engleski

Fruit firmness prediction using multiple linear regression

Smart agriculture is a term used to describe the utilization of digital technologies used in optimizing agricultural food production systems. In order to increase the efficiency of manufacturing process, modern tools for collecting, storing and analyzing electronic data are used. The focus of this paper is creation and comparison of peach firmness prediction models using various machine learning algorithms. The size of the data set, which is used to construct machine learning models described in this paper, is small. Because size of the data set has a large impact on the performance of the machine learning algorithm, models of different complexities were developed to tackle this problem. Simple linear regression is used as fundamental techniques for predicting numerical outcome variable, the peach firmness using a single predictor variable. By extending simple linear regression model so that is can accommodate multiple predictors, multiple linear regression model is obtained, which is the top performing model when applied to the dataset described in this paper. The backpropagation neural network model is developed and its performance is compared to the performance of regression models.

smart agriculture ; BP neural networks ; machine learning ; linear regression ; prediction models ; fruit firmness

nije evidentirano