The Prediction of Electrical Grid Stability Using Naïve Bayes and K-Means Algorithm

Baik Budi
Ilhamdi Rusydi, Muhammad
Arya Witama, Reivan
Hesti Ramadhamy, Queen
Budiman, Refki

Andalasian International Journal of Applied Science, Engineering, and Technology
Vol. 5 No. 2 (2025): July 2025

This study explores the use of Naive Bayes and k-means algorithms to predict and analyzed stability of the electrical grid. Data set for this research is public dataset from Kaggle. The main goal of the research is to develop an accurate and efficient predictive model. Naive Bayes was chosen it has ability to handle independent features and also have a compatibility with highdimensional data. The implementation was carried out using Python in Google Colab, with data preprocessing that included feature normalization and an 80:20 train-test split. The Gaussian Naive Bayes model was used for system stability classification. The results demonstrate excellent model performance, with an accuracy of 97.35%, precision of 98.91%, recall of 97.02%, and an F1-score of 97.95%. The confusion matrix reveals the model's ability to classify "stable" and "unstable" conditions with minimal prediction errors.