Implementation of Human Cognitive Bias on Naïve Bayes

Abstract

We propose a human-cognition inspired classification model based on Naïve Bayes. Our previous study showed that human-cognitively inspired heuristics is able to enhance the prediction accuracy of text classifier based on Naïve Bayes. In the study, our classification model showed higher performance than conventional Naïve Bayes under specific conditions. In this paper, to investigate the mechanism that realizes the higher performance of classification, we further tested our model and its modified variant. As a result, our two models showed slightly different behaviors, but both of them achieved higher performance than conventional Naïve Bayes.

Keywords

• naïve bayes
• text classification
• attribute independence assumption
• cognition-inspired model
• bayesian spam filtering