The Optimized K-Means Clustering Algorithms to Analyzed the Budget Revenue Expenditure in Padang

Dony Novaliendy (National Kaohsiung University of Taiwan)
Yeka Hendriyani (Universitas Negeri Padang)
Cheng-Hong Yang (National Kaohsiung University of Taiwan)
Hafilah Hamimi (Universitas Negeri Padang)

Proceeding of the Electrical Engineering Computer Science and Informatics
Vol 2: EECSI 2015

APBD is a systematic detailed list of receipts,expenditures and local spending within a year arranged inPERMENDAGRI No. 16 of 2006, so that the data of APBD canbe used as guidelines for governments and local expenditures incarrying out activities to raise revenue to maintain economicstability and to avoid inflation and deflation. Governmentfinancial institutions in areas such as DPKA Padang, experienceddifficulties in identifying the relevance of each archive data onAPBD, that result in a data warehouse. In addition, to theadministration, APBD in the government of Padang have notbeen effective. To minimize the difficulty in identifying dataarchive of APBD, then the data warehouse can be used toproduce knowledge using the techniques of Data Mining (DM).The method that is used are clustering and forecasting.Clustering performed using the K-Means Algorithm whileforecasting is done by using multiple linear regressions. Thesemethods intended to classify and identify the data in the budgetthat have certain characteristics in common, and can predict thevalue of APBD for the following years.