Fraud detection based-on data mining on Indonesian E-Procurement System (SPSE)

This paper focuses on detection of potential fraud that occurs in the procurement process via the Indonesian EProcurement System (SPSE). Potential frauds in procurement take very diverse forms such as corruption, collusion and tender fixation and more importantly, they are found in various stages ranging from the budgeting to the utilization stages. By analyzing the data, we show that there are several techniques that may work effectively to serve the goal of detecting these frauds. Furthermore, we also obtain that SPSE data contain a huge number of data-related issues such as inconsistencies in hierarchical structure among official agencies and company names among institutions as well as missing data points. In addition to that, the size of data to deal with is gigantic (about 515,069 projects), which renders the fraud detection mechanism has been a non-trivial problem. In this paper, we implement a fraud detection mechanism using data mining techniques based on supervised learning. The use of supervised learning depends on the availability of data labeling (fraud and non-fraud) which are extracted using string matching and manual extraction procedure from several data sources including court rulings, Komisi Pemberantasan Korupsi (KPK) publication and public comment. Test results show that Naive Bayes algorithm with 14 attributes obtained from dimension reduction produces the best performance with promising result compared to the other fraud detection techniques available to date. Besides, the sensitivity analysis deployed in the dimension reduction process has not only significantly reduced the dimension of the problem but also has improved the performance of the fraud detection technique.