ELiRF-UPV at SemEval-2018 Tasks 1 and 3: Affect and Irony Detection in Tweets

Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018), pages 565–569 New Orleans, Louisiana, June 5–6, 2018. ©2018 Association for Computational Linguistics ELiRF-UPV at SemEval-2018 Tasks 1 and 3: Affect and Irony Detection in Tweets Jos´e- ´Angel Gonz´alez, Llu´ıs-F. Hurtado, Ferran Pla Departament de Sistemes Inform`atics i Computaci´o Universitat Polit`ecnica de Val`encia Cam´ı de Vera, sn 46022, Val`encia {jogonba2, lhurtado, fpla}@dsic.upv.es Abstract This paper describes the participation of ELiRF-UPV team at tasks 1 and 3 of Semeval- 2018. We present a deep learning based sys- tem that assembles Convolutional Neural Net- works and Long Short-Term Memory neu- ral networks. This system has been used with slight modifications for the two tasks ad- dressed both for English and Spanish. Finally, the results obtained in the competition are re- ported and discussed. 1 Introduction The study of figurative language and affective in- formation expressed in texts is of great interest in sentiment analysis applications because they can change the polarity of a message. The objective of tasks 1 and 3 of Semeval 2018 is the study of these phenomena on Twitter. Task 1 (Mohammad et al., 2018) is related to Affect in Tweets. Systems have to automatically determine the intensity of emotions and intensity of sentiment or valence of the tweeters from their tweets. The task is divided in five subtasks: emo- tion intensity regression (EI-Reg), emotion inten- sity ordinal classification (EI-Oc), sentiment in- tensity regression (V-Reg), sentiment analysis or- dinal classification (V-Oc) and emotion classifica- tion (E-C). Task 3 (Van Hee et al., 2018) addresses the problem of Irony detection in English Tweets. It consists of two subtasks. The first subtask is a two-class (or binary) classification task where the system has to predict whether a tweet is ironic or not. The second subtask is a multiclass classifica- tion task where the system has to predict one out of four labels describing i) verbal irony realized through a polarity contrast, ii) verbal irony without such a polarity contrast (i.e., other verbal irony), iii) descriptions of situational irony, iv) non-irony. This paper describes the main characteristics of the developed system by the ELiRF-UPV team for tasks 1 and 3. We addressed all subtasks of task 1 both for English and Spanish, and all subtasks of task 3. 2 Data Preprocessing In this work we have taken into account different aspects when preprocessing the tweets. First we removed the accents and converted all the text to lowercase. In general, emoticons, web links, hash- tags, numbers, and user mentions, were substi- tuted by generic tokens. For instance, “#hashtag” → “hashtag”, � → “Slightly Smiling Face”, etc. After that, we used TweetMotif (Krieger and Ahn, 2010) as tweet tokenizer, moreover we adapted it to work with Spanish tweets. 3 Resources On the one hand, for English, we used the follow- ing polarity/emotion lexicons: AFFIN (Nielsen, 2011), Bing Liu’s Opinion Lexicon (Hu and Liu, 2004), MPQA (Wilson et al., 2005), Sentiment140 (Go et al., 2009), SentiWordnet (Baccianella et al., 2010), NRC Emotion Lexicon (Mohammad and …

Content truncated to 500 words