The wealth of all human created data is in the form of text and being able to detect emotion
in text and classify them provides great insight into how people express themselves on
various topics. We consider the emotion classication of text by resolving the strength of
association of certain parts of speech and n-grams of those parts of speech to an emotion,
e.g. how strongly the noun `dark' is associated to the emotions happy, fear, sad, surprise,
and anger. Using the method of emotion hash tags and seed terms we collected a corpus
from a micro blogging site, twitter. Twitter is an extremely popular communications tool
and is a great resource for emotional data. Using our method we are able to identify
certain terms which can be related to an emotion and use the presence of these terms
as a feature. We built a classier that can categorise a tweet into one of ve emotion
categories. The contribution of this paper is a method which can classify tweets which
express emotion and are highly implicit, having no emotion keywords. Previously these
tweets may be eliminated from the training sets and test sets or be labeled as neutral, we
aim to classify them into an emotion category.

The wealth of all human created data is in the form of text and being able to detect emotion
in text and classify them provides great insight into how people express themselves on
various topics. We consider the emotion classication of text by resolving the strength of
association of certain parts of speech and n-grams of those parts of speech to an emotion,
e.g. how strongly the noun `dark' is associated to the emotions happy, fear, sad, surprise,
and anger. Using the method of emotion hash tags and seed terms we collected a corpus
from a micro blogging site, twitter. Twitter is an extremely popular communications tool
and is a great resource for emotional data. Using our method we are able to identify
certain terms which can be related to an emotion and use the presence of these terms
as a feature. We built a classier that can categorise a tweet into one of ve emotion
categories. The contribution of this paper is a method which can classify tweets which
express emotion and are highly implicit, having no emotion keywords. Previously these
tweets may be eliminated from the training sets and test sets or be labeled as neutral, we
aim to classify them into an emotion category.

Summary 1
Acknowledgments 2
List of Tables 4
List of Figures 5
1 Introduction 6
2 Related Work 9
3 Methodology 11
3.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.1 Feature Explanation . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.2 Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 Emotion classication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 Results 18
5 Conclusion 22
3
List of Tables
3.1 Emotion hash tag words with sample tweets . . . . . . . . . . . . . . . . . 12
3.2 Seed words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3 Emotion break down of test data set . . . . . . . . . . . . . . . . . . . . . 13
3.4 TF-ICF score of terms per emotion category . . . . . . . . . . . . . . . . . 16
4.1 classication (implicit tweets/2 features) . . . . . . . . . . . . . . . . . . . 20
4.2 classication (all tweets/3 features) . . . . . . . . . . . . . . . . . . . . . . 21
4.3 classication (all tweets/ 2 features) . . . . . . . . . . . . . . . . . . . . . . 21
4
List of Figures
3.1 Process Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.1 Fear noun list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.2 Happy noun list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.3 Angry noun list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.4 Surpise noun list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.5 Sad noun list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

[1] A. Agarwal, B. Xie, I. Vovsha, O. Rambow, and R. Passonneau. Sentiment analysis of
twitter data. In Proceedings of the Workshop on Languages in Social Media, LSM '11,
pages 30{38, Stroudsburg, PA, USA, 2011. Association for Computational Linguistics.
[2] C. O. Alm, D. Roth, and R. Sproat. Emotions from text: machine learning for
text-based emotion prediction. In Proceedings of the conference on Human Language
Technology and Empirical Methods in Natural Language Processing, HLT '05, pages
579{586, Stroudsburg, PA, USA, 2005. Association for Computational Linguistics.
[3] S. Aman and S. Szpakowicz. Identifying expressions of emotion in text. In Text,
Speech and Dialogue, pages 196{205. Springer, 2007.
[4] S. Bhattacharyya and P. Bhattacharyya. Emotion analysis of internet chat. In Pro-
ceedings of ICON-2008: 6th International Conference on Natural Language Process-
ing. Macmillan Publishers, India., 2008.
[5] A. Bifet and E. Frank. Sentiment knowledge discovery in twitter streaming data. In
Discovery Science, pages 1{15. Springer, 2010.
[6] D. Davidov and A. Rappoport. Ecient unsupervised discovery of word categories
using symmetric patterns and high frequency words. In Proceedings of the 21st In-
ternational Conference on Computational Linguistics and the 44th annual meeting of
the Association for Computational Linguistics, ACL-44, pages 297{304, Stroudsburg,
PA, USA, 2006. Association for Computational Linguistics.
[7] D. Davidov, O. Tsur, and A. Rappoport. Enhanced sentiment learning using twitter
hashtags and smileys. In Proceedings of the 23rd International Conference on Com-
putational Linguistics: Posters, COLING '10, pages 241{249, Stroudsburg, PA, USA,
2010. Association for Computational Linguistics.
[8] P. Ekman. Universals and cultural dierences in facial expressions of emotion. In
Nebraska symposium on motivation. University of Nebraska Press, 1971.
[9] P. Ekman et al. Facial expression and emotion. American Psychologist, 48:384{384,
1993.
24
[10] A. Go, R. Bhayani, and L. Huang. Twitter sentiment classication using distant
supervision. CS224N Project Report, Stanford, pages 1{12, 2009.
[11] P. Melville, W. Gryc, and R. D. Lawrence. Sentiment analysis of blogs by combining
lexical knowledge with text classication. In Proceedings of the 15th ACM SIGKDD
international conference on Knowledge discovery and data mining, KDD '09, pages
1275{1284, New York, NY, USA, 2009. ACM.
[12] S. Mohammad. Portable features for classifying emotional text. In Proceedings of
the 2012 Conference of the North American Chapter of the Association for Computa-
tional Linguistics: Human Language Technologies, NAACL HLT '12, pages 587{591,
Stroudsburg, PA, USA, 2012. Association for Computational Linguistics.
[13] S. M. Mohammad. #emotional tweets. In Proceedings of the First Joint Conference
on Lexical and Computational Semantics - Volume 1: Proceedings of the main con-
ference and the shared task, and Volume 2: Proceedings of the Sixth International
Workshop on Semantic Evaluation, SemEval '12, pages 246{255, Stroudsburg, PA,
USA, 2012. Association for Computational Linguistics.
[14] A. Pak and P. Paroubek. Twitter as a corpus for sentiment analysis and opinion
mining. In Proceedings of LREC, volume 2010, 2010.
[15] B. Pang, L. Lee, and S. Vaithyanathan. Thumbs up?: sentiment classication using
machine learning techniques. In Proceedings of the ACL-02 conference on Empiri-
cal methods in natural language processing - Volume 10, EMNLP '02, pages 79{86,
Stroudsburg, PA, USA, 2002. Association for Computational Linguistics.
[16] J. Read. Using emoticons to reduce dependency in machine learning techniques
for sentiment classication. In Proceedings of the ACL Student Research Workshop,
ACLstudent '05, pages 43{48, Stroudsburg, PA, USA, 2005. Association for Computational
Linguistics.
[17] K. Roberts, M. A. Roach, J. Johnson, J. Guthrie, and S. M. Harabagiu. Empatweet:
Annotating and detecting emotions on twitter. In Proceedings of the LREC, volume
12, pages 3806{3813, 2012.
[18] G. A. Rodrigues Barbosa, I. S. Silva, M. Zaki, W. Meira, Jr., R. O. Prates, and
A. Veloso. Characterizing the eectiveness of twitter hashtags to detect and track
online population sentiment. In CHI '12 Extended Abstracts on Human Factors in
Computing Systems, CHI EA '12, pages 2621{2626, New York, NY, USA, 2012. ACM.
[19] C. Strapparava and R. Mihalcea. Semeval-2007 task 14: aective text. In Proceedings
of the 4th International Workshop on Semantic Evaluations, SemEval '07, pages 70{
74, Stroudsburg, PA, USA, 2007. Association for Computational Linguistics.
25
[20] C. Strapparava and R. Mihalcea. Learning to identify emotions in text. In Proceedings
of the 2008 ACM symposium on Applied computing, SAC '08, pages 1556{1560, New
York, NY, USA, 2008. ACM.
[21] W. Wang, L. Chen, K. Thirunarayan, and A. Sheth. Harnessing twitter 'big data' for
automatic emotion identication. In Privacy, Security, Risk and Trust (PASSAT),
2012 International Conference on and 2012 International Confernece on Social Com-
puting (SocialCom), pages 587{592, 2012.