RiTUAL-UH at SemEval-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 877–882, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics RiTUAL-UH at SemEval-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks Sudipta Kar∗and Suraj Maharjan∗and Thamar Solorio Department of Computer Science University of Houston Houston, TX 77204-3010 {skar3, smaharjan2}@uh.edu, solorio@cs.uh.edu Abstract In this paper, we present our systems for “SemEval-2017 Task-5 on Fine-Grained Sentiment Analysis on Financial Mi- croblogs and News”. In our system, we combined hand-engineered lexical, senti- ment, and metadata features with the rep- resentations learned from Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Unit (Bi-GRU) having Attention model applied on top. With this architecture we obtained weighted co- sine similarity scores of 0.72 and 0.74 for subtask-1 and subtask-2, respectively. Us- ing the official scoring system, our system ranked in the second place for subtask-2 and in the eighth place for the subtask-1. However, it ranked first in both subtasks when evaluated with an alternate scoring metric1. 1 Introduction Predicting sentiments of financial data has a wide range of applications. The most important ap- plication is being able to predict the ups and downs of the share market as the changes in sen- timents and opinions can change the market dy- namics (Goonatilake and Herath, 2007; Van de Kauter et al., 2015). Stock market related informa- tion is typically found in newspapers (Malo et al., 2013) and people discuss them in social media platforms like Twitter and StockTwits. Positive news has the capability to boost the market by in- creasing optimism among people (Van de Kauter et al., 2015; Schuster, 2003). SemEval-2017 Task ∗Both authors contributed equally. 1http://alt.qcri.org/semeval2017/ task5/data/uploads/description_second_ approach.pdf 5 on ‘Fine-Grained Sentiment Analysis on Finan- cial Microblogs and News’ aims at analyzing the polarity of public sentiments from financial data found in newspapers and social media. In this paper, we describe our systems, which exploit automatically learned representations using deep learning architecture based methods along with hand-engineered features in order to predict the sentiment polarity of financial data. 2 Dataset Task Training Trial Test Subtask-1 1,694 10 799 Subtask-2 1,142 14 491 Table 1: Data distribution for subtask-1 and subtask-2. Table 1 shows the distribution of training, trial, and test data for subtask-1 and subtask-2. For the subtask-1, the financial microblogs and tweets were collected from Twitter2 and StockTwits3 whereas for subtask-2, the financial news head- lines were collected from different financial news sources such as Yahoo Finance4. Each instance was labeled with a floating point value ranging from -1 to +1, indicating the sentiment score. A score of -1 means very negative or bearish whereas, a score of +1 means very positive or bullish. A score of 0 means neutral sentiment. The dataset is noisy and contains URLs, cashtags, digits, usernames, and emoticons. The messages are short with an average number of 13 tokens for the microblog data and 10 tokens for the headlines data. 2https://twitter.com 3https://stocktwits.com 4https://finance.yahoo.com 877 3 Methodology We designed two systems for predicting …

Content truncated to 500 words