Unsupervised Emotion Detection from Text Using Semantic and Syntactic Relations

Agrawal, Ameeta; An, Aijun

doi:10.1109/wi-iat.2012.170

Cited by 122 publications

(55 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PMI, SVM and RESOLVE all used the same corpus. Note that NAVA words (noun, adjective, verb and adverb) are the major sentiment-bearing terms (Agrawal and An, 2012). Hence for comparison with the feature set of extracted patterns we selected NAVA words as the additional feature set.…”

Section: Comparison To Mi/ml Methodsmentioning

confidence: 99%

“…Related work typically attempts this using classifiers. These classifiers are trained with features such as n-grams (Tokuhisa et al, 2008), word-level pointwise mutual information (PMI) values (Agrawal et al, 2012;Bullinaria et al, 2007;and Church et al, 1990) or a combination of word POS and sentence dependency relations (Ghazi et al, 2012). The remained works of emotion classification in above mentioned research to deal with emotions aroused by events inspires us to relate events to emotion words in RESOLVE.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Embarrassed or Awkward? Ranking Emotion Synonyms for ESL Learners' Appropriate Wording

Chen

2015

Proceedings of the Tenth Workshop on Innovative Use of NLP for Building Educational Applications

View full text Add to dashboard Cite

We introduce a novel framework based on the probabilistic model for emotion wording assistance. The example sentences from the online dictionary, Vocabulary.com are utilized as the training data; and the writings in a designed ESL's writing task are the testing corpus. The emotion events are captured by extracting patterns of the example sentences. Our approach learns the joint probability of contextual emotion events and the emotion words from the training corpus. After extracting patterns in the testing corpus, we then aggregate their probabilities to suggest the emotion word that describes the ESL's context most appropriately. We evaluate the proposed approach by the NDCG@5 of the suggested words for the writings in the testing corpus. The experiment result shows our approach can more appropriately suggest the emotion words compared to SVM, PMI and two representative on-line reference tools, PIGAI and Thesaurus.com.

show abstract

Section: Comparison To Mi/ml Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Embarrassed or Awkward? Ranking Emotion Synonyms for ESL Learners' Appropriate Wording

Chen

2015

Proceedings of the Tenth Workshop on Innovative Use of NLP for Building Educational Applications

View full text Add to dashboard Cite

show abstract

“…Semi-supervised algorithms have also been utilized, especially for automatic generation of sentiment lexicons [16]. Some studies have opted for unsupervised methods exploiting a variety of semantic and syntactic techniques and lexicons [17], [18].…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Approach for Emotion Detection in Support of Affective Interaction

Gievska

Koroveshovski

Chavdarova

2014

2014 IEEE International Conference on Data Mining Workshop

View full text Add to dashboard Cite

Affective interaction is a new emerging area of interest for interaction designers. This research explores the potential of our hybrid approach that relies on both, lexical and machine learning techniques for detection of Ekman's six emotional categories in user's text. The initial results of the performance evaluation of the proposed hybrid approach are encouraging and comparable to related research. A demonstrative mobile application that employs the proposed approach was developed to engage the users in a dialogue that solicits their reflections on various daily events and provides appropriate affective responses.

show abstract

“…This proposal used the Ekman's basic emotions. (Agrawal 2012) proposes a novel unsupervised contextbased approach based on a methodology that does not depend on any existing affect lexicon, thereby their model is flexible enough to classify sentences beyond Ekman's model of six basic emotions. (Calvo 2013) presents different categorical approaches based on Vector Space Model (VSM) with three dimensionality reduction techniques: Latent Semantic Analysis (LSA), Probabilistic Latent Semantic Analysis (PLSA) and Non-negative Matrix Factorization (NMF).…”

Section: Machine Learning-based Approachesmentioning

confidence: 99%

Proceedings of the Workshop on Natural Language Processing in the 5th Information Systems Research Working Days (JISIC)

2014

View full text Add to dashboard Cite

(NAACL) has endorsed the event.The lecturers have been invited to write papers on all aspects of computational approaches to Natural Language Processing. The papers received have been revised and prepared to compose this issue.We thank all lecturers and participants who have contributed and made this publication possible. We also appreciate and give a special thanks to the support from the Association for Computational Linguistics (ACL). Conference Program Workshop NLP Towards the production of a corpus for content analysis of references in scientific literatureIng. MSc. M. Hernández. Escuela Politécnica Nacional. Ecuador. Bilingual sentence alignment of a parallel corpus by using English as a pivot languageDr. Josafá Pontes. Escuela Politécnica Nacional. Ecuador. Language Technologies for Suicide Prevention in Social MediaDr. José Gómez. University of Alicante. Spain. Since, at the moment there is not a goldstandard annotated corpus for this objective, it is necessary to build one, to allow generation and testing of automatic systems for classifying the purpose or function of a citation referenced in an article. The development of this kind of corpus is subject to two conditions: the first one is to present a clear and unambiguous classification scheme. The second one is to secure an initial manual process of labeling to reach a sufficient inter-coder agreement among annotators to validate the annotation scheme and to be able to reproduce it even with coders who do not know in depth the topic of the analyzed articles. This paper proposes and validates a methodology for corpus annotation for citation classification in scientific literature that facilitate annotation and produces substantial interannotator agreement. Sentiment analysis in the Web

show abstract

Unsupervised Emotion Detection from Text Using Semantic and Syntactic Relations

Cited by 122 publications

References 12 publications

Embarrassed or Awkward? Ranking Emotion Synonyms for ESL Learners' Appropriate Wording

Embarrassed or Awkward? Ranking Emotion Synonyms for ESL Learners' Appropriate Wording

A Hybrid Approach for Emotion Detection in Support of Affective Interaction

Proceedings of the Workshop on Natural Language Processing in the 5th Information Systems Research Working Days (JISIC)

Contact Info

Product

Resources

About