PENGGUNAAN METODE REGRESI LINIER UNTUK ESTIMASI ANGKA PERCERAIAN : STUDI KASUS PENGADILAN AGAMA KABUPATEN GROBOGAN

Abstract

Data mining is the process of analyzing large datasets to discover hidden patterns, trends, 
and valuable information. This study utilizes data mining to address a social issue, 
specifically estimating divorce rates in the Religious Court of Grobogan Regency. The 
method used is multiple linear regression, with the dependent variable being the number of 
divorces and independent variables including 'cerai talak' (divorce initiated by the husband), 
'cerai gugat' (divorce initiated by the wife), and 'dispensasi kawin' (marriage dispensation). 
The objective of this research is to test and develop a data mining method to estimate divorce 
rates, thereby aiding the Religious Court of Grobogan Regency in formulating more effective 
policies, based on divorce data from 2023-2024. The research process includes data 
collection, pre-processing, algorithm implementation, and result evaluation. The analysis 
shows that the multiple linear regression model provides reasonably accurate estimates, with 
a Root Mean Square Error (RMSE) of 6.505 and a Relative Root Squared Error (RRSE) of 
0.070. Further analysis reveals that 'cerai talak,' 'cerai gugat,' and 'dispensasi kawin' 
significantly affect divorce rates, with 'cerai gugat' being the most dominant factor. These 
findings provide a solid foundation for developing strategic policies to handle divorce cases 
in Grobogan Regency. To improve model accuracy, data enrichment and additional 
variables are needed. Collaboration between academics and the Religious Court of 
Grobogan Regency is also crucial to ensure the successful implementation of this research’s 
findings.