Detecting expert's eye using a multiple-kernel Relevance Vector Machine

Giuseppe Boccignone, Mario Ferraro, Sofia Crespi, Carlo Robino, Claudio De'Sperati

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Decoding mental states from the pattern of neural activity or overt behavior is an intensely pursued goal. Here we applied machine learning to detect expertise from the oculomotor behavior of novice and expert billiard players during free viewing of a filmed billiard match with no specific task, and in a dynamic trajectory prediction task involving ad-hoc, occluded billiard shots. We have adopted a ground framework for feature space fusion and a Bayesian sparse classifier, namely, a Relevance Vector Machine. By testing different combinations of simple oculomotor features (gaze shifts amplitude and direction, and fixation duration), we could classify on an individual basis which group - novice or expert - the observers belonged to with an accuracy of 82% and 87%, respectively for the match and the shots. These results provide evidence that, at least in the particular domain of billiard sport, a signature of expertise is hidden in very basic aspects of oculomotor behavior, and that expertise can be detected at the individual level both with ad-hoc testing conditions and under naturalistic conditions - and suitable data mining. Our procedure paves the way for the development of a test for the "expert's eye", and promotes the use of eye movements as an additional signal source in Brain-Computer-Interface (BCI) systems.

Original languageEnglish
Article number3
JournalJournal of Eye Movement Research
Volume7
Issue number2
Publication statusPublished - 2014

Fingerprint

Brain-Computer Interfaces
Data Mining
Computer Systems
Eye Movements
Sports
Direction compound
Machine Learning

Keywords

  • Billiards
  • Expertise
  • Eye movements
  • Feature fusion
  • Machine learning
  • Mind reading
  • Relevance vector machine

ASJC Scopus subject areas

  • Ophthalmology
  • Sensory Systems

Cite this

Detecting expert's eye using a multiple-kernel Relevance Vector Machine. / Boccignone, Giuseppe; Ferraro, Mario; Crespi, Sofia; Robino, Carlo; De'Sperati, Claudio.

In: Journal of Eye Movement Research, Vol. 7, No. 2, 3, 2014.

Research output: Contribution to journalArticle

Boccignone, Giuseppe ; Ferraro, Mario ; Crespi, Sofia ; Robino, Carlo ; De'Sperati, Claudio. / Detecting expert's eye using a multiple-kernel Relevance Vector Machine. In: Journal of Eye Movement Research. 2014 ; Vol. 7, No. 2.
@article{6acd9b6dc8b04603bd4c4b2edeef1b0d,
title = "Detecting expert's eye using a multiple-kernel Relevance Vector Machine",
abstract = "Decoding mental states from the pattern of neural activity or overt behavior is an intensely pursued goal. Here we applied machine learning to detect expertise from the oculomotor behavior of novice and expert billiard players during free viewing of a filmed billiard match with no specific task, and in a dynamic trajectory prediction task involving ad-hoc, occluded billiard shots. We have adopted a ground framework for feature space fusion and a Bayesian sparse classifier, namely, a Relevance Vector Machine. By testing different combinations of simple oculomotor features (gaze shifts amplitude and direction, and fixation duration), we could classify on an individual basis which group - novice or expert - the observers belonged to with an accuracy of 82{\%} and 87{\%}, respectively for the match and the shots. These results provide evidence that, at least in the particular domain of billiard sport, a signature of expertise is hidden in very basic aspects of oculomotor behavior, and that expertise can be detected at the individual level both with ad-hoc testing conditions and under naturalistic conditions - and suitable data mining. Our procedure paves the way for the development of a test for the {"}expert's eye{"}, and promotes the use of eye movements as an additional signal source in Brain-Computer-Interface (BCI) systems.",
keywords = "Billiards, Expertise, Eye movements, Feature fusion, Machine learning, Mind reading, Relevance vector machine",
author = "Giuseppe Boccignone and Mario Ferraro and Sofia Crespi and Carlo Robino and Claudio De'Sperati",
year = "2014",
language = "English",
volume = "7",
journal = "Journal of Eye Movement Research",
issn = "1995-8692",
publisher = "International Group for Eye Movement Research",
number = "2",

}

TY - JOUR

T1 - Detecting expert's eye using a multiple-kernel Relevance Vector Machine

AU - Boccignone, Giuseppe

AU - Ferraro, Mario

AU - Crespi, Sofia

AU - Robino, Carlo

AU - De'Sperati, Claudio

PY - 2014

Y1 - 2014

N2 - Decoding mental states from the pattern of neural activity or overt behavior is an intensely pursued goal. Here we applied machine learning to detect expertise from the oculomotor behavior of novice and expert billiard players during free viewing of a filmed billiard match with no specific task, and in a dynamic trajectory prediction task involving ad-hoc, occluded billiard shots. We have adopted a ground framework for feature space fusion and a Bayesian sparse classifier, namely, a Relevance Vector Machine. By testing different combinations of simple oculomotor features (gaze shifts amplitude and direction, and fixation duration), we could classify on an individual basis which group - novice or expert - the observers belonged to with an accuracy of 82% and 87%, respectively for the match and the shots. These results provide evidence that, at least in the particular domain of billiard sport, a signature of expertise is hidden in very basic aspects of oculomotor behavior, and that expertise can be detected at the individual level both with ad-hoc testing conditions and under naturalistic conditions - and suitable data mining. Our procedure paves the way for the development of a test for the "expert's eye", and promotes the use of eye movements as an additional signal source in Brain-Computer-Interface (BCI) systems.

AB - Decoding mental states from the pattern of neural activity or overt behavior is an intensely pursued goal. Here we applied machine learning to detect expertise from the oculomotor behavior of novice and expert billiard players during free viewing of a filmed billiard match with no specific task, and in a dynamic trajectory prediction task involving ad-hoc, occluded billiard shots. We have adopted a ground framework for feature space fusion and a Bayesian sparse classifier, namely, a Relevance Vector Machine. By testing different combinations of simple oculomotor features (gaze shifts amplitude and direction, and fixation duration), we could classify on an individual basis which group - novice or expert - the observers belonged to with an accuracy of 82% and 87%, respectively for the match and the shots. These results provide evidence that, at least in the particular domain of billiard sport, a signature of expertise is hidden in very basic aspects of oculomotor behavior, and that expertise can be detected at the individual level both with ad-hoc testing conditions and under naturalistic conditions - and suitable data mining. Our procedure paves the way for the development of a test for the "expert's eye", and promotes the use of eye movements as an additional signal source in Brain-Computer-Interface (BCI) systems.

KW - Billiards

KW - Expertise

KW - Eye movements

KW - Feature fusion

KW - Machine learning

KW - Mind reading

KW - Relevance vector machine

UR - http://www.scopus.com/inward/record.url?scp=84901201535&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901201535&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84901201535

VL - 7

JO - Journal of Eye Movement Research

JF - Journal of Eye Movement Research

SN - 1995-8692

IS - 2

M1 - 3

ER -