Sentiment Classification by Combining Triplet Belief Functions

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

Abstract

Sentiment analysis is an emerging technique that caters for semantic orientation and opinion mining. It is increasingly used to anal- yse online product reviews for identifying customers’ opinions and atti- tudes to products or services in order to improve business performance of companies. This paper presents an innovative approach to combining outputs of sentiment classifiers under the framework of belief functions. The approach is composed of the formulation of outputs of sentiment classifiers in the triplet structure and adoption of its formulas to combin- ing simple support functions derived from triplet functions by evidential combination rules. The empirical studies have been conducted on the performance of sentiment classification individually and in combination, the experimental results show that the best combined classifiers made by these combination rules outperform the best individual classifiers over the MP3 and Movie-Review datasets.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages234-245
Number of pages12
Volume8793
Publication statusPublished - 16 Oct 2014
EventInternational Conference on Knowledge Science, Engineering and Management 2014 - Romania
Duration: 16 Oct 2014 → …

Conference

ConferenceInternational Conference on Knowledge Science, Engineering and Management 2014
Period16/10/14 → …

Fingerprint

Classifiers
Industry
Semantics

Keywords

  • Sentiment analysis
  • opinion mining
  • triplet belief functions and combination rules

Cite this

Bi, Y., Mulvenna, M., & Jurek, A. (2014). Sentiment Classification by Combining Triplet Belief Functions. In Unknown Host Publication (Vol. 8793, pp. 234-245)
Bi, Yaxin ; Mulvenna, Maurice ; Jurek, Anna. / Sentiment Classification by Combining Triplet Belief Functions. Unknown Host Publication. Vol. 8793 2014. pp. 234-245
@inproceedings{44f8f8501dc641c0bdb301e30824fe2a,
title = "Sentiment Classification by Combining Triplet Belief Functions",
abstract = "Sentiment analysis is an emerging technique that caters for semantic orientation and opinion mining. It is increasingly used to anal- yse online product reviews for identifying customers’ opinions and atti- tudes to products or services in order to improve business performance of companies. This paper presents an innovative approach to combining outputs of sentiment classifiers under the framework of belief functions. The approach is composed of the formulation of outputs of sentiment classifiers in the triplet structure and adoption of its formulas to combin- ing simple support functions derived from triplet functions by evidential combination rules. The empirical studies have been conducted on the performance of sentiment classification individually and in combination, the experimental results show that the best combined classifiers made by these combination rules outperform the best individual classifiers over the MP3 and Movie-Review datasets.",
keywords = "Sentiment analysis, opinion mining, triplet belief functions and combination rules",
author = "Yaxin Bi and Maurice Mulvenna and Anna Jurek",
note = "Reference text: 1. Feldnan, R.: Techniques and Applications for Sentiment Analysis. Communications of the ACM 56(4), 82–89 (2013) 2. Pang, B., Lee, L., Vaithyanathan, S.: Thumbs up? Sentiment Classification using Machine Learning Techniques. In: ACL 2002 Conference on Empirical methods in Natural Language Processing, vol. 10, pp. 79–86. Association for Computational Linguistics, Morristown (2002) 3. Jijkoun, V., de Rijke, M., Weerkamp, W.: Generating Focused Topic-specific Sen- timent Lexicons. In: Annual Meeting of the Association for Computational Lin- guistics (ACL 2010), pp. 585–594 (2010) 4. Hu,M.,Liu,B.:Miningandsummarizingcustomerreviews.In:2004ACMSIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 168–177. ACM, New York (2004) 5. Kim, S., Hovy, E.: Determining the Sentiment of Opinions. In: The 20th Inter- national Conference on Computational Linguistics Association for Computational Linguistics, Morristown, NJ, USA (2004) 6. Blitzer, J., Dredze, M., Pereira, F.: Biographies, Bollywood, Boom-boxes and Blenders: Domain adaptation for sentiment classification. In: ACL 2007 (2007) 7. Li, S., Huang, C., Zong, C.: Multi-domain Sentiment Classification with Classifier Combination. Journal of Computer Science and Technology 26(1), 25–33 (2011) 8. Dredze, M., Crammer, K.: Online Methods for Multi-Domain Learning and Adap- tation. In: EMNLP 2008 (2008) 9. Burns, N., Bi, Y., Wang, H., Anderson, T.: Sentiment Analysis of Customer Re- views: Balanced versus Unbalanced Datasets. In: Ko ̈nig, A., Dengel, A., Hinkel- mann, K., Kise, K., Howlett, R.J., Jain, L.C. (eds.) KES 2011, Part I. LNCS, vol. 6881, pp. 161–170. Springer, Heidelberg (2011) 10. Bi, Y., Guan, J.W., Bell, D.: The combination of multiple classifiers using an evidential approach. Artificial Intelligence 17, 1731–1751 (2008)",
year = "2014",
month = "10",
day = "16",
language = "English",
volume = "8793",
pages = "234--245",
booktitle = "Unknown Host Publication",

}

Bi, Y, Mulvenna, M & Jurek, A 2014, Sentiment Classification by Combining Triplet Belief Functions. in Unknown Host Publication. vol. 8793, pp. 234-245, International Conference on Knowledge Science, Engineering and Management 2014, 16/10/14.

Sentiment Classification by Combining Triplet Belief Functions. / Bi, Yaxin; Mulvenna, Maurice; Jurek, Anna.

Unknown Host Publication. Vol. 8793 2014. p. 234-245.

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

TY - GEN

T1 - Sentiment Classification by Combining Triplet Belief Functions

AU - Bi, Yaxin

AU - Mulvenna, Maurice

AU - Jurek, Anna

N1 - Reference text: 1. Feldnan, R.: Techniques and Applications for Sentiment Analysis. Communications of the ACM 56(4), 82–89 (2013) 2. Pang, B., Lee, L., Vaithyanathan, S.: Thumbs up? Sentiment Classification using Machine Learning Techniques. In: ACL 2002 Conference on Empirical methods in Natural Language Processing, vol. 10, pp. 79–86. Association for Computational Linguistics, Morristown (2002) 3. Jijkoun, V., de Rijke, M., Weerkamp, W.: Generating Focused Topic-specific Sen- timent Lexicons. In: Annual Meeting of the Association for Computational Lin- guistics (ACL 2010), pp. 585–594 (2010) 4. Hu,M.,Liu,B.:Miningandsummarizingcustomerreviews.In:2004ACMSIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 168–177. ACM, New York (2004) 5. Kim, S., Hovy, E.: Determining the Sentiment of Opinions. In: The 20th Inter- national Conference on Computational Linguistics Association for Computational Linguistics, Morristown, NJ, USA (2004) 6. Blitzer, J., Dredze, M., Pereira, F.: Biographies, Bollywood, Boom-boxes and Blenders: Domain adaptation for sentiment classification. In: ACL 2007 (2007) 7. Li, S., Huang, C., Zong, C.: Multi-domain Sentiment Classification with Classifier Combination. Journal of Computer Science and Technology 26(1), 25–33 (2011) 8. Dredze, M., Crammer, K.: Online Methods for Multi-Domain Learning and Adap- tation. In: EMNLP 2008 (2008) 9. Burns, N., Bi, Y., Wang, H., Anderson, T.: Sentiment Analysis of Customer Re- views: Balanced versus Unbalanced Datasets. In: Ko ̈nig, A., Dengel, A., Hinkel- mann, K., Kise, K., Howlett, R.J., Jain, L.C. (eds.) KES 2011, Part I. LNCS, vol. 6881, pp. 161–170. Springer, Heidelberg (2011) 10. Bi, Y., Guan, J.W., Bell, D.: The combination of multiple classifiers using an evidential approach. Artificial Intelligence 17, 1731–1751 (2008)

PY - 2014/10/16

Y1 - 2014/10/16

N2 - Sentiment analysis is an emerging technique that caters for semantic orientation and opinion mining. It is increasingly used to anal- yse online product reviews for identifying customers’ opinions and atti- tudes to products or services in order to improve business performance of companies. This paper presents an innovative approach to combining outputs of sentiment classifiers under the framework of belief functions. The approach is composed of the formulation of outputs of sentiment classifiers in the triplet structure and adoption of its formulas to combin- ing simple support functions derived from triplet functions by evidential combination rules. The empirical studies have been conducted on the performance of sentiment classification individually and in combination, the experimental results show that the best combined classifiers made by these combination rules outperform the best individual classifiers over the MP3 and Movie-Review datasets.

AB - Sentiment analysis is an emerging technique that caters for semantic orientation and opinion mining. It is increasingly used to anal- yse online product reviews for identifying customers’ opinions and atti- tudes to products or services in order to improve business performance of companies. This paper presents an innovative approach to combining outputs of sentiment classifiers under the framework of belief functions. The approach is composed of the formulation of outputs of sentiment classifiers in the triplet structure and adoption of its formulas to combin- ing simple support functions derived from triplet functions by evidential combination rules. The empirical studies have been conducted on the performance of sentiment classification individually and in combination, the experimental results show that the best combined classifiers made by these combination rules outperform the best individual classifiers over the MP3 and Movie-Review datasets.

KW - Sentiment analysis

KW - opinion mining

KW - triplet belief functions and combination rules

M3 - Conference contribution

VL - 8793

SP - 234

EP - 245

BT - Unknown Host Publication

ER -

Bi Y, Mulvenna M, Jurek A. Sentiment Classification by Combining Triplet Belief Functions. In Unknown Host Publication. Vol. 8793. 2014. p. 234-245