To anticipate these crises, you can use i predictive models but they are based on risk measures that are often delayed, obsolete or incomplete. The New York University study tried to understand how to exploit predictive algorithms in an optimal way.
The study showed that by compiling the text of 11,2 million articles on food insecure countries published between 1980 and 2020, and taking advantage of recent advances in deep learning: comforting results can be obtained. The elaboration allowed to extract high-frequency precursors of food crises that are both interpretable and validated by traditional risk indicators.
The algorithm of deep learning highlighted that over the period from July 2009 to July 2020, crisis indicators substantially improve forecasts in 21 food insecure countries, up to 12 months earlier than baseline models that do not include textual information.
The study focuses on the Integrated Phase Classification (IPC) prediction of food insecurity published by the Famine Early Warning Systems Network (FEWS NET). This classification is available at the district level in 37 food insecure countries in Africa, Asia and Latin America and was reported four times a year between 2009 and 2015 and three times a year thereafter.
Food insecurity is classified according to an ordinal scale consisting of five stages: low, stress, crisis, emergency and famine.
BlogInnovazione.it
Last Monday, the Financial Times announced a deal with OpenAI. FT licenses its world-class journalism…
Millions of people pay for streaming services, paying monthly subscription fees. It is common opinion that you…
Coveware by Veeam will continue to provide cyber extortion incident response services. Coveware will offer forensics and remediation capabilities…
Predictive maintenance is revolutionizing the oil & gas sector, with an innovative and proactive approach to plant management.…