AffecThor at SemEval-2018 Task 1: A cross-linguistic approach to sentiment intensity quantification in tweets

Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018), pages 210–217 New Orleans, Louisiana, June 5–6, 2018. ©2018 Association for Computational Linguistics AffecThor at SemEval-2018 Task 1: A cross-linguistic approach to sentiment intensity quantification in tweets Mostafa Abdou and Artur Kulmizev and Joan Gin´es i Ametll´e {m.abdou, a.kulmizev, j.gines.i.ametlle}@student.rug.nl CLCG, University of Groningen Abstract In this paper we describe our submission to SemEval-2018 Task 1: Affects in Tweets. The model which we present is an ensemble of var- ious neural architectures and gradient boosted trees, and employs three different types of vec- torial tweet representations. Furthermore, our system is language-independent and ranked first in 5 out of the 12 subtasks in which we participated, while achieving competitive re- sults in the remaining ones. Comparatively re- markable performance is observed on both the Arabic and Spanish languages. 1 Introduction The Affects in Tweets shared task (Mohammad et al., 2018) is the second iteration of a task which offers a new approach to Sentiment Analysis - one that concerns itself with emotion and sentiment intensity, rather than simple categorical classifi- cation. The shared task is divided into a set of subtasks, where the aim is to predict the emotion intensity of a predetermined emotion (fear, anger, sadness, joy) or sentiment (valence) intensity of a given set of tweets. Such predictions are either for- mulated as a regression problem where the output is a continuous-valued score in the interval (0, 1), or as ordinal classification into a given number of classes representing intensity. Additionally, each one of the subtasks targets a particular language: English, Arabic or Spanish. In total, we participated in 12 different subtasks and our system achieved the best performance on the test set out of all participants in 5 out of those, ranked second in 3 others, and performed compet- itively in the rest. Moreover, our system can ar- guably be considered the best overall performing system for both Arabic and Spanish1. It should be noted, however, that the shared task includes tradi- tional emotion classification subtasks in which we did not participate. The system described in this paper builds upon a survey of some of the best performing systems from previous related shared tasks (Mohammad and Bravo-Marquez, 2017; Rosenthal et al., 2017). In particular, we draw inspiration from the systems described in (John and Vechtomova, 2017), which makes use of gradient boosted trees for regres- sion; (Goel et al., 2017), which employs an ensem- ble of various neural models; and (Baziotis et al., 2017), which features Long Short Term Memory (LSTM) networks with an attention mechanism. Our work contributes to the aforementioned ap- proaches by further developing a variety of neu- ral architectures, using transfer learning via pre- trained sentence encoders, testing methods of en- sembling neural and non-neural models, and gaug- ing the performance and stability of a regressor across languages. The rest of this paper describes the pipeline of the system used for our submission, which is an ensemble of neural and non-neural models. 2 Data …

Content truncated to 500 words