Vesicular stomatitis forecasting based on Google Trends.

Area of Science:

• Veterinary Medicine
• Epidemiology
• Data Science

Background:

• Vesicular stomatitis (VS) is a significant viral disease affecting livestock, characterized by blisters on various body parts.
• Human infection with VS can lead to meningitis.
• Accurate prediction of VS outbreaks is crucial for disease management.

Purpose of the Study:

• To analyze 2014 American Vesicular Stomatitis outbreaks.
• To develop accurate prediction models for VS outbreak trends.
• To evaluate the effectiveness of Google Trends data in disease forecasting.

Main Methods:

• Collected American VS outbreak data from OIE.
• Utilized Google Trends data by inputting 24 disease-related keywords.
• Calculated Pearson and Spearman correlation coefficients to establish relationships.
• Constructed qualitative classification and quantitative regression models for prediction.

Main Results:

• Pearson correlation coefficients for the regression model were high (0.953 and 0.948).
• The best classification model achieved 78.52% sensitivity, 72.5% specificity, and 77.14% accuracy.
• A significant relationship was found between VS outbreaks and Google Trends data.

Conclusions:

• Google search data can be effectively applied to predict vesicular stomatitis outbreaks.
• Both qualitative and quantitative models provide accurate forecasts.
• The study demonstrates a novel approach for real-time disease surveillance.