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

By Keith Cortis, André Freitas, Tobias Daudert, M. Hürlimann, Manel Zarrouk, Siegfried Handschuh, Brian Davis. 2017

199 citations recorded by the source corpus.

This paper discusses the “Fine-Grained Sentiment Analysis on Financial Microblogs and News” task as part of SemEval-2017, specifically under the “Detecting sentiment, humour, and truth” theme. This task contains two tracks, where the first one concerns Microblog messages and the second one covers News Statements and Headlines. The main goal behind both tracks was to predict the sentiment score for each of the mentioned companies/stocks. The sentiment scores for each text instance adopted floating point values in the range of -1 (very negative/bearish) to 1 (very positive/bullish), with 0 designating neutral sentiment. This task attracted a total of 32 participants, with 25 participating in Track 1 and 29 in Track 2.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Mathias Kraus
2. Pushpak Bhattacharyya
3. John Gerard Breslin
4. Asif Ekbal
5. Paul Rayson
6. Abhishek Kumar
7. Stefan Feuerriegel
8. Deepanway Ghosal

In Management Science and Operations Research

1. Mathias Kraus
2. Pushpak Bhattacharyya
3. John Gerard Breslin
4. Stefan Feuerriegel
5. Deepanway Ghosal
6. Guangyuan Piao
7. Dogu Araci
8. M. Ugur Gudelek

Citations in the corpus

• Decision support from financial disclosures with deep neural networks and transfer learning (2017, background): “The SemEval-2017 competition is currently raising awareness in this regard [28].”
• A Multilayer Perceptron based Ensemble Technique for Fine-grained Financial Sentiment Analysis (2017, methodology): “For evaluation purpose we use datasets of SemEval-2017 ‘Fine-Grained Sentiment Analysis on Financial Microblogs and News’ shared task (Keith Cortis and Davis, 2017).”
• Lancaster A at SemEval-2017 Task 5: Evaluation metrics matter: predicting sentiment from financial news headlines (2017, methodology): “…this is what the first version of the results were actually based on, where we were ranked 4th), which then changed by the organisers to their equation as presented in Cortis et al. (2017) (which we term metric 3 and what the second version of the results were based on, where we were ranked 5th).”
• Financial Aspect and Sentiment Predictions with Deep Neural Networks: An Ensemble Approach (2018, methodology): “As we can see from the table, CNN [1, 2, 3] provides the best performance as a single predictor followed by CNN [3, 4, 5] and CNN [5, 6, 7].”