A Delphi-based rough sets fusion model for extracting payment rules of vehicle license tax in the government sector

It is a problematic issue faced by the government sector to effectively discover potentially owed taxes (overdue payments) and continually promote the principle of taxation justices. However, due to the high economic development over the past 30 years in Taiwan, the quantity of vehicles recorded contingent with the mounds of data generated and collected in the Tax Bureau is growing at a fast rate concurrently; therefore, the mission-critical nature of the data and the speed with which analyses need to be made now increase the requirements for a more reliable way to dig out a government’s taxation information hidden. Based on the reasons above, this study proposes a hybrid model, which combines the Delphi method and rough sets classifier approaches, for intelligently classifying the vehicle license tax payment (called VLTP) to solve real-world problems that are faced by taxation agencies. The proposed hybrid model is illustrated by examining a practically collected dataset, and the experimental results reveal that this hybrid model outperforms the listing methods in terms of accuracy and its standard deviation. More importantly, the output created by rough sets LEM2 (Learning from Examples Module, version 2) algorithm is a set of comprehensible and meaningful rules applied readily in knowledge-based systems of payment classification of vehicle license tax for tax authority.