INF-UFRGS at SemEval-2017 Task 5: A Supervised Identification of Sentiment Score in Tweets and Headlines

Abstract

This paper describes a supervised solution for detecting the polarity scores of tweets or headline news in the financial domain, submitted to the SemEval 2017 Fine-Grained Sentiment Analysis on Financial Microblogs and News Task. The premise is that it is possible to understand market reaction over a company stock by measuring the positive/negative sentiment contained in the financial tweets and news headlines, where polarity is measured in a continuous scale ranging from -1.0 (very bearish) to 1.0 (very bullish). Our system receives as input the textual content of tweets or news headlines, together with their ids, stock cashtag or name of target company, and the polarity score gold standard for the training dataset. Our solution retrieves features from these text instances using n-gram, hashtags, sentiment score calculated by a external APIs and others features to train a regression model capable to detect continuous score of these sentiments with precision.

SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogs and News

“…For each instance vector in GS/PS which has a length of 1, the absolute distance between both scores ( 4) is added to the total similarity score (6). s similarity(G, P ) = 1 − |G 0 − P 0 | (4) Tool System scikit-learn 18 Nasim (2017), Moore and Rayson (2017), John and Vechtomova (2017), Cabanski et al (2017), Kumar et al (2017), Symeonidis et al (2017), Kar et al (2017), Jiang et al (2017) word2vec 19 Li (2017), Ghosal et al (2017), Kumar et al (2017), Saleiro et al (2017) Weka 20 Seyeditabari et al (2017, Zini et al (2017) GloVe 21 Seyeditabari et al (2017), Mansar et al (2017), Rotim et al (2017), Pivovarova et al (2017), Ghosal et al (2017), Kumar et al ( 2017) LIBSVM 22 Rotim et al (2017) LIBLINEAR 23 Rotim et al (2017), Jiang et al (2017) Keras 24 Moore and Rayson (2017), Ghosal et al (2017) XGBoost 25 John and Vechtomova (2017), Jiang et al ( 2017) gensim 26 John and Vechtomova (2017), Cabanski et al (2017) TensorFlow 27 John and Vechtomova (2017), Pivovarova et al (2017), Cabanski et al (2017) Figure 8: Tools used by systems in both tracks…”