Enhanced senticnet with affective labels for concept-based opinion mining

Soujanya Poria; Alexander Gelbukh; Amir Hussain; Newton Howard; Dipankar Das; Sivaji Bandyopadhyay

doi:10.1109/MIS.2013.40

Enhanced senticnet with affective labels for concept-based opinion mining

Soujanya Poria, Alexander Gelbukh, Amir Hussain, Newton Howard, Dipankar Das, Sivaji Bandyopadhyay

Centro de Investigación en Computación (CIC)

Research output: Contribution to journal › Article › peer-review

156 Scopus citations

Abstract

A methodology to automatically assign emotion labels to SenticNet concepts is proposed. SenticNet was supplied with affective WordNet-Affect (WNA) compatible labels using a machine learning algorithm. The WNA list dataset consists of six word lists corresponding to Ekman's six basic emotions, anger, disgust, fear, joy, sadness, and surprise. The dataset contains 606 synsets, of which all but two are assigned exactly one label each. If the synsets are broken down into individual concepts, the dataset contains 1,536 concepts. For each SenticNet concept, ISEAR statistical features of its occurrences and cooccurrences with other SenticNet concepts in ISEAR statements were extracted. English WordNet 3.0 was used to measure the semantic distance between two words. This work opens up multiple directions for future research, such as using other types of monolingual or multilingual corpora as a source of features to improve the accuracy or to label more concepts.

Original language	English
Pages (from-to)	2-9
Number of pages	8
Journal	IEEE Intelligent Systems
Volume	28
Issue number	2
DOIs	https://doi.org/10.1109/MIS.2013.40
State	Published - 2013

Access to Document

10.1109/MIS.2013.40

Cite this

@article{ca83054eb2944a619f4d25ef6da4f84e,

title = "Enhanced senticnet with affective labels for concept-based opinion mining",

abstract = "A methodology to automatically assign emotion labels to SenticNet concepts is proposed. SenticNet was supplied with affective WordNet-Affect (WNA) compatible labels using a machine learning algorithm. The WNA list dataset consists of six word lists corresponding to Ekman's six basic emotions, anger, disgust, fear, joy, sadness, and surprise. The dataset contains 606 synsets, of which all but two are assigned exactly one label each. If the synsets are broken down into individual concepts, the dataset contains 1,536 concepts. For each SenticNet concept, ISEAR statistical features of its occurrences and cooccurrences with other SenticNet concepts in ISEAR statements were extracted. English WordNet 3.0 was used to measure the semantic distance between two words. This work opens up multiple directions for future research, such as using other types of monolingual or multilingual corpora as a source of features to improve the accuracy or to label more concepts.",

author = "Soujanya Poria and Alexander Gelbukh and Amir Hussain and Newton Howard and Dipankar Das and Sivaji Bandyopadhyay",

year = "2013",

doi = "10.1109/MIS.2013.40",

language = "Ingl{\'e}s",

volume = "28",

pages = "2--9",

journal = "IEEE Intelligent Systems",

issn = "1541-1672",

number = "2",

}

TY - JOUR

T1 - Enhanced senticnet with affective labels for concept-based opinion mining

AU - Poria, Soujanya

AU - Gelbukh, Alexander

AU - Hussain, Amir

AU - Howard, Newton

AU - Das, Dipankar

AU - Bandyopadhyay, Sivaji

PY - 2013

Y1 - 2013

N2 - A methodology to automatically assign emotion labels to SenticNet concepts is proposed. SenticNet was supplied with affective WordNet-Affect (WNA) compatible labels using a machine learning algorithm. The WNA list dataset consists of six word lists corresponding to Ekman's six basic emotions, anger, disgust, fear, joy, sadness, and surprise. The dataset contains 606 synsets, of which all but two are assigned exactly one label each. If the synsets are broken down into individual concepts, the dataset contains 1,536 concepts. For each SenticNet concept, ISEAR statistical features of its occurrences and cooccurrences with other SenticNet concepts in ISEAR statements were extracted. English WordNet 3.0 was used to measure the semantic distance between two words. This work opens up multiple directions for future research, such as using other types of monolingual or multilingual corpora as a source of features to improve the accuracy or to label more concepts.

AB - A methodology to automatically assign emotion labels to SenticNet concepts is proposed. SenticNet was supplied with affective WordNet-Affect (WNA) compatible labels using a machine learning algorithm. The WNA list dataset consists of six word lists corresponding to Ekman's six basic emotions, anger, disgust, fear, joy, sadness, and surprise. The dataset contains 606 synsets, of which all but two are assigned exactly one label each. If the synsets are broken down into individual concepts, the dataset contains 1,536 concepts. For each SenticNet concept, ISEAR statistical features of its occurrences and cooccurrences with other SenticNet concepts in ISEAR statements were extracted. English WordNet 3.0 was used to measure the semantic distance between two words. This work opens up multiple directions for future research, such as using other types of monolingual or multilingual corpora as a source of features to improve the accuracy or to label more concepts.

UR - http://www.scopus.com/inward/record.url?scp=84898734877&partnerID=8YFLogxK

U2 - 10.1109/MIS.2013.40

DO - 10.1109/MIS.2013.40

M3 - Artículo

SN - 1541-1672

VL - 28

SP - 2

EP - 9

JO - IEEE Intelligent Systems

JF - IEEE Intelligent Systems

IS - 2

ER -

Enhanced senticnet with affective labels for concept-based opinion mining

Abstract

Access to Document

Other files and links

Fingerprint

Cite this