Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon

Nir Ofek, Corneli Caragea, Prakhaa Biyani, Lior Rokach, Prasenjit Mitra, John Yen, Kenneth Portier, Greta Greer

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

10 Citations (Scopus)

Abstract

Online Health Communities is a major source for patients and their family members in the process of gathering information and seeking social support. The American Cancer Society Cancer Survivors Network has many users and presents a large number of users' interactions with regards to coping with cancer. Sentiment analysis is an important step in understanding participants' needs and concerns and the impact of users' responses on other members. We present an automated approach for sentiment analysis in an online cancer survivor community and compare it with a previous sentiment analysis approach. Both approaches are machine learning based and are tested on the same dataset. However, this work uses features derived from a dynamic sentiment lexicon, whereas the previous work uses a general sentiment lexicon to extract features. Tested on several classifiers, with only six features (versus thirteen), our results show 2.3% improvement on average, in terms of accuracy, and greater improvement in F-measure and AUC. An additional experiment was conducted that showed a positive impact of dimensionality reduction by extracting abstract features, instead of using term frequency (TF) vector space as attribute values.

Original languageEnglish
Title of host publicationProceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013
Pages109-113
Number of pages5
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event2013 International Conference on Social Intelligence and Technology, SOCIETY 2013 - State College, PA
Duration: 8 May 201310 May 2013

Other

Other2013 International Conference on Social Intelligence and Technology, SOCIETY 2013
CityState College, PA
Period8/5/1310/5/13

Fingerprint

Vector spaces
Learning systems
Classifiers
Health
Experiments
Sentiment analysis
Sentiment
Cancer
Survivors

Keywords

  • Abstract features
  • Dynamic sentiment lexicon
  • Sentiment analysis

ASJC Scopus subject areas

  • Management of Technology and Innovation
  • Artificial Intelligence

Cite this

Ofek, N., Caragea, C., Biyani, P., Rokach, L., Mitra, P., Yen, J., ... Greer, G. (2013). Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon. In Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013 (pp. 109-113). [6545971] https://doi.org/10.1109/SOCIETY.2013.20

Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon. / Ofek, Nir; Caragea, Corneli; Biyani, Prakhaa; Rokach, Lior; Mitra, Prasenjit; Yen, John; Portier, Kenneth; Greer, Greta.

Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013. 2013. p. 109-113 6545971.

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

Ofek, N, Caragea, C, Biyani, P, Rokach, L, Mitra, P, Yen, J, Portier, K & Greer, G 2013, Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon. in Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013., 6545971, pp. 109-113, 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013, State College, PA, 8/5/13. https://doi.org/10.1109/SOCIETY.2013.20
Ofek N, Caragea C, Biyani P, Rokach L, Mitra P, Yen J et al. Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon. In Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013. 2013. p. 109-113. 6545971 https://doi.org/10.1109/SOCIETY.2013.20
Ofek, Nir ; Caragea, Corneli ; Biyani, Prakhaa ; Rokach, Lior ; Mitra, Prasenjit ; Yen, John ; Portier, Kenneth ; Greer, Greta. / Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon. Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013. 2013. pp. 109-113
@inproceedings{605c487384344611aa07933e6124a39c,
title = "Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon",
abstract = "Online Health Communities is a major source for patients and their family members in the process of gathering information and seeking social support. The American Cancer Society Cancer Survivors Network has many users and presents a large number of users' interactions with regards to coping with cancer. Sentiment analysis is an important step in understanding participants' needs and concerns and the impact of users' responses on other members. We present an automated approach for sentiment analysis in an online cancer survivor community and compare it with a previous sentiment analysis approach. Both approaches are machine learning based and are tested on the same dataset. However, this work uses features derived from a dynamic sentiment lexicon, whereas the previous work uses a general sentiment lexicon to extract features. Tested on several classifiers, with only six features (versus thirteen), our results show 2.3{\%} improvement on average, in terms of accuracy, and greater improvement in F-measure and AUC. An additional experiment was conducted that showed a positive impact of dimensionality reduction by extracting abstract features, instead of using term frequency (TF) vector space as attribute values.",
keywords = "Abstract features, Dynamic sentiment lexicon, Sentiment analysis",
author = "Nir Ofek and Corneli Caragea and Prakhaa Biyani and Lior Rokach and Prasenjit Mitra and John Yen and Kenneth Portier and Greta Greer",
year = "2013",
doi = "10.1109/SOCIETY.2013.20",
language = "English",
pages = "109--113",
booktitle = "Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013",

}

TY - GEN

T1 - Improving sentiment analysis in an online cancer survivor community using dynamic sentiment lexicon

AU - Ofek, Nir

AU - Caragea, Corneli

AU - Biyani, Prakhaa

AU - Rokach, Lior

AU - Mitra, Prasenjit

AU - Yen, John

AU - Portier, Kenneth

AU - Greer, Greta

PY - 2013

Y1 - 2013

N2 - Online Health Communities is a major source for patients and their family members in the process of gathering information and seeking social support. The American Cancer Society Cancer Survivors Network has many users and presents a large number of users' interactions with regards to coping with cancer. Sentiment analysis is an important step in understanding participants' needs and concerns and the impact of users' responses on other members. We present an automated approach for sentiment analysis in an online cancer survivor community and compare it with a previous sentiment analysis approach. Both approaches are machine learning based and are tested on the same dataset. However, this work uses features derived from a dynamic sentiment lexicon, whereas the previous work uses a general sentiment lexicon to extract features. Tested on several classifiers, with only six features (versus thirteen), our results show 2.3% improvement on average, in terms of accuracy, and greater improvement in F-measure and AUC. An additional experiment was conducted that showed a positive impact of dimensionality reduction by extracting abstract features, instead of using term frequency (TF) vector space as attribute values.

AB - Online Health Communities is a major source for patients and their family members in the process of gathering information and seeking social support. The American Cancer Society Cancer Survivors Network has many users and presents a large number of users' interactions with regards to coping with cancer. Sentiment analysis is an important step in understanding participants' needs and concerns and the impact of users' responses on other members. We present an automated approach for sentiment analysis in an online cancer survivor community and compare it with a previous sentiment analysis approach. Both approaches are machine learning based and are tested on the same dataset. However, this work uses features derived from a dynamic sentiment lexicon, whereas the previous work uses a general sentiment lexicon to extract features. Tested on several classifiers, with only six features (versus thirteen), our results show 2.3% improvement on average, in terms of accuracy, and greater improvement in F-measure and AUC. An additional experiment was conducted that showed a positive impact of dimensionality reduction by extracting abstract features, instead of using term frequency (TF) vector space as attribute values.

KW - Abstract features

KW - Dynamic sentiment lexicon

KW - Sentiment analysis

UR - http://www.scopus.com/inward/record.url?scp=84881186112&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881186112&partnerID=8YFLogxK

U2 - 10.1109/SOCIETY.2013.20

DO - 10.1109/SOCIETY.2013.20

M3 - Conference contribution

SP - 109

EP - 113

BT - Proceedings - 2013 International Conference on Social Intelligence and Technology, SOCIETY 2013

ER -