Data Mining with Algorithmic Transparency

Simple item page
dc.contributor.ORCID0000-0001-6423-4533 (Kantarcioglu, M)
dc.contributor.VIAF305367293 (Kantarcioglu, M)
dc.contributor.authorZhou, Yan
dc.contributor.authorAlufaisan, Yasmeen
dc.contributor.authorKantarcioglu, Murat
dc.contributor.utdAuthorZhou, Yan
dc.contributor.utdAuthorAlufaisan, Yasmeen
dc.contributor.utdAuthorKantarcioglu, Murat
dc.date.accessioned2019-07-25T22:35:29Z
dc.date.available2019-07-25T22:35:29Z
dc.date.created2018-06-19
dc.description.abstractIn this paper, we investigate whether decision trees can be used to interpret a black-box classifier without knowing the learning algorithm and the training data. Decision trees are known for their transparency and high expressivity. However, they are also notorious for their instability and tendency to grow excessively large. We present a classifier reverse engineering model that outputs a decision tree to interpret the black-box classifier. There are two major challenges. One is to build such a decision tree with controlled stability and size, and the other is that probing the black-box classifier is limited for security and economic reasons. Our model addresses the two issues by simultaneously minimizing sampling cost and classifier complexity. We present our empirical results on four real datasets, and demonstrate that our reverse engineering learning model can effectively approximate and simplify the black box classifier.
dc.identifier.bibliographicCitationZhou, Y., Y. Alufaisan, and M. Kantarcioglu. 2018. "Data mining with algorithmic transparency." Lecture Notes In Computer Science 10937:130-142; doi:10.1007/978-3-319-93034-3_11
dc.identifier.isbn9783319930336
dc.identifier.urihttps://hdl.handle.net/10735.1/6724
dc.identifier.volume10937
dc.language.isoen
dc.publisherSpringer Verlag
dc.relation.urihttp://dx.doi.org/10.1007/978-3-319-93034-3_11
dc.rights©2018 Springer International Publishing AG, part of Springer Nature
dc.source.journalLecture Notes In Computer Science
dc.subjectInformation organization
dc.subjectDecision trees
dc.subjectAlgorithms
dc.subjectReverse engineering
dc.subjectTransparency
dc.subjectData mining
dc.titleData Mining with Algorithmic Transparency
dc.type.genrearticle

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
JECS-2796-279814.35-LINK.pdf
Size:
162.97 KB
Format:
Adobe Portable Document Format
Description:
Link to Article
Download

Collections

Kantarcioglu, Murat
JECS Staff and Student Research