TSA-INF at SemEval-2017 Task 4: An Ensemble of Deep Learning Architectures Including Lexicon Features for Twitter Sentiment Analysis

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 802–806, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics TSA-INF at SemEval-2017 Task 4: An Ensemble of Deep Learning Architectures Including Lexicon Features for Twitter Sentiment Analysis Amit Ajit Deshmane Infosys Limited Pune, Maharashtra 411057, India amitad87@gmail.com Jasper Friedrichs Infosys Limited Newark, CA 94560, USA jasper friedrichs@infosys.com Abstract This paper describes the submission of team TSA-INF to SemEval-2017 Task 4 Subtask A. The submitted system is an ensemble of three varying deep learning architectures for sentiment analysis. The core of the architecture is a convolutional neural network that performs well on text classification as is. The second subsystem is a gated recurrent neural network imple- mentation. Additionally, the third system integrates opinion lexicons directly into a convolution neural network architecture. The resulting ensemble of the three archi- tectures achieved a top ten ranking with a macro-averaged recall of 64.3%. Ad- ditional results comparing variations of the submitted system are not conclusive enough to determine a best architecture, but serve as a benchmark for further im- plementations. 1 Introduction The SemEval competitions continually offer suit- able dataset and resulting benchmarks for a va- riety of natural language processing tasks. The SemEval-2017 Task 4 Subtask A addresses the po- larity classification task of informal texts (Rosen- thal et al., 2017). Tweets serve as a very accessi- ble sample of the abundant social media content. Submitted systems must classify tweets into the categories of negative, positive and neutral opin- ion. Submitted results are compared over macro- averaged recall. In recent benchmarks across this task, deep learning implementations achieved top re- sults (Nakov et al., 2016). We seek to combine three varying deep learning approaches in an ensemble. Conventional methods seem to become obsolete since convolutional neural networks (CNN) have first shown state-of-the-art results in sentiment analysis (Kim, 2014). SemEval has since seen successful results by similar models (Severyn and Moschitti, 2015) as well as ensembles of CNNs (Deriu et al., 2016). Long term short term recurrent neural networks (LSTM) (Hochreiter and Schmidhuber, 1997) have also been applied successfully in ensemble with a CNN (Xu et al., 2016). As an alternative to LSTMs gated recurrent neural networks (GRNN) have been shown to be competitive in other domains (Chung et al., 2014). These models are well suited to model sequential data and were successfully applied for sentiment analysis of larger documents (Yang et al., 2016). The core contribution of a recent non deep learning system to win this task (Kiritchenko et al., 2014) back to back in 2013 and 2014, was the integration of opinion lexicons into a support vector machine system. Opinion lexicons have since then also been integrated into CNN architectures (Rouvier and Favre, 2016; Shin et al., 2016). In this work we combine these diverse architectures. We use a CNN, thoroughly optimized for text classification, as the foundation of our ensemble approach. We add a lexicon integrated CNN to take advantage of lexicon features. In order to …

Content truncated to 500 words