Haeffner, Sonja and Hofer, Martin and Nagl, Maximilian and Walterskirchen, Julian (2023) Introducing an Interpretable Deep Learning Approach to Domain-Specific Dictionary Creation: A Use Case for Conflict Prediction. POLITICAL ANALYSIS, 31 (4): PII S10471. pp. 481-499. ISSN 1047-1987, 1476-4989
Full text not available from this repository. (Request a copy)Abstract
Recent advancements in natural language processing (NLP) methods have significantly improved their performance. However, more complex NLP models are more difficult to interpret and computationally expensive. Therefore, we propose an approach to dictionary creation that carefully balances the trade-off between complexity and interpretability. This approach combines a deep neural network architecture with techniques to improve model explainability to automatically build a domain-specific dictionary. As an illustrative use case of our approach, we create an objective dictionary that can infer conflict intensity from text data. We train the neural networks on a corpus of conflict reports and match them with conflict event data. This corpus consists of over 14,000 expert-written International Crisis Group (ICG) CrisisWatch reports between 2003 and 2021. Sensitivity analysis is used to extract the weighted words from the neural network to build the dictionary. In order to evaluate our approach, we compare our results to state-of-the-art deep learning language models, text-scaling methods, as well as standard, nonspecialized, and conflict event dictionary approaches. We are able to show that our approach outperforms other approaches while retaining interpretability.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | ; natural language processing; objective dictionaries; deep learning; transformers; conflict dynamics |
| Subjects: | 300 Social sciences > 330 Economics |
| Divisions: | Business, Economics and Information Systems > Institut für Betriebswirtschaftslehre > Lehrstuhl für Statistik und Risikomanagement (Prof. Dr. Rösch) |
| Depositing User: | Dr. Gernot Deinzer |
| Date Deposited: | 13 Mar 2024 13:04 |
| Last Modified: | 13 Mar 2024 13:04 |
| URI: | https://pred.uni-regensburg.de/id/eprint/59930 |
Actions (login required)
 |
View Item |
