A Machine Learning Framework for Olive Farms Profit Prediction

Abstract

Agricultural systems are constantly stressed due to higher demands for products. Consequently, water resources consumed on irrigation are increased. In combination with the climatic change, those are major obstacles to maintaining sustainable development, especially in a semi-arid land. This paper presents an end-to-end Machine Learning framework for predicting the potential profit from olive farms. The objective is to estimate the optimal economic gain while preserving water resources on irrigation by considering various related factors such as climatic conditions, crop management practices, soil characteristics, and crop yield. The case study focuses on olive tree farms located on the Hellenic Island of Crete. Real data from the farms and the weather in the area will be used. The target is to build a framework that will preprocess input data, compare the results among a group of Machine Learning algorithms and propose the best-predicted value of economic profit. Various aspects during this process will be thoroughly examined such as the bias-variance tradeoff and the problem of overfitting, data transforms, feature engineering and selection, ensemble methods as well as pursuing optimal resampling towards better model accuracy. Results indicated that through data preprocessing and resampling, Machine Learning algorithms performance is enhanced. Ultimately, prediction accuracy and reliability are greatly improved compared to algorithms’ performances without the framework’s operation.

Automatic Optimization of Deep Learning Training through Feature-Aware-Based Dataset Splitting

“…Resampling, including normal, repeated, nested, leave-one-out KFCV, was used by [38] to split a dataset for training different ML algorithms (linear regression, Bayes ridge regression, ridge regression, LASSO regression, K-nearest neighbors, CART decision trees, support vector machines regression-SVMR, extreme gradient boosting, gradient boosting, random forests, and extra trees) to predict profits in olive farms. Their results presented a better performance for SVMR compared to the others, and the resampling technique outperformed the other splitting techniques.…”

An effective crop recommendation method using machine learning techniques

“…While trying to make accurate predictions, resampling throughout the training phase was crucial since it made it possible to determine which algorithms generalized best depending on our data. Also, it considerably uses the process of hyperparameter tuning, which involves modifying specific parameters of algorithms to improve outcomes [35]. The commonly used variations on cross-validation are train/test split, leave one out cross-validation (LOOCV), k-fold cross-validation, etc.…”

“…Although it is similar to k-fold cross-validation, the main distinction is that n different data splits are carried out. Simple crossvalidation uses well-known k values (5 and 10) to reduce complexity [35]. Train-test split typically divides the dataset into training and test data in an 80:20 ratio and mimics how a model would perform on new and unseen data.…”

Automated prediction of income from farming of a commodity: An ARIMA based framework

“…The former makes the input data trend stationary, while the latter checks the effect of any seasonal parameter in the model. Christias and Mocanu (2021) applies various machine learning algorithms for Olive farm profit prediction. However, it did not experiment with time-series based prediction.…”