PENERAPAN TEKNIK LOGISTIC REGRESSION DALAM PREDIKSI KUALITAS APEL PADA BERBAGAI VARIABEL FISIK

Abstract

In the agricultural industry, improving the quality of apples has a direct impact on the selling price and customer attractiveness. Logistic regression method is used to predict the quality of apples. Physical variables such as size, weight, sweetness, and acidity level were used in the study. This study uses a quantitative approach with descriptive design, and Apple samples obtained through purposive sampling method. The independent variable includes the relevant physical factors, while the dependent variable is called good or bad. The results showed that size, weight, sweetness, and juiciness contributed negatively to Apple quality, while ripeness and acidity contributed positively. With an accuracy rate of 73.03%, the logistic regression model used to detect poor quality apples better than good quality apples. However, this model also has considerable misclassification in good quality apples, which can lead to the product being discarded. It is hoped that this research will contribute to the field of Agriculture and statistics, as well as offer practical solutions for farmers and the apple industry to improve the efficiency of Apple quality prediction using machine learning technology. It is estimated that the application of this technology in agriculture can reduce losses and increase overall productivity.