TY - GEN
T1 - EmotiWord
T2 - International Workshop on Multimedia Understanding Through Semantics, Computation, and Learning, MUSCLE 2011
AU - Malandrakis, Nikos
AU - Potamianos, Alexandros
AU - Iosif, Elias
AU - Narayanan, Shrikanth
PY - 2012
Y1 - 2012
N2 - We present a fully automated algorithm for expanding an affective lexicon with new entries. Continuous valence ratings are estimated for unseen words using the underlying assumption that semantic similarity implies affective similarity. Starting from a set of manually annotated words, a linear affective model is trained using the least mean squares algorithm followed by feature selection. The proposed algorithm performs very well on reproducing the valence ratings of the Affective Norms for English Words (ANEW) and General Inquirer datasets. We then propose three simple linear and non-linear fusion schemes for investigating how lexical valence scores can be combined to produce sentence-level scores. These methods are tested on a sentence rating task of the SemEval 2007 corpus, on the ChIMP politeness and frustration detection dialogue task and on a movie subtitle polarity detection task.
AB - We present a fully automated algorithm for expanding an affective lexicon with new entries. Continuous valence ratings are estimated for unseen words using the underlying assumption that semantic similarity implies affective similarity. Starting from a set of manually annotated words, a linear affective model is trained using the least mean squares algorithm followed by feature selection. The proposed algorithm performs very well on reproducing the valence ratings of the Affective Norms for English Words (ANEW) and General Inquirer datasets. We then propose three simple linear and non-linear fusion schemes for investigating how lexical valence scores can be combined to produce sentence-level scores. These methods are tested on a sentence rating task of the SemEval 2007 corpus, on the ChIMP politeness and frustration detection dialogue task and on a movie subtitle polarity detection task.
KW - affect
KW - affective lexicon
KW - emotion
KW - language understanding
UR - http://www.scopus.com/inward/record.url?scp=84867138198&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-32436-9_3
DO - 10.1007/978-3-642-32436-9_3
M3 - Conference contribution
AN - SCOPUS:84867138198
SN - 9783642324352
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 30
EP - 41
BT - Computational Intelligence for Multimedia Understanding - International Workshop, MUSCLE 2011, Revised Selected Papers
Y2 - 13 December 2011 through 15 December 2011
ER -