Skip to Main content Skip to Navigation
New interface
Conference papers

Mining the Lattice of Binary Classifiers for Identifying Duplicate Labels in Behavioral Data

Quentin Labernia 1 Victor Codocedo 1 Céline Robardet 1 Mehdi Kaytoue 1 
1 DM2L - Data Mining and Machine Learning
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : Analysis of behavioral data represents today a big issue, as so many domains generate huge quantity of activity and mobility traces. When traces are labeled by the user that generates it, models can be learned to accurately predict the user of an unknown trace. In online systems however, users may have several virtual identities, or duplicate labels. By ignoring them, the prediction accuracy drastically drops, as the set of all virtual identities of a single person is not known beforehand. In this article, we tackle this duplicate labels identification problem, and present an original approach that explores the lattice of binary classifiers. Each subset of labels is learned as the positive class against the others (the negative class), and constraints make possible to identify duplicate labels while pruning the search space. We experiment this original approach with data of the video game Starcraft 2 in the new context of Electronic Sports (eSport) with encouraging results.
Document type :
Conference papers
Complete list of metadata

Cited literature [8 references]  Display  Hide  Download
Contributor : Mehdi Kaytoue Connect in order to contact the contributor
Submitted on : Friday, June 30, 2017 - 10:53:04 AM
Last modification on : Friday, September 30, 2022 - 11:34:16 AM
Long-term archiving on: : Monday, January 22, 2018 - 7:57:57 PM


Files produced by the author(s)



Quentin Labernia, Victor Codocedo, Céline Robardet, Mehdi Kaytoue. Mining the Lattice of Binary Classifiers for Identifying Duplicate Labels in Behavioral Data. 30th International Conference on Industrial Engineering and Other Applications of Applied Intelligent Systems, Jun 2017, Arras, France. pp.40-21, ⟨10.1007/978-3-319-60045-1_2⟩. ⟨hal-01551395⟩



Record views


Files downloads