TopicThunder at SemEval-2017 Task 4: Sentiment Classification Using a Convolutional Neural Network with Distant Supervision

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 766–770, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics TopicThunder at SemEval-2017 Task 4: Sentiment Classification Using a Convolutional Neural Network with Distant Supervision Simon M¨uller Zurich University of Applied Sciences Switzerland muellsi3@students.zhaw.ch Tobias Huonder Zurich University of Applied Sciences Switzerland huondtob@students.zhaw.ch Jan Deriu Zurich University of Applied Sciences Switzerland deri@zhaw.ch Mark Cieliebak Zurich University of Applied Sciences Switzerland ciel@zhaw.ch Abstract In this paper, we propose a classifier for predicting topic-specific sentiments of En- glish Twitter messages. Our method is based on a 2-layer CNN. With a distant su- pervised phase we leverage a large amount of weakly-labelled training data. Our sys- tem was evaluated on the data provided by the SemEval-2017 competition in the Topic-Based Message Polarity Classifica- tion subtask, where it ranked 4th place. 1 Introduction The goal of sentiment analysis is to teach the ma- chine the ability to understand emotions and opin- ions expressed in text. There are numerous chal- lenges which make this task very difficult, for in- stance the complexity of natural language which makes use of intricate concepts like irony, sar- casm, and metaphors to name a few. Usually we are not interested in the overall sentiment of a tweet but rather in the sentiment the tweet ex- presses towards a topic of interest. Subtask B in SemEval-2017 (Rosenthal et al., 2017) consists of predicting the sentiment of a tweet towards a given topic as either positive or negative. Convolutional neural networks (CNN) have shown to be very efficient at tackling the task of senti- ment analysis, as they are able to learn features themselves instead of relying on manually crafted ones (Kalchbrenner et al., 2014; Kim, 2014). Our work is based on the system proposed by (De- riu et al., 2016), which in turn is based on (Sev- eryn and Moschitti, 2015). CNNs typically have a large number of parameters which need sufficient data to be trained efficiently. In this work, we leverage a large amount of data to train a multi- layer CNN. The training is based on the follow- ing 3-phase procedure (see Figure 1): (i) creation of word embeddings based on an unsupervised corpus of 200M English tweets, (ii) a distant su- pervised phase using an additional 100M tweets, where the labels are inferred by weak sentiment indicators (e.g. smileys), and (iii) a supervised phase, where the pre-trained network is trained on the provided data. We evaluated the approach on the datasets of SemEval-2017 for the subtask B, where it ranked 4th out of 23 submissions. Figure 1: Overview of the 3-Phase training proce- dure. 766 2 Model Our system is based on the 2-layer CNN proposed by (Deriu et al., 2016), as depicted in Figure 2 and described in details below. We refer to a layer as one convolutional and one pooling layer, thus, a 2- layer network consists of two consecutive pairs of convolutional-pooling layers. Word embeddings. The word embeddings are initialized …

Content truncated to 500 words