Detection of Avocado Fruit Ripeness Level Using Classification Algorithm And Stratified K-Fold Cross Validation Method

Abstract

In the production of avocados, there are several problems, including classifying the fruit and whether the fruit is ripe, raw, or damaged. So the possible way that can be used is using an electronic nose (e-nose). The purpose of this study was to determine the level of ripeness of avocados, raw and ripe based on physical properties and chemical properties using classification algorithm and Stratified K-Fold Cross Validation methods. The method used in this research is experimental quantitative, testing the effect of one or more variables on other variables. The parameters tested in this study were gas content, weight loss, total dissolved solids, and water content. The results showed increased weight loss in avocados stored at room temperature from day 2 to day 5. The increase in fruit weight loss was mainly caused by the transpiration process or the release of water in the form of steam through the skin surface that occurred during thestorage period. An increase in weight loss in fruits is one sign that the fruit is starting to experience a decrease in its freshness quality. The water content of avocados from day 1 to day 5 fluctuated and tended to increase. Changes in water content are affected by room temperature and humidity during storage.