deepSA at SemEval-2017 Task 4: Interpolated Deep Neural Networks for Sentiment Analysis in Twitter

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 616–620, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics System Implementation for SemEval-2017 Task 4 Subtask A Based on Interpolated Deep Neural Networks 1Tzu-Hsuan Yang, 2Tzu-Hsuan Tseng, and 3Chia-Ping Chen Department of Computer Science and Engineering National Sun Yat-sen University Kaohsiung, Taiwan 1m043040003@student.nsysu.edu.tw 2m043040013@student.nsysu.edu.tw 3cpchen@cse.nsysu.edu.tw Abstract In this paper, we describe our system implementation for sentiment analysis in Twitter. This system combines two models based on deep neural networks, namely a convolutional neural network (CNN) and a long short-term memory (LSTM) re- current neural network, through interpola- tion. Distributed representation of words as vectors are input to the system, and the output is a sentiment class. The neural net- work models are trained exclusively with the data sets provided by the organizers of SemEval-2017 Task 4 Subtask A. Overall, this system has achieved 0.618 for the av- erage recall rate, 0.587 for the average F1 score, and 0.618 for accuracy. 1 Introduction Analysis of digital content created and spread in social networks are becoming instrumental in pub- lic affairs. Twitter is one of the popular social networks, so there are more and more researches on Twitter recently, including sentiment analysis, which predicts the polarity of a message. A message submitted to Twitter is called a tweet. Millions of tweets are created every hour, expressing users’ views or emotions towards all sorts of topics. Different from a document or an article, a tweet is limited in length to 140 charac- ters. In addition, tweets are often colloquial and may contain emotional symbols called emoticons. For sentiment analysis, deep learning-based ap- proaches have performed well in recent years. For example, convolution neural networks (CNN) with word embeddings have been implemented for text classification (Kim, 2014), and have achieved state-of-the-art results in SemEval 2015 (Severyn and Moschitti, 2015). In this paper, we describe our system for SemEval-2017 Task 4 Subtask A for message po- larity classification (Rosenthal et al., 2017). classifies the sentiment of a tweet as positive, neu-It tral, or negative. Our system combines a CNN and a recurrent neural network (RNN) based on long short-term memory (LSTM) cells. We use word embeddings in both models and interpolate them. Our submission achieved 0.618 for average recall, which ranked 19th out of 39 participating teams for subtask A. This paper is organized as follows. In Section 2, we review previous studies on sentiment analysis in Twitter. In Section 3, we describe data, pre- processing steps, model architectures, and tools used in developing our system. In Section 4, we present the evaluation results along with our com- ments. In Section 5, we draw conclusion and dis- cuss future works. 2 Related Works In this section, we briefly review the research works of sentiment analysis in Twitter based on deep neural networks. A one-layer convolu- tion neural network with embeddings can achieve high performance on sentiment analysis (Kim, 2014). In SemEval 2016, quite a few submissions were based on neural …

Content truncated to 500 words