ECNU at SemEval-2018 Task 1: Emotion Intensity Prediction Using Effective Features and Machine Learning Models

Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018), pages 231–235 New Orleans, Louisiana, June 5–6, 2018. ©2018 Association for Computational Linguistics ECNU at SemEval-2018 Task 1: Emotion Intensity Prediction Using Effective Features and Machine Learning Models Huimin Xu1, Man Lan1,2∗ ,Yuanbin Wu1,2 1Department of Computer Science and Technology, East China Normal University, Shanghai, P.R.China 2Shanghai Key Laboratory of Multidimensional Information Processing 51174506035@stu.ecnu.edu.cn, {mlan, ybwu}@cs.ecnu.edu.cn Abstract In this paper we describe our systems submit- ted to Semeval 2018 Task 1 “Affect in Tweet” (Mohammad et al., 2018). We participated in all subtasks of English tweets, including emo- tion intensity classification and quantification, valence intensity classification and quantifica- tion. In our systems, we extracted four type- s of features, including linguistic, sentiment lexicon, emotion lexicon and domain-specific features, then fed them to different regressors, finally combined the models to create an en- semble for the better performance. Officially released results showed that our system can be further extended. 1 Introduction The Semeval 2018 Task 1 aims to automatically determine the intensity of emotions of the tweeters from their tweets, including five subtasks. That is, given a tweet and one of the four emotions (anger, fear, joy, sadness), the subtask 1 and 2 are to deter- mine the intensity and classify the tweet into one of the four ordinal classes of intensity of the e- motion respectively. Similarly, the subtask 3 and 4 determine the intensity and classify the tweet into one of seven ordinal classes of intensity of valance. Subtask 5 is a multi-label emotion clas- sification task which classifies the tweets as neu- tral or no emotion or as one, or more, of eleven given emotions (anger, anticipation, disgust, fear, joy, love, optimism, pessimism, sadness, surprise, trust) that best represent the mental state of the tweeter. For each task, training and test dataset- s are divided into English, Arabic, and Spanish tweets. We participated in all subtasks of English tweets. Traditional sentiment classification is a coarse- grained task in sentiment analysis which focuses on sentiment polarity classification of the whole sentence (i.e., positive, negative, neutral, mixed). Semeval 2018 Task 1 subtask 5 takes basic human emotion proposed by Ekman (Ekman, 1999) into consideration, including Anger, Anticipation, Dis- gust, Fear, Joy, Sadness, Surprise, and Trust. The difference between these subtasks lies in the emotion granularity and classification or quan- tification, so in our work, the similar method is adopted for five subtasks. We extracted a rich set of elaborately designed features. In addition to lin- guistic features, sentiment lexicon features and e- motion lexicon features, we also extracted some domain specific features. Also, we conducted a series of experiments on different machine learn- ing algorithms and ensemble methods to obtain the better performing for each subtask. For subask 5, we adopted multiple binary classification and con- structed a model for each emotion. 2 System Description We first performed data preprocessing, then ex- tracted several types of features from tweets and constructed supervised models for this task. 2.1 Data Preprocessing Firstly, …

Content truncated to 500 words