Predicting Methane Emissions Using Artificial Neural Networks and Decision Tree

A study of models like Artificial Neural Network (ANN) and Decision Tree for the prediction of methane emissions from agricultural activities has been examined in the paper. The strengths and limitations of these models are brought into light in the paper. The performance of these models discussed in the paper will aid researchers and stakeholders and scientists to predict various environmental activities. In the present study, a sample set of 100 was taken. Features like population of livestock and usage of fertilizers were taken into consideration. For the aforesaid study, metrices like Mean Squared, Accuracy, Precision, Recall, R-squared F1-score, Error (MSE)and the Area Under the Curve (AUC) were employed. A comparative analysis of ANN model and Decision Tree was done. There was a higher rate of accuracy and precision of the Decision Tree model as that of the ANN model. The ANN model showcased moderate predictive results because according to the study, it could not handle non-linear relationships and interactions within the data like the Decision Tree. The comparative study of both the models in this study would provide a wise recommendation for researchers and environmental stakeholders.

Перевод пока недоступен