Using Brain Data for Sentiment Analysis

Yuqiao Gu, Fabio Celli, Josef Steinberger, Andrew James Anderson, Massimo Poesio, Carlo Strapparava, Brian Murphy

Research output: Contribution to journalArticle

Abstract

We present the results of exploratory experiments using lexical valence extracted from brain using electroencephalography (EEG) for sentiment analysis. We selected 78 English words (36 for training and 42 for testing), presented as stimuli to 3 English native speakers. EEG signals were recorded from the subjects while they performed a mental imaging task for each word stimulus. Wavelet decomposition was employed to extract EEG features from the time-frequency domain. The extracted features were used as inputs to a sparse multinomial logistic regression (SMLR) classifier for valence classification, after univariate ANOVA feature selection. After mapping EEG signals to sentiment valences, we exploited the lexical polarity extracted from brain data for the prediction of the valence of 12 sentences taken from the SemEval-2007 shared task, and compared it against existing lexical resources.
Original languageEnglish
Pages (from-to)79-94
Number of pages16
JournalJournal for Language Technology and Computational Linguistics
Volume29
Issue number1
Publication statusPublished - 2014

Fingerprint

Electroencephalography
Brain
Wavelet decomposition
Classifiers
Imaging techniques
Testing
Experiments

Cite this

Gu, Y., Celli, F., Steinberger, J., Anderson, A. J., Poesio, M., Strapparava, C., & Murphy, B. (2014). Using Brain Data for Sentiment Analysis. Journal for Language Technology and Computational Linguistics, 29(1), 79-94.
Gu, Yuqiao ; Celli, Fabio ; Steinberger, Josef ; Anderson, Andrew James ; Poesio, Massimo ; Strapparava, Carlo ; Murphy, Brian. / Using Brain Data for Sentiment Analysis. In: Journal for Language Technology and Computational Linguistics. 2014 ; Vol. 29, No. 1. pp. 79-94.
@article{e5e8672d970642bbab7ee134415c9e31,
title = "Using Brain Data for Sentiment Analysis",
abstract = "We present the results of exploratory experiments using lexical valence extracted from brain using electroencephalography (EEG) for sentiment analysis. We selected 78 English words (36 for training and 42 for testing), presented as stimuli to 3 English native speakers. EEG signals were recorded from the subjects while they performed a mental imaging task for each word stimulus. Wavelet decomposition was employed to extract EEG features from the time-frequency domain. The extracted features were used as inputs to a sparse multinomial logistic regression (SMLR) classifier for valence classification, after univariate ANOVA feature selection. After mapping EEG signals to sentiment valences, we exploited the lexical polarity extracted from brain data for the prediction of the valence of 12 sentences taken from the SemEval-2007 shared task, and compared it against existing lexical resources.",
author = "Yuqiao Gu and Fabio Celli and Josef Steinberger and Anderson, {Andrew James} and Massimo Poesio and Carlo Strapparava and Brian Murphy",
year = "2014",
language = "English",
volume = "29",
pages = "79--94",
journal = "Journal for Language Technology and Computational Linguistics",
issn = "2190-6858",
number = "1",

}

Gu, Y, Celli, F, Steinberger, J, Anderson, AJ, Poesio, M, Strapparava, C & Murphy, B 2014, 'Using Brain Data for Sentiment Analysis', Journal for Language Technology and Computational Linguistics, vol. 29, no. 1, pp. 79-94.

Using Brain Data for Sentiment Analysis. / Gu, Yuqiao; Celli, Fabio; Steinberger, Josef; Anderson, Andrew James; Poesio, Massimo; Strapparava, Carlo; Murphy, Brian.

In: Journal for Language Technology and Computational Linguistics, Vol. 29, No. 1, 2014, p. 79-94.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Using Brain Data for Sentiment Analysis

AU - Gu, Yuqiao

AU - Celli, Fabio

AU - Steinberger, Josef

AU - Anderson, Andrew James

AU - Poesio, Massimo

AU - Strapparava, Carlo

AU - Murphy, Brian

PY - 2014

Y1 - 2014

N2 - We present the results of exploratory experiments using lexical valence extracted from brain using electroencephalography (EEG) for sentiment analysis. We selected 78 English words (36 for training and 42 for testing), presented as stimuli to 3 English native speakers. EEG signals were recorded from the subjects while they performed a mental imaging task for each word stimulus. Wavelet decomposition was employed to extract EEG features from the time-frequency domain. The extracted features were used as inputs to a sparse multinomial logistic regression (SMLR) classifier for valence classification, after univariate ANOVA feature selection. After mapping EEG signals to sentiment valences, we exploited the lexical polarity extracted from brain data for the prediction of the valence of 12 sentences taken from the SemEval-2007 shared task, and compared it against existing lexical resources.

AB - We present the results of exploratory experiments using lexical valence extracted from brain using electroencephalography (EEG) for sentiment analysis. We selected 78 English words (36 for training and 42 for testing), presented as stimuli to 3 English native speakers. EEG signals were recorded from the subjects while they performed a mental imaging task for each word stimulus. Wavelet decomposition was employed to extract EEG features from the time-frequency domain. The extracted features were used as inputs to a sparse multinomial logistic regression (SMLR) classifier for valence classification, after univariate ANOVA feature selection. After mapping EEG signals to sentiment valences, we exploited the lexical polarity extracted from brain data for the prediction of the valence of 12 sentences taken from the SemEval-2007 shared task, and compared it against existing lexical resources.

M3 - Article

VL - 29

SP - 79

EP - 94

JO - Journal for Language Technology and Computational Linguistics

JF - Journal for Language Technology and Computational Linguistics

SN - 2190-6858

IS - 1

ER -

Gu Y, Celli F, Steinberger J, Anderson AJ, Poesio M, Strapparava C et al. Using Brain Data for Sentiment Analysis. Journal for Language Technology and Computational Linguistics. 2014;29(1):79-94.