Back

Using Random Forest feature importance results to predict zoonosis

Gonzalez, R. G.

2023-05-05 health informatics
10.1101/2023.05.04.23289530 medRxiv
Show abstract

This study fills a gap in the literature regarding using machine learning techniques within the field of zoonoses. Instead of using linear and logistic inference modeling like in previous (Knowledge, Attitudes, and Practices (KAP) studies, this study incorporates Random Forest (RF) to identify important features that predict zoonotic diseases using survey and blood serology data. Using RF, we found that the most important features related to zoonoses were villages where households were 46 or larger and where participants owned many animals such as ducks, cats, and pigs. Compared to previous KAP studies in other countries where ethnicity, age, and education background were important features regarding knowledge, awareness, and practices relating to zoonoses, the KAP Cambodia case was different because these features were not found to be important.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.