Learning Discriminative Tree Edit Similarities for Linear Classification - Application to Melody Recognition

Aurélien Bellet 1 José Bernabeu 2 Amaury Habrard 3 Marc Sebban 3
1 MAGNET - Machine Learning in Information Networks
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189
Abstract : Similarity functions are a fundamental component of many learning algorithms. When dealing with string or tree-structured data, measures based on the edit distance are widely used, and there exist a few methods for learning them from data. In this context, we recently proposed GESL [3], an approach to string edit similarity learning based on loss minimization which offers theoretical guarantees as to the generalization ability and discriminative power of the learned similarities. In this paper, we argue that GESL, which has been originally dedicated to deal with strings, can be extended to trees and lead to powerful and competitive similarities. We illustrate this claim on a music recognition task, namely melody classification, where each piece is represented as a tree modeling its structure as well as rhythm and pitch information. The results show that GESL outperforms standard as well as probabilistically-learned edit distances, and that it is able to describe consistently the underlying melodic similarity model.
Document type :
Journal articles
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01330492
Contributor : Marc Sebban <>
Submitted on : Sunday, November 20, 2016 - 11:26:17 AM
Last modification on : Tuesday, April 2, 2019 - 2:27:33 AM
Long-term archiving on : Monday, March 27, 2017 - 9:15:07 AM

File

ldTree2015.pdf
Files produced by the author(s)

Identifiers

Citation

Aurélien Bellet, José Bernabeu, Amaury Habrard, Marc Sebban. Learning Discriminative Tree Edit Similarities for Linear Classification - Application to Melody Recognition. Neurocomputing, Elsevier, 2016, 214, pp.155-161. ⟨10.1016/j.neucom.2016.06.006⟩. ⟨hal-01330492⟩

Share

Metrics

Record views

688

Files downloads

177