Human Development Index Prediction in East Kalimantan uses the FFNN (Feed Forward Neural Network) optimized MLR (Multiple Linear Regression) Method

A decent standard of living is an adequate standard for each person for himself and his family, which is related to the basic needs, and the improvement of sustainable living conditions. The poverty line is the minimum income level to obtain an adequate standard of living. On the other hand, HDI (Human Development Index) is an aggregate indicator that provides a complete picture of the quality of life. Two critical points shown by HDI are the ability of humans to improve their health and education, and the ability of humans to enjoy life on the results of their efforts that are reflected by their income. Of course, the Prosperous Family has fulfilled a decent standard of living, and vice versa for Poor Families. NPF (the Number of Prosperous Families) is the number of families that meet PFI (Prosperous Family Index) requirements. The number of Poor Families contributes to %PP (the Percentage of Poor People). Therefore it could be assumed that NPF has a positive influence on HDI, while the %PP has a negative influence. The results of this study showed that the assumption has been proven, where NPF has a positive influence on HDI (𝛽1=0.0525), and vice versa %PP has a negative influence (𝛽2=−0.1253). Both variables (NPF and %PP) simultaneously have an influence on HDI of 47.75% (𝑅𝑠𝑞𝑢𝑎𝑟𝑒=0.4775). For predictive needs, the MLR (Multiple Linear Regression) model was optimized by using FFNN (Feed Forward Neural network). The results of studies have shown that the MLR model with optimization provides much better performance (%𝑅𝑠𝑞𝑢𝑎𝑟𝑒=98.66%) compared to the MLR model without optimization (%𝑅𝑠𝑞𝑢𝑎𝑟𝑒=54.65%) in terms of predictive activity.