Embedded Files
Epidemiological semaphore prediction
Epidemiological semaphore prediction
This task consists of determining the Mexican Epidemiological Semaphore through the covid news in Mexico. The epidemiological semaphore, among other things, allows the activities for the tourism sector. It is a system of four ordered colors: red, orange, yellow, and green, where red is the most restrictive, and green is the most permissive. The color of the semaphore changes weekly and is independent in each state of the Mexican Republic. This color is determined by the government taking into account various factors such as the capacity of hospitals, the contagion and death curve, available ventilators, among others. Coincidentally, these factors are shared through the news of each state. If it is possible to predict the color of the epidemiological semaphore in advance, this will allow sellers and tourism service providers to take preventive measures.
This task aims to generate classification models that help determine the color of the epidemiological semaphore in Mexico and thus be better prepared for the different changes in the evolution of the pandemic. The corpus consists of 77,983 news related to covid-19 collected from June 1, 2020, in the 32 regions of Mexico.
This is a classification task of four classes: red, orange, yellow, and green. Each instance in the corpus consists of the news and the Mexican state where it was published. Each participant team must classify the semaphore color for the week in which the news was published (week 0), the semaphore color 2 weeks later (week 2), the semaphore color one month later (week 4), and finally, the semaphore color of 2 months later (week 8).
Example
Example
"Coronavirus en México: ¿qué hay detrás del súbito aumento de muertes por la pandemia de covid-19? Se registra la cifra más alta de muertes por covid-19. La noticia alarmó a muchos en México: de un día a otro el registro de personas muertas en la pandemia de coronavirus se duplicó."
State: Estado de México
Labels:
Week 0: Red
Week 2: Red
Week 4: Orange
Week 8: Yellow
Page updated
Report abuse