Prediction of moisture in cheese of commercial production using neural networks

Detecting heterogeneity parameters and hybrid models for precision farming

Precision farming (PF) plays a crucial role in the field of agriculture to solve the challenges of food shortages in society. Heterogeneity, multicollinearity, and outliers are problems in PF because they can cause bias and lead to incorrect inferences. However, traditional methods typically assume it to be a homogenous model, and in machine learning, data scientists ignore heterogeneity. In this study, the aim is to identify the heterogeneity parameters and develop hybrid models before and after heterogeneity. Data on seaweed is collected using sensor smart farming technology attached to v-Groove Hybrid Solar Drier (v-GHSD). There are 29 drying parameters, and each parameter has 1914 observations. We considered the highest order up to the second order interaction, and the parameters increased to 435 parameters from 29 parameters. In high-dimensional data, the number of observations is less than the number of parameters. The authors proposed a method using the variance inflation factor to identify the heterogeneity parameters. Seven predictive models such as ridge, random forest, support vector machine, bagging, boosting, LASSO and elastic net are used to select the 15, 25, 35 and 45 significant drying parameters for the moisture content removal of the seaweed, and hybrid models are developed using robust statistical methods. For before heterogeneity, the hybrid model random forest M Hampel with 19 outliers is the best, because it performs better when compared to other models. For after heterogeneity, the hybrid model boosting M Hampel with 19 outliers is the best, because it performs better when compared to other models. These results are vital to seaweed precision farming. The study of heterogeneity will not only help us to comprehend the dynamics of the large number of the drying parameters, but also gives a way to leverage the data for efficient predictive modelling.

pH prediction for a semi-batch cream cheese fermentation using a grey-box model

Cream cheese, a popular condiment, is widely used in people’s daily diet and in dessert making. To ensure high-quality cream cheese production, the pH value is generally used as the indicator to determine the end point of cream cheese fermentation. The inoculation time and time-dependent concentrations of biomass, lactose, lactic acid are all crucial for pH prediction. However, the inoculation time could vary for industrial applications with multiple fermenters. Moreover, the inoculation time impact on fermentation has not been investigated. This paper aims to build a cream cheese fermentation model predicting pH. The model includes a semi-batch kinetic model and an artificial neural network (ANN) model. The outcome of the model will help the cream cheese industries understand the inoculation time impact on fermentation time and organise better fermenter scheduling.

Artificial Neural Network Simulated Elman Models for Predicting Shelf Life of Processed Cheese

Elman artificial neural network models with single and multilayer for predicting shelf life of processed cheese stored at 7-8ºC were developed. Input parameters were: Body & texture, aroma & flavour, moisture, and free fatty acid, while sensory score was output parameter. Bayesian regularization was training algorithm for the models. The network was trained up to 100 epochs, and neurons in each hidden layers varied from 1 to 20. Transfer function for hidden layer was tangent sigmoid, while for the output layer it was pure linear function. Mean Square Error, Root Mean Square Error, Coefficient of Determination and Nash - Sutcliffo Coefficient were used for comparing the prediction ability of the developed models. Elman model with combination of 4-17-17-1 performed significantly well for predicting the shelf life of processed cheese stored at 7-8º C.

A Model Describing Debaryomyces hansenii Growth and Substrate Consumption During a Smear Soft Cheese Deacidification and Ripening

A mechanistic model for Debaryomyces hansenii growth and substrate consumption, lactose conversion into lactate by lactic acid bacteria, as well as lactose and lactate transfer from the core toward the rind was established. The model described the first step (14 d) of the ripening of a smear soft cheese and included the effects of temperature and relative humidity of the ripening chamber on the kinetic parameters. Experimental data were collected from experiments carried out in an aseptic pilot scale ripening chamber under 9 different combinations of temperature (8, 12, and 16 degrees C) and relative humidity (85, 93, and 99%) according to a complete experimental design. The model considered the cheese as a system with 2 compartments (rind and core) and included 5 state evolution equations and 16 parameters. The model succeeded in predicting D. hansenii growth and lactose and lactate concentrations during the first step of ripening (curd deacidification) in core and rind. The nonlinear data-fitting method allowed the determination of tight confidence intervals for the model parameters. The residual standard error (RSE) between model predictions and experimental data was close to the experimental standard deviation between repeated experiments.