Classification and regression tree (CART) has been widely used in data mining to solve classification and prediction problems. In this paper, we propose a novel numerical ant-colony decision tree (nACDT) algorithm that combines CART with ant-colony optimisation (ACO). The combination works not only in inducing decision trees but also in incorporating the discretisation of attributes during the process to cope with continuous attributes. The proposed algorithm is used to estimate the duration of ship maintenance, with the aim of improving service quality and competitive advantage in the shipyard industry. The results show that the predictive accuracy of the proposed algorithm is statistically significantly higher than the accuracy of CART, which is the most well-known conventional decision tree algorithm.
|Number of pages||12|
|Journal||International Journal of Applied Management Science|
|Publication status||Published - 1 Jan 2015|
- Ant colony
- Data mining
- Decision tree
- Dry docking