EICA at SemEval-2017 Task 4: A Simple Convolutional Neural Network for Topic-based Sentiment Classification

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 737–740, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics EICA at SemEval-2017 Task 4: A Convolutional Neural Network for Topic-based Sentiment Classification Maoquan Wang, Shiyun Chen, Yufei Xie, Jing Ma, Zhao Lu Department of Computer Science and Technology, East China Normal University, Shanghai, P.R.China {maoquanwang,yufeixie,liangma,lzhao}@ica.stc.sh.cn, csyecnu@outlook.com Abstract This paper describes our approach for SemEval-2017 Task 4 - Sentiment Anal- ysis in Twitter (SAT). Its five subtasks are divided into two categories: (1) sentiment classification, i.e., predicting topic-based tweet sentiment polarity, and (2) sentiment quantification, that is, estimating the senti- ment distributions of a set of given tweets. We build a convolutional sentence classifi- cation system for the task of SAT. Official results show that the experimental results of our system are comparative. 1 Introduction With the rapid growth of social media such as Twitter, sentiment classification towards the user generated texts has attracted increasing research interest. The objective of sentiment classification is identifying the sentiment of a text into binary polarity (Positive vs. Negative) or single-label multi-class (e.g., Very positive, Positive, Neutral, Negative, Very negative). Feature representation is one of key points for this kind of classification, which generally falls into two categories: (1) tra- ditional feature engineering (Liu, 2012; Moham- mad et al., 2013), such as sentiment lexicon, n- grams, dependency triple, etc. (2) deep learning methods (Zhao et al., 2015; Yang et al., 2016), which use exquisitely designed neural network to encode input texts and to get text feature represen- tation. Recently, deep learning approaches emerge as powerful computational models for text senti- ment classification, and have achieved new state- of-the-art result in some datasets. SemEval-2017 provides a universal platform for researchers to explore the task of twitter sentiment analysis. In this paper, we explore Task 4 (Rosen- thal et al., 2017), which includes five subtasks: subtask A, B and C are related to the task of sen- timent classification, and subtask D and E are re- lated to sentiment quantification (that is distribu- tions of sentiments). Considering the length lim- itations of tweets, we view the subtasks of SAT as sentence-level sentiment analysis. We design a convolutional neural network for topic-based sen- timent classification. 2 System Description In this section, we describe the neural network ar- chitecture of our system. As shown in Figure 1, our system consists of six layers, an input layer, a convolutional layer, a max-pooling layer, a topic embedding layer, a concatenate layer, and an out- put layer. Input layer. A tweet text can be denoted as a sentence sequence x with n words, x [w1, w2, · · ·, wi, · · ·, wn]. To obtain word vector= of word wi , we look-up word embedding ma- trix E, where e(wi) ∈ Rd, E ∈ R|V |×d, |V | is the vocabulary size. Then, we get an input matrix X = [e(w1); · · ·; e(wn)], where X ∈ Rn×d. Convolution layer. The convolution action has been …

Content truncated to 500 words