Prediction of Olive Oil Productivity using Machine Learning Decision Tree Algorithm

Abstract

The production of olive oil is of great importance in the Palestinian economy, it considers an important consumable resource and one of the most important elements of food security. On another hand, it’s one of the main sources of income as it contributes to about 13% of the value of the annual agricultural production. It is noted that the production of olive oil in Palestine varies from year to year and varies from one governorate to another due to the climatic factors that contribute to fluctuating production based on rain and other climate elements. Applying machine learning and data mining techniques to analyze data in agriculture is a promising approach for the prediction of agricultural yield management. Most of the challenges that we faced in this study focused on the availability of recorded data for olive oil productivity for 25 years, which we were able to collect in cooperation with many sources. In this paper, we endeavored to study the impact of climatic factors on olive oil productivity using the decision tree algorithm. We aim to determine the most important climatic parameters that play a major role in determining the level of the past and the future production of olive oil. The decision tree algorithm achieved excellent results in specifying the factors that the most important and which influence the productions. Our results showed that this method will be awesome for the classification of climatic parameters and prediction of the future olive oil quantity. These study findings will serve the interest of researchers, decision-makers in predicting future yields of the olive oil.