A policy knowledge- and reasoning-based method for data-analytic city policymaking

Abstract

Efforts are being directed towards the implementation of data analysis in various areas of policymaking. In many studies, data analysis has been conducted by applying scientific methods on objective data. However, very few studies have dealt with this aspect pragmatically, starting from the data collection stage. This paper presents knowledge and reasoning systems for establishing city policies based on data analysis. First, city policy-related data are collected, and a clustering method is used for analysis. Next, Shapley value theory is used to determine the levels of inter-variable influence, and machine learning techniques, such as the decision tree, Bayesian analysis, and regression analysis, are implemented using the major variables to determine policies. Finally, a system dynamics model is designed to review the policy reasoning and assess its practicality.