Skip to Main Content
A rough set method is presented in this paper to assess the credit of contractors. Unlike traditional methods, the rough set method deduces credit-classifying rules from actual data to predict new cases. The method uses a contractors' database with a genetic algorithm and an exhaustive reduction implemented using ROSETTA software that integrates rough set method. The classification accuracy of the rough set model is not as good as that of a decision tree, logistic regression, and neural network models, but the rough set model more accurately predicts contractors with bad credit. The results show that the rough set model is especially useful for detecting corporations with bad credit in the currently disordered Chinese construction market.