ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features

Zhihua Zhang; Guoshun Wu; Man Lan

doi:10.18653/v1/s15-2094

ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features

Zhihua Zhang
, Guoshun Wu
, Man Lan^*

^*Corresponding author for this work

School of Computer Science and Technology

East China Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

13 Scopus citations

Abstract

This paper reports our submission to task 10 (Sentiment Analysis on Tweet, SAT) (Rosenthal et al., 2015) in SemEval 2015, which contains five subtasks, i.e., contextual polarity disambiguation (subtask A: expression-level), message polarity classification (subtask B: message-level), topic-based message polarity classification and detecting trends towards a topic (subtask C and D: topic-level), and determining sentiment strength of twitter terms (subtask E: term-level). For the first four subtasks, we built supervised models using traditional features and word embedding features to perform sentiment polarity classification. For subtask E, we first expanded the training data with the aid of external sentiment lexicons and then built a regression model to estimate the sentiment strength. Despite the simplicity of features, our systems rank above the average.

Original language	English
Title of host publication	SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics
Subtitle of host publication	Human Language Technologies, NAACL-HLT 2015 - Proceedings
Editors	Preslav Nakov, Torsten Zesch, Daniel Cer, David Jurgens
Publisher	Association for Computational Linguistics (ACL)
Pages	561-567
Number of pages	7
ISBN (Electronic)	9781941643402
DOIs	https://doi.org/10.18653/v1/s15-2094
State	Published - 2015
Event	9th International Workshop on Semantic Evaluation, SemEval 2015 co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Denver, United States Duration: 4 Jun 2015 → 5 Jun 2015

Publication series

Name	SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings

Conference

Conference	9th International Workshop on Semantic Evaluation, SemEval 2015 co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015
Country/Territory	United States
City	Denver
Period	4/06/15 → 5/06/15

Access to Document

10.18653/v1/s15-2094

Cite this

Zhang, Z., Wu, G., & Lan, M. (2015). ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features. In P. Nakov, T. Zesch, D. Cer, & D. Jurgens (Eds.), SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings (pp. 561-567). (SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/s15-2094

Zhang, Zhihua ; Wu, Guoshun ; Lan, Man. / ECNU : Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features. SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings. editor / Preslav Nakov ; Torsten Zesch ; Daniel Cer ; David Jurgens. Association for Computational Linguistics (ACL), 2015. pp. 561-567 (SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings).

@inproceedings{5c54cfd77be242199cb234a71d1b857b,

title = "ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features",

abstract = "This paper reports our submission to task 10 (Sentiment Analysis on Tweet, SAT) (Rosenthal et al., 2015) in SemEval 2015, which contains five subtasks, i.e., contextual polarity disambiguation (subtask A: expression-level), message polarity classification (subtask B: message-level), topic-based message polarity classification and detecting trends towards a topic (subtask C and D: topic-level), and determining sentiment strength of twitter terms (subtask E: term-level). For the first four subtasks, we built supervised models using traditional features and word embedding features to perform sentiment polarity classification. For subtask E, we first expanded the training data with the aid of external sentiment lexicons and then built a regression model to estimate the sentiment strength. Despite the simplicity of features, our systems rank above the average.",

author = "Zhihua Zhang and Guoshun Wu and Man Lan",

note = "Publisher Copyright: {\textcopyright} 2015 Association for Computational Linguistics; 9th International Workshop on Semantic Evaluation, SemEval 2015 co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 ; Conference date: 04-06-2015 Through 05-06-2015",

year = "2015",

doi = "10.18653/v1/s15-2094",

language = "英语",

series = "SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings",

publisher = "Association for Computational Linguistics (ACL)",

pages = "561--567",

editor = "Preslav Nakov and Torsten Zesch and Daniel Cer and David Jurgens",

booktitle = "SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics",

address = "澳大利亚",

}

Zhang, Z, Wu, G & Lan, M 2015, ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features. in P Nakov, T Zesch, D Cer & D Jurgens (eds), SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings. SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings, Association for Computational Linguistics (ACL), pp. 561-567, 9th International Workshop on Semantic Evaluation, SemEval 2015 co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015, Denver, United States, 4/06/15. https://doi.org/10.18653/v1/s15-2094

ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features. / Zhang, Zhihua; Wu, Guoshun; Lan, Man.
SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings. ed. / Preslav Nakov; Torsten Zesch; Daniel Cer; David Jurgens. Association for Computational Linguistics (ACL), 2015. p. 561-567 (SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - ECNU

T2 - 9th International Workshop on Semantic Evaluation, SemEval 2015 co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015

AU - Zhang, Zhihua

AU - Wu, Guoshun

AU - Lan, Man

PY - 2015

Y1 - 2015

N2 - This paper reports our submission to task 10 (Sentiment Analysis on Tweet, SAT) (Rosenthal et al., 2015) in SemEval 2015, which contains five subtasks, i.e., contextual polarity disambiguation (subtask A: expression-level), message polarity classification (subtask B: message-level), topic-based message polarity classification and detecting trends towards a topic (subtask C and D: topic-level), and determining sentiment strength of twitter terms (subtask E: term-level). For the first four subtasks, we built supervised models using traditional features and word embedding features to perform sentiment polarity classification. For subtask E, we first expanded the training data with the aid of external sentiment lexicons and then built a regression model to estimate the sentiment strength. Despite the simplicity of features, our systems rank above the average.

AB - This paper reports our submission to task 10 (Sentiment Analysis on Tweet, SAT) (Rosenthal et al., 2015) in SemEval 2015, which contains five subtasks, i.e., contextual polarity disambiguation (subtask A: expression-level), message polarity classification (subtask B: message-level), topic-based message polarity classification and detecting trends towards a topic (subtask C and D: topic-level), and determining sentiment strength of twitter terms (subtask E: term-level). For the first four subtasks, we built supervised models using traditional features and word embedding features to perform sentiment polarity classification. For subtask E, we first expanded the training data with the aid of external sentiment lexicons and then built a regression model to estimate the sentiment strength. Despite the simplicity of features, our systems rank above the average.

UR - https://www.scopus.com/pages/publications/85122015485

U2 - 10.18653/v1/s15-2094

DO - 10.18653/v1/s15-2094

M3 - 会议稿件

AN - SCOPUS:85122015485

T3 - SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings

SP - 561

EP - 567

BT - SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics

A2 - Nakov, Preslav

A2 - Zesch, Torsten

A2 - Cer, Daniel

A2 - Jurgens, David

PB - Association for Computational Linguistics (ACL)

Y2 - 4 June 2015 through 5 June 2015

ER -

Zhang Z, Wu G, Lan M. ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features. In Nakov P, Zesch T, Cer D, Jurgens D, editors, SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings. Association for Computational Linguistics (ACL). 2015. p. 561-567. (SemEval 2015 - 9th International Workshop on Semantic Evaluation, co-located with the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2015 - Proceedings). doi: 10.18653/v1/s15-2094

ECNU: Multi-level Sentiment Analysis on Twitter Using Traditional Linguistic Features and Word Embedding Features

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this