Open Data Analytical Model for Human Development Index Optimization to Support Government Policy

A. Alamsyah,T.T. Gustyana,A.D. Fajaryanto,D. Septiafani

DOI: https://doi.org/10.48550/arXiv.1809.00189

2018-09-01

Abstract:The transparency nature of Open Data is beneficial for citizens to evaluate government work performance. In Indonesia, each government bodies or ministry have their own standard operating procedure on data treatment resulting in incoherent information between agent and likely to miss valuable insight. Therefore, our motivation is to show the advantage of Open Data movement to support unified government decision making. We use the dataset from <a class="link-external link-http" href="http://data.go.id" rel="external noopener nofollow">this http URL</a> which publish official data from each government bodies. The idea is by using those official but limited data, we can find important pattern. The case study is on Human Development Index value prediction and its clustered nature. We explore the data pattern using two important data analytics methods classification and clustering procedure. Data analytics is the collection of activities to reveal unknown data pattern. Specifically, we use Artificial Neural Network classification and K-means clustering. The classification objective is to categorize different level of Human Development Index of cities or region in Indonesia based on Gross Domestic Product, Number of Population in Poverty, Number of Internet User, Number of Labors and Number of Population indicators data. We determined which city belongs to four categories of Human Development stated by UNDP standard. The clustering objective is to find the group characteristics between Human Development Index and Gross Domestic Product.

Computers and Society