Open AccessInformation System of Agricultural Commodities Mapping Based on Machine Learning
Author(s) -
Devi Fauziah Nur,
C. Riyanti,
Meylanie Olivya
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1577/1/012008
Subject(s) - artificial neural network , agriculture , relative humidity , artificial intelligence , computer science , backpropagation , climate change , precipitation , machine learning , training set , agricultural engineering , meteorology , geography , engineering , ecology , archaeology , biology
Climate change that occurs from year to year affects the agricultural sector. People who work in agriculture need to know the compatibility of plants with climate conditions in the region. Back Propagation Neural Network (BPNN) is a multilayer Artificial Neural Network (ANN) that is used to train neural networks with input in the form of precipitation, relative humidity, and temperature data. The training produced a model that was able to classify climate data based on plant growth requirements. The model for Soybean got the highest average accuracy of 96.53% and 90.87% for Rice. Variables that influence training include the number of neurons in the hidden layer, the value of learning rate, the number of folds, and the number of epochs. Prediction results generated from the model are used as a reference to display markers on maps that can be accessed by users.