A rule-based approach to implicit emotion detection in text

Udochukwu Orizu, Yulan He

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Most research in the area of emotion detection in written text focused on detecting explicit expressions of emotions in text. In this paper, we present a rule-based pipeline approach for detecting implicit emotions in written text without emotion-bearing words based on the OCC Model. We have evaluated our approach on three different datasets with five emotion categories. Our results show that the proposed approach outperforms the lexicon matching method consistently across all the three datasets by a large margin of 17–30% in F-measure and gives competitive performance compared to a supervised classifier. In particular, when dealing with formal text which follows grammatical rules strictly, our approach gives an average F-measure of 82.7% on “Happy”, “Angry-Disgust” and “Sad”, even outperforming the supervised baseline by nearly 17% in F-measure. Our preliminary results show the feasibility of the approach for the task of implicit emotion detection in written text.

Original languageEnglish
Title of host publicationNatural language processing and information systems
Subtitle of host publication20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings
EditorsChris Biemann, Siegfried Handschuh, André Freitas, et al
Place of PublicationCham (CHE)
PublisherSpringer
Pages197-203
Number of pages7
ISBN (Electronic)978-3-319-19581-0
ISBN (Print)978-3-319-19580-3
DOIs
Publication statusPublished - 4 Jun 2015
Event20th International Conference on Applications of Natural Language to Information Systems - Passau, Germany
Duration: 17 Jun 201519 Jun 2015

Publication series

NameLecture notes in computer science
PublisherSpringer
Volume9103
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference20th International Conference on Applications of Natural Language to Information Systems
Abbreviated titleNLDB 2015
CountryGermany
CityPassau
Period17/06/1519/06/15

Fingerprint

Bearings (structural)
Classifiers
Pipelines
Margin
Emotion
Text
Baseline
Strictly
Classifier

Bibliographical note

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-19581-0_17

Keywords

  • emotion detection
  • implicit emotions
  • OCC model
  • rule-based approach

Cite this

Orizu, U., & He, Y. (2015). A rule-based approach to implicit emotion detection in text. In C. Biemann, S. Handschuh, A. Freitas, & et al (Eds.), Natural language processing and information systems: 20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings (pp. 197-203). (Lecture notes in computer science; Vol. 9103). Cham (CHE): Springer. https://doi.org/10.1007/978-3-319-19581-0_17
Orizu, Udochukwu ; He, Yulan. / A rule-based approach to implicit emotion detection in text. Natural language processing and information systems: 20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings. editor / Chris Biemann ; Siegfried Handschuh ; André Freitas ; et al. Cham (CHE) : Springer, 2015. pp. 197-203 (Lecture notes in computer science).
@inproceedings{bc5c945795da4c129ef6315fb931e355,
title = "A rule-based approach to implicit emotion detection in text",
abstract = "Most research in the area of emotion detection in written text focused on detecting explicit expressions of emotions in text. In this paper, we present a rule-based pipeline approach for detecting implicit emotions in written text without emotion-bearing words based on the OCC Model. We have evaluated our approach on three different datasets with five emotion categories. Our results show that the proposed approach outperforms the lexicon matching method consistently across all the three datasets by a large margin of 17–30{\%} in F-measure and gives competitive performance compared to a supervised classifier. In particular, when dealing with formal text which follows grammatical rules strictly, our approach gives an average F-measure of 82.7{\%} on “Happy”, “Angry-Disgust” and “Sad”, even outperforming the supervised baseline by nearly 17{\%} in F-measure. Our preliminary results show the feasibility of the approach for the task of implicit emotion detection in written text.",
keywords = "emotion detection, implicit emotions, OCC model, rule-based approach",
author = "Udochukwu Orizu and Yulan He",
note = "The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-19581-0_17",
year = "2015",
month = "6",
day = "4",
doi = "10.1007/978-3-319-19581-0_17",
language = "English",
isbn = "978-3-319-19580-3",
series = "Lecture notes in computer science",
publisher = "Springer",
pages = "197--203",
editor = "Chris Biemann and Siegfried Handschuh and Andr{\'e} Freitas and {et al}",
booktitle = "Natural language processing and information systems",
address = "Germany",

}

Orizu, U & He, Y 2015, A rule-based approach to implicit emotion detection in text. in C Biemann, S Handschuh, A Freitas & et al (eds), Natural language processing and information systems: 20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings. Lecture notes in computer science, vol. 9103, Springer, Cham (CHE), pp. 197-203, 20th International Conference on Applications of Natural Language to Information Systems, Passau, Germany, 17/06/15. https://doi.org/10.1007/978-3-319-19581-0_17

A rule-based approach to implicit emotion detection in text. / Orizu, Udochukwu; He, Yulan.

Natural language processing and information systems: 20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings. ed. / Chris Biemann; Siegfried Handschuh; André Freitas; et al. Cham (CHE) : Springer, 2015. p. 197-203 (Lecture notes in computer science; Vol. 9103).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A rule-based approach to implicit emotion detection in text

AU - Orizu, Udochukwu

AU - He, Yulan

N1 - The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-19581-0_17

PY - 2015/6/4

Y1 - 2015/6/4

N2 - Most research in the area of emotion detection in written text focused on detecting explicit expressions of emotions in text. In this paper, we present a rule-based pipeline approach for detecting implicit emotions in written text without emotion-bearing words based on the OCC Model. We have evaluated our approach on three different datasets with five emotion categories. Our results show that the proposed approach outperforms the lexicon matching method consistently across all the three datasets by a large margin of 17–30% in F-measure and gives competitive performance compared to a supervised classifier. In particular, when dealing with formal text which follows grammatical rules strictly, our approach gives an average F-measure of 82.7% on “Happy”, “Angry-Disgust” and “Sad”, even outperforming the supervised baseline by nearly 17% in F-measure. Our preliminary results show the feasibility of the approach for the task of implicit emotion detection in written text.

AB - Most research in the area of emotion detection in written text focused on detecting explicit expressions of emotions in text. In this paper, we present a rule-based pipeline approach for detecting implicit emotions in written text without emotion-bearing words based on the OCC Model. We have evaluated our approach on three different datasets with five emotion categories. Our results show that the proposed approach outperforms the lexicon matching method consistently across all the three datasets by a large margin of 17–30% in F-measure and gives competitive performance compared to a supervised classifier. In particular, when dealing with formal text which follows grammatical rules strictly, our approach gives an average F-measure of 82.7% on “Happy”, “Angry-Disgust” and “Sad”, even outperforming the supervised baseline by nearly 17% in F-measure. Our preliminary results show the feasibility of the approach for the task of implicit emotion detection in written text.

KW - emotion detection

KW - implicit emotions

KW - OCC model

KW - rule-based approach

UR - http://www.scopus.com/inward/record.url?scp=84948844177&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-19581-0_17

DO - 10.1007/978-3-319-19581-0_17

M3 - Conference contribution

AN - SCOPUS:84948844177

SN - 978-3-319-19580-3

T3 - Lecture notes in computer science

SP - 197

EP - 203

BT - Natural language processing and information systems

A2 - Biemann, Chris

A2 - Handschuh, Siegfried

A2 - Freitas, André

A2 - et al,

PB - Springer

CY - Cham (CHE)

ER -

Orizu U, He Y. A rule-based approach to implicit emotion detection in text. In Biemann C, Handschuh S, Freitas A, et al, editors, Natural language processing and information systems: 20th international conference on applications of Natural Language to Information Systems, NLDB 2015, Passau, Germany, June 17-19, 2015, proceedings. Cham (CHE): Springer. 2015. p. 197-203. (Lecture notes in computer science). https://doi.org/10.1007/978-3-319-19581-0_17