AUTHOR=Cerchiello Paola , Nicola Giancarlo , Rönnqvist Samuel , Sarlin Peter TITLE=Assessing Banks' Distress Using News and Regular Financial Data JOURNAL=Frontiers in Artificial Intelligence VOLUME=Volume 5 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2022.871863 DOI=10.3389/frai.2022.871863 ISSN=2624-8212 ABSTRACT=n this paper we focus our attention on leveraging the information contained in financialnews to enhance the performance of a bank distress classifier. The news information shouldbe analyzed and inserted into the predictive model in the most efficient way and this taskdeals with the issues related to Natural Language interpretation and to the analysis of newsmedia. Among the different models proposed for such purpose, we investigate a deep learningapproach. The methodology is based on a distributed representation of textual data obtainedfrom a model (Doc2Vec) that maps the documents and the words contained within a textonto a reduced latent semantic space. Afterwards, a second supervised feed forward fullyconnected neural network is trained combining news data distributed representations withstandard financial figures in input. The goal of the model is to classify the correspondingbanks in distressed or tranquil state. The final aim is to comprehend both the improvementof the predictive performance of the classifier and to assess the importance of news data inthe classification process. This to understand if news data really bring useful informationnot contained in standard financial variables.