Mango is an important horticulture crop with high economic value. The manual classification and postharvest handling cause high postharvest losses. The classification of mango using emerging machine learning techniques offers an advanced solution to reduce postharvest losses. The fruits were classified into three different categories; raw, intermediate, and ripe, using L*, a*, and b* (LAB) color characteristics. The non-invasive classification was performed using Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbour (K-NN). The K-fold cross-validation (CV) was used to assess the model's performance. Further, the precision, recall, Root Mean Square Error (RMSE), F1 measure, and confusion matrices were used to assess model performance. The SVM resampling using 10-fold cross-validation and the tuning parameter “sigma = 1.577 and C = 1.0” showed the highest accuracy of 98.67% among all four algorithms used across the cross-validation set. Thus, suggesting its efficient potential in classifying mangoes with varying maturity levels.