"As the coronavirus crisis evolves, we recognise that comprehensive, real-time information is essential for decision-makers," said Keith Berry, managing director of Moody's Analytics Accelerator. "Coronavirus Pulse complements quantitative analysis by more effectively helping market participants identify emerging risk categories in the industries they monitor. »Coronavirus Pulse uses machine learning and natural language processing to find English-language news articles related to COVID-19 and to identify companies, sector-specific categories and events mentioned in each article. It then assigns "article" and "credit" sentiment to each news item by assessing the overall context of the article. An unfavourable, neutral or positive "article" sentiment is assigned based on an approach that captures reputational, political, medical and macroeconomic risks. At the same time, an unfavourable credit sentiment is assigned to each item based on an approach developed for use in Moody's Analytics Credit Sentiment solution Score™, which identifies credit risks. Developed by Moody's Analytics' Accelerator, Coronavirus Pulse is built on transferable artificial intelligence models, is customized and uses relevant data from Moody's Analytics. It is part of a suite of scalable solutions that use advanced machine learning techniques to scan unstructured data and provide sentiment analysis of current events. Translated from COVID-19 : Moody's Analytics présente un outil d’analyse de sentiment vis-à-vis de l’actualité
Our articles will then appear first in Google Top Stories.