Abstract
In this paper, we perform modality prediction in financial dialogue. To this end, we introduce a new dataset and develop a binary classifier to detect strong or weak modal answers depending on surface, lexical, and semantic representations of the preceding question and financial features. To do so, we contrast different algorithms, feature categories, and fusion methods. Perhaps counter-intuitively, our results indicate that the strongest features for the given task are financial uncertainty measures such as market and individual firm risk.- Anthology ID:
- 2020.fnp-1.35
- Volume:
- Proceedings of the 1st Joint Workshop on Financial Narrative Processing and MultiLing Financial Summarisation
- Month:
- December
- Year:
- 2020
- Address:
- Barcelona, Spain (Online)
- Editors:
- Dr Mahmoud El-Haj, Dr Vasiliki Athanasakou, Dr Sira Ferradans, Dr Catherine Salzedo, Dr Ans Elhag, Dr Houda Bouamor, Dr Marina Litvak, Dr Paul Rayson, Dr George Giannakopoulos, Nikiforos Pittaras
- Venue:
- FNP
- SIG:
- Publisher:
- COLING
- Note:
- Pages:
- 226–234
- Language:
- URL:
- https://aclanthology.org/2020.fnp-1.35
- DOI:
- Cite (ACL):
- Kilian Theil and Heiner Stuckenschmidt. 2020. Predicting Modality in Financial Dialogue. In Proceedings of the 1st Joint Workshop on Financial Narrative Processing and MultiLing Financial Summarisation, pages 226–234, Barcelona, Spain (Online). COLING.
- Cite (Informal):
- Predicting Modality in Financial Dialogue (Theil & Stuckenschmidt, FNP 2020)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2020.fnp-1.35.pdf