Keele Research Repository

Recently random forest classification became a popular choice machine learning applications aimed to detect spam content in online social networks. In this paper, we report a systematic analysis of random forest classification for this purpose. We assessed the impact of key parameters, such as number of trees, depth of trees and minimum size of leaf nodes on classification performance. Our results show that controlling the complexity of random forest classifiers applied to social media spam is important in order to avoid overfitting and optimize performance We also conclude that in order to support reproducibility of experimental results it is important to report key parameters of random forest classifiers.