Indonesian affective word resources construction in valence and arousal dimension for sentiment analysis
Research in the field of text analysis will always be related to words, either the selection of words to be used or the position of the words in a sentence. Furthermore, a hypothesis that each language difference can cause different meanings, makes some researchers interested in doing research c...
| Main Authors: | , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
IEEE
2019
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/69472/ http://irep.iium.edu.my/69472/ http://irep.iium.edu.my/69472/ http://irep.iium.edu.my/69472/7/69472%20Indonesian%20Affective%20Word%20Resources.pdf http://irep.iium.edu.my/69472/8/69472%20Indonesian%20Affective%20Word%20Resources%20SCOPUS.pdf |
| Summary: | Research in the field of text analysis will always be
related to words, either the selection of words to be used or the
position of the words in a sentence. Furthermore, a hypothesis
that each language difference can cause different meanings,
makes some researchers interested in doing research classifying
words based on emotion or affective words. Research focuses on
affective states as a continuous numerical value to the dimensions
of valence and arousal. Sentiment analysis that is usually done
with positive and negative category approaches, nowadays, the
dimensional approach can provide more analysis of grained
sentiments. On the other hand, the affective words dataset with
valence and arousal rating are still very rare, especially for the
Indonesian language. Therefore, this research does an affective
lexicon dataset called Indonesian Valence and Arousal Words
(IVAW) containing 1024 words by Self-Assessment Manikin
(SAM) surveys. Furthermore, for the next study, we will also
crawls status in twitter based on selected words from IVAW to
get Indonesian Valence and Arousal Text (IVAT). To predict VA
rating for obtaining the advance of annotation quality,
experiment will be compared by brain signal using EEG tool. |
|---|