Study on Illegal Wildlife Trade Based on Hierarchical Analysis and Random Forest Modeling

Abstract

With the goal of solving the problem of illegal wildlife trade, this study proposes a data-driven five-year project by combining hierarchical analytical modeling and random forest modeling. First, 16 potential factors were analyzed through a hierarchical analytical model to identify key customers. Then, a linear regression model was used to develop a five-year program suitable for the client. Subsequently, a random forest model was used to verify that the program met the client's needs, and model training was used to predict the impact and probability of success of the program. Finally, in conjunction with the AHP model, the program was confirmed to significantly reduce illegal trade. This study proposes an integrated approach that provides an effective solution for wildlife conservation, and hopefully will lead to a win-win situation for the protection of ecological balance and human interests.