Application of MOS Gas Sensors Coupled with Chemometrics Methods to Predict the Amount of Sugar and Carbohydrates in Potatoes.

Five potato varieties were studied using an electronic nose with nine MOS sensors. Parameters measured included carbohydrate content, sugar level, and the toughness of the potatoes. Routine tests were carried out while the signals for each potato were measured, simultaneously, using an electronic nose. The signals obtained indicated the concentration of various chemical components. In addition to support vector machines (SVMs that were used for the classification of the samples, chemometric methods, such as the partial least squares regression (PLSR) method, the principal component regression (PCR) method, and the multiple linear regression (MLR) method, were used to create separate regression models for sugar and carbohydrates. The predictive power of the regression models was characterized by a coefficient of determination (R2), a root-mean-square error of prediction (RMSEP), and offsets. PLSR was able to accurately model the relationship between the smells of different types of potatoes, sugar, and carbohydrates. The highest and lowest accuracy of models for predicting sugar and carbohydrates was related to Marfona potatoes and Sprite cultivar potatoes. In general, in all cultivars, the accuracy in predicting the amount of carbohydrates was somewhat better than the accuracy in predicting the amount of sugar. Moreover, the linear function had 100% accuracy for training and validation in the C-SVM method for classification of five potato groups. The electronic nose could be used as a fast and non-destructive method for detecting different potato varieties. Researchers in the food industry will find this method extremely useful in selecting the desired product and samples.

Implementation of Inertia Sensor and Machine Learning Technologies for Analyzing the Behavior of Individual Laying Hens.

Welfare-oriented regulations cause farmers worldwide to shift towards more welfare-friendly, e.g., loose housing systems such as aviaries with litter. In contrast to the traditional cage housing systems, good technical results can only be obtained if the behavior of hens is considered. With increasing flock sizes, the automation of behavioural assessment can be beneficial. This research aims to show a proof of principle of tools for analyzing laying-hen behaviors by using wearable inertia sensor technology and a machine learning model (ML). For this aim, the behaviors of hens were classified into three classes: static, semi-dynamic, and highly dynamic behavior. The activities of hens were continuously recorded on video and synchronized with the sensor signals. Two hens were equipped with sensors, one marked green and one blue, for five days to collect the data. The training data set indicated that the ML model can accurately classify the highly dynamic behaviors with a one-second time window; a four-second time window is accurate for static and semi-dynamic behaviors. The Bagged Trees model, with an overall accuracy of 89% was the best ML model with the F1-scores of 89%, 91%, and 87% for static, semi-dynamic, and highly dynamic behaviors. The Bagged Trees model also performed well in classifying the behaviors of the hen in the validation data set with an overall F1-score of 0.92 (uniform either % or decimals). This research illustrates that the combination of wearable inertia sensors and machine learning is a viable technique for analyzing the laying-hen behaviors and supporting farmers in the management of hens in loose housing systems.