YZU-NLP Team at SemEval-2016 Task 4: Ordinal Sentiment Classification Using a Recurrent Convolutional Network

Proceedings of SemEval-2016, pages 251–255, San Diego, California, June 16-17, 2016. c⃝2016 Association for Computational Linguistics YZU-NLP Team at SemEval-2016 Task 4: Ordinal Sentiment Classifi- cation Using a Recurrent Convolutional Network Yunchao He2,3,4, Liang-Chih Yu1,3, Chin-Sheng Yang1,3, K. Robert Lai2,3, Weiyi Liu4 1Department of Information Management, Yuan Ze University, Taiwan 2Department of Computer Science & Engineering, Yuan Ze University, Taiwan 3Innovation Center for Big Data and Digital Convergence, Yuan Ze University, Taiwan 4School of Information Science and Engineering, Yunnan University, Yunnan, P.R. China Abstract Sentiment analysis of tweets has attracted considerable attention recently for potential use in commercial and public sector applica- tions. Typical sentiment analysis classifies the sentiment of sentences into several discrete classes (e.g., positive and negative). The aim of Task 4 subtask C of SemEval-2016 is to classify the sentiment of tweets into an ordinal five-point scale. In this paper, we present a system that uses word embeddings and recur- rent convolutional networks to complete the competition task. The word embeddings pro- vide a continuous vector representation of words for the recurrent convolutional network to use in building sentence vectors for multi- point classification. The proposed method ranked second among eleven teams in terms of micro-averaged MAE (mean absolute error) and eighth for macro-averaged MAE. 1 Introduction Sentiment analysis seeks to detect and analyze sen- timent within texts. Following the rapid increase of user generated content in the form of social media, sentiment analysis has attracted considerable inter- est. Typical approaches to sentiment analysis clas- sify the sentiment of a sentence into several dis- crete classes such as positive and negative polari- ties, or six basic emotions: anger, happiness, fear, sadness, disgust and surprise (Ekman, 1992). Based on this representation, various techniques have been investigated including supervised learn- ing and lexicon-based approaches. Supervised learning approaches require training data for sen- timent classification (Go et al., 2009; Yu et al., 2009; Saif et al., 2016), while lexicon-based ap- proaches do not require training data but use a sen- timent lexicon to determine the overall sentiment of a sentence (Liu, 2010; Hu et al., 2013). A five-point scale (Nakov et al., 2016) is also a popular way to evaluate sentiment. Many compa- nies, such as Amazon, Google, and Alibaba all use a multi-point scale to evaluate product or APP re- views. Unlike typical classification approaches, ordinal classification can assign different ratings (e.g., very negative, negative, neutral, positive and very positive) according to sentiment strength (Taboada et al., 2011; Li et al., 2011; Yu et al., 2013; Wang and Ester, 2014). Task 4 subtask C of SemEval-2016 seeks to classify the sentiment of tweets into an ordinal five-point scale. This paper presents a system that uses word embeddings (Mikolov et al., 2013) and recurrent convolutional networks to this end. The word embeddings can capture both semantic and syntactic information of words to provide a contin- uous vector representation of those words. These word vectors are then used to build sentence vec- tors through a recurrent convolutional neural net- work. For …

Content truncated to 500 words