name1 Berlingerio name2 M. editor name1 Bonchi name2 F. editor name1 Gärtner name2 T. editor name1 Hurley name2 N. editor name1 Ifrim name2 G.
keyword	Multiple instance learning
keyword	randomized trees
keyword	classification
author (primary)	ARLID cav_un_auth*0352238 name1 Komárek name2 T. country CZ
author	ARLID cav_un_auth*0101197 name1 Somol name2 Petr full_dept (cz) Rozpoznávání obrazu full_dept Department of Pattern Recognition department (cz) RO department RO institution UTIA-B full_dept Department of Pattern Recognition fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
source	url http://library.utia.cas.cz/separaty/2019/RO/somol-0507111.pdf
cas_special	abstract (eng) Knowledge discovery in databases with a flexible structure poses a great challenge to machine learning community. Multiple Instance Learning (MIL) aims at learning from samples (called bags) represented by multiple feature vectors (called instances) as opposed to single feature vectors characteristic for the traditional data representation. This relaxation turns out to be useful in formulating many machine learning problems including classification of molecules, cancer detection from tissue images or identification of malicious network communications. However, despite the recent progress in this area, the current set of MIL tools still seems to be very application specific and/or burdened with many tuning parameters or processing steps. In this paper, we propose a simple, yet effective tree-based algorithm for solving MIL classification problems. Empirical evaluation against 28 classifiers on 29 publicly available benchmark datasets shows a high level performance of the proposed solution even with its default parameter settings. action ARLID cav_un_auth*0377819 name Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD) dates 20180910 mrcbC20-s 20180914 place Dublin country IE RIV BC FORD0 20000 FORD1 20200 FORD2 20204 reportyear 2020 num_of_auth 2 presentation_type PR inst_support RVO:67985556 permalink http://hdl.handle.net/11104/0298754 confidential S arlyear 2019 mrcbU14 85061163500 SCOPUS mrcbU24 PUBMED mrcbU34 WOS mrcbU63 cav_un_epca*0507110 Machine Learning and Knowledge Discovery in Databases Springer International Publishing 2019 Cham 259 272 978-3-030-10925-7 Lecture Notes in Computer Science 11051 mrcbU67 340 Berlingerio M. mrcbU67 340 Bonchi F. mrcbU67 340 Gärtner T. mrcbU67 340 Hurley N. mrcbU67 340 Ifrim G.