Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis

Jürgen Bernard, Christian Ritter, David Sessler, Matthias Zeppelzauer, Jörn Kohlhammer, Dieter Fellner

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

The definition of similarity is a key prerequisite when analyzing complex data types in data mining, information retrieval, or machine learning. However, the meaningful definition is often hampered by the complexity of data objects and particularly by different notions of subjective similarity latent in targeted user groups. Taking the example of soccer players, we present a visual-interactive system that learns users' mental models of similarity. In a visual-interactive interface, users are able to label pairs of soccer players with respect to their subjective notion of similarity. Our proposed similarity model automatically learns the respective concept of similarity using an active learning strategy. A visual-interactive retrieval technique is provided to validate the model and to execute downstream retrieval tasks for soccer player analysis. The applicability of the approach is demonstrated in different evaluation strategies, including usage scenarions and cross-validation tests.

Conference

Conference8th International Conference on Information Visualization Theory and Applications
Abbreviated titleIVAPP 2017
CountryPortugal
CityPorto
Period27/02/171/03/17

Keywords

  • cs.LG
  • cs.IR

Fields of Expertise

  • Information, Communication & Computing

Cite this

Bernard, J., Ritter, C., Sessler, D., Zeppelzauer, M., Kohlhammer, J., & Fellner, D. (2017). Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. In IVAPP 2017 DOI: 10.5220/0006116400750087

Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. / Bernard, Jürgen; Ritter, Christian; Sessler, David; Zeppelzauer, Matthias; Kohlhammer, Jörn; Fellner, Dieter.

IVAPP 2017. 2017.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Bernard, J, Ritter, C, Sessler, D, Zeppelzauer, M, Kohlhammer, J & Fellner, D 2017, Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. in IVAPP 2017. 8th International Conference on Information Visualization Theory and Applications, Porto, Portugal, 27/02/17. DOI: 10.5220/0006116400750087
Bernard J, Ritter C, Sessler D, Zeppelzauer M, Kohlhammer J, Fellner D. Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. In IVAPP 2017. 2017. Available from, DOI: 10.5220/0006116400750087
Bernard, Jürgen ; Ritter, Christian ; Sessler, David ; Zeppelzauer, Matthias ; Kohlhammer, Jörn ; Fellner, Dieter. / Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis. IVAPP 2017. 2017.
@inproceedings{e12e5bfa8b6947d382df306ff629ec1c,
title = "Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis",
abstract = "The definition of similarity is a key prerequisite when analyzing complex data types in data mining, information retrieval, or machine learning. However, the meaningful definition is often hampered by the complexity of data objects and particularly by different notions of subjective similarity latent in targeted user groups. Taking the example of soccer players, we present a visual-interactive system that learns users' mental models of similarity. In a visual-interactive interface, users are able to label pairs of soccer players with respect to their subjective notion of similarity. Our proposed similarity model automatically learns the respective concept of similarity using an active learning strategy. A visual-interactive retrieval technique is provided to validate the model and to execute downstream retrieval tasks for soccer player analysis. The applicability of the approach is demonstrated in different evaluation strategies, including usage scenarions and cross-validation tests.",
keywords = "cs.LG, cs.IR",
author = "J{\"u}rgen Bernard and Christian Ritter and David Sessler and Matthias Zeppelzauer and J{\"o}rn Kohlhammer and Dieter Fellner",
year = "2017",
month = "3",
day = "9",
doi = "10.5220/0006116400750087",
language = "English",
booktitle = "IVAPP 2017",

}

TY - GEN

T1 - Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis

AU - Bernard,Jürgen

AU - Ritter,Christian

AU - Sessler,David

AU - Zeppelzauer,Matthias

AU - Kohlhammer,Jörn

AU - Fellner,Dieter

PY - 2017/3/9

Y1 - 2017/3/9

N2 - The definition of similarity is a key prerequisite when analyzing complex data types in data mining, information retrieval, or machine learning. However, the meaningful definition is often hampered by the complexity of data objects and particularly by different notions of subjective similarity latent in targeted user groups. Taking the example of soccer players, we present a visual-interactive system that learns users' mental models of similarity. In a visual-interactive interface, users are able to label pairs of soccer players with respect to their subjective notion of similarity. Our proposed similarity model automatically learns the respective concept of similarity using an active learning strategy. A visual-interactive retrieval technique is provided to validate the model and to execute downstream retrieval tasks for soccer player analysis. The applicability of the approach is demonstrated in different evaluation strategies, including usage scenarions and cross-validation tests.

AB - The definition of similarity is a key prerequisite when analyzing complex data types in data mining, information retrieval, or machine learning. However, the meaningful definition is often hampered by the complexity of data objects and particularly by different notions of subjective similarity latent in targeted user groups. Taking the example of soccer players, we present a visual-interactive system that learns users' mental models of similarity. In a visual-interactive interface, users are able to label pairs of soccer players with respect to their subjective notion of similarity. Our proposed similarity model automatically learns the respective concept of similarity using an active learning strategy. A visual-interactive retrieval technique is provided to validate the model and to execute downstream retrieval tasks for soccer player analysis. The applicability of the approach is demonstrated in different evaluation strategies, including usage scenarions and cross-validation tests.

KW - cs.LG

KW - cs.IR

U2 - 10.5220/0006116400750087

DO - 10.5220/0006116400750087

M3 - Conference contribution

BT - IVAPP 2017

ER -