In increasing sales by increasing consumer satisfaction with the quality of coffee sold. A way is needed to make it easier to predict the determination of quality coffee so as to increase the efficiency of the coffee sorting process which does not take a long time and can increase the productivity of companies that have competitiveness. Several developments have been made to improve the performance of the algorithm which has the potential to produce good quality predictions. Import Copy Data into a format that can be processed to a later stage or with a Machine Learning algorithm. Copy data that can be processed is then modified in such a way as to ensure that the data is suitable for use in Data Science or Machine Learning processes. By using coffee data specifications from the plantation to the coffee beans produced, it is expected that coffee quality can be predicted quickly without the need for manual calculations or analysis by humans. The working procedures for selecting the quality of coffee beans are coffee import data, coffee data processing, split test-train coffee data, light gradient enhancement machine, yield prediction, and Performance Prediction Evaluation. The amount of data used is 1,339 data. The dependent variable in this data is Coffee Quality while the rest will be cleaned and processed to serve as an independent variable. The accuracy rate of the algorithm in predicting coffee quality is 72%.