ETD system

Electronic theses and dissertations repository

 

Tesi etd-09192016-170443


Thesis type
Tesi di laurea magistrale
Author
CATRAMBONE, VINCENZO
email address
catrambone.vincenzo@gmail.com
URN
etd-09192016-170443
Title
Computational modeling and analysis of ECoG data through Support Vector Machines algorithms.
Struttura
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA BIOMEDICA
Commissione
relatore Prof. Vanello, Nicola
controrelatore Prof. Landini, Luigi
relatore Prof. Formisano, Elia
Parole chiave
  • SVM
  • ECoG
  • Multivariate Analysis
  • Machine Learning
Data inizio appello
07/10/2016;
Consultabilità
parziale
Data di rilascio
07/10/2019
Riassunto analitico
In this work some ElectoCorticography data has been taken in account, a multivariate analysis has been proposed considering spatial and time-frequency features. The Dataset has been recorded following the Recalibration and Adaptation paradigm about ambigous and not-ambigous audio-visual stimulation. A classification between these two conditions through a Support Vector Machine algorithm using a RBF kernel has been implemented. An intra- and inter- subject discussion of the results will close the dissertation.
File