Evaluating the performance of machine learning methods and variable selection methods for predicting difficult-to-measure traits in Holstein dairy cattle using milk infrared spectral data

Fourier-transform infrared (FTIR) spectroscopy is a powerful high-throughput phenotyping tool for predicting traits that are expensive and difficult to measure in dairy cattle. Calibration equations are often developed using standard methods, such as partial least squares (PLS) regression. Methods that employ penalization, rank-reduction, and variable selection, as well as being able to model the nonlinear relations between phenotype and FTIR, might offer improvements in predictive ability and model robustness. This study aimed to compare the predictive ability of 2 machine learning methods, namely random forest (RF) and gradient boosting machine (GBM), and penalized regression against PLS regression for predicting 3 phenotypes differing in terms of biological meaning and relationships with milk composition (i.e., phenotypes measurable directly and not directly in milk, reflecting different biological processes which can be captured using milk spectra) in Holstein-Friesian cattle under 2 cross-validation scenarios. The data set comprised phenotypic information from 471 Holstein-Friesian cows, and 3 target phenotypes were evaluated: (1) body condition score (BCS), (2) blood β-hydroxybutyrate (BHB, mmol/L), and (3) κ-casein expressed as a percentage of nitrogen (κ-CN, % N). The data set was split considering 2 cross-validation scenarios: samples-out random in which the population was randomly split into 10-folds (8-folds for training and 1-fold for validation and testing); and herd/date-out in which the population was randomly assigned to training (70% herd), validation (10%), and testing (20% herd) based on the herd and date in which the samples were collected. The random grid search was performed using the training subset for the hyperparameter optimization and the validation set was used for the generalization of prediction error. The trained model was then used to assess the final prediction in the testing subset. The grid search for penalized regression evidenced that the elastic net (EN) was the best regularization with increase in predictive ability of 5%. The performance of PLS (standard model) was compared against 2 machine learning techniques and penalized regression using 2 cross-validation scenarios. Machine learning methods showed a greater predictive ability for BCS (0.63 for GBM and 0.61 for RF), BHB (0.80 for GBM and 0.79 for RF), and κ-CN (0.81 for GBM and 0.80 for RF) in samples-out cross-validation. Considering a herd/date-out cross-validation these values were 0.58 (GBM and RF) for BCS, 0.73 (GBM and RF) for BHB, and 0.77 (GBM and RF) for κ-CN. The GBM model tended to outperform other methods in predictive ability around 4%, 1%, and 7% for EN, RF, and PLS, respectively. The prediction accuracies of the GBM and RF models were similar, and differed statistically from the PLS model in samples-out random cross-validation. Although, machine learning techniques outperformed PLS in herd/date-out cross-validation, no significant differences were observed in terms of predictive ability due to the large standard deviation observed for predictions. Overall, GBM achieved the highest accuracy of FTIR-based prediction of the different phenotypic traits across the cross-validation scenarios. These results indicate that GBM is a promising method for obtaining more accurate FTIR-based predictions for different phenotypes in dairy cattle.     

VFC-SMOTE: very fast continuous synthetic minority oversampling for evolving data streams

Data Mining and Knowledge Discovery - The world is constantly changing, and so are the massive amount of data produced. However, only a few studies deal with online class imbalance learning that...     

Skew Gaussian processes for classification

Machine Learning - Gaussian processes (GPs) are distributions over functions, which provide a Bayesian nonparametric approach to regression and classification. In spite of their success, GPs have...     

Detection,localisation and tracking of pallets using machine learning techniques and 2D range data

The problem of autonomous transportation in industrial scenarios is receiving a renewed interest due to the way it can revolutionise internal logistics, especially in unstructured environments. This paper presents a novel architecture allowing a robot to detect, localise, and track (possibly multiple) pallets using machine learning techniques based on an on-board 2D laser rangefinder only. The architecture is composed of two main components: the first stage is a pallet detector employing a Faster Region-Based Convolutional Neural Network (Faster R-CNN) detector cascaded with a CNN-based classifier; the second stage is a Kalman filter for localising and tracking detected pallets, which we also use to defer commitment to a pallet detected in the first stage until sufficient confidence has been acquired via a sequential data acquisition process. For fine-tuning the CNNs, the architecture has been systematically evaluated using a real-world dataset containing 340 labelled 2D scans, which have been made freely available in an online repository. Detection performance has been assessed on the basis of the average accuracy over k-fold cross-validation, and it scored 99.58% in our tests. Concerning pallet localisation and tracking, experiments have been performed in a scenario where the robot is approaching the pallet to fork. Although data have been originally acquired by considering only one pallet as per specification of the use case we consider, artificial data have been generated as well to mimic the presence of multiple pallets in the robot workspace. Our experimental results confirm that the system is capable of identifying, localising and tracking pallets with a high success rate while being robust to false positives.

     

Corrosion and mechanical properties of plasma electrolytic oxidation‐coated AZ80 magnesium alloy

Plasma electrolytic oxidation (PEO) coatings were produced on AZ80 magnesium alloy in a solution containing silicates and phosphates and working at high current densities with short treatment times. The effect of a sealing treatment in boiling water on corrosion and mechanical properties of the coatings were investigated. Moreover, the corrosion mechanism of the samples with and without the sealing treatment was evaluated. The microstructure of the coatings was characterized with scanning electron microscope observation and X‐ray diffraction analysis. The mechanical properties were evaluated with nanoindentation tests and the corrosion resistance was studied by potentiodynamic polarization, electrochemical impedance spectroscopy, and scanning vibrating electrode technique. The results showed that the sealing did not influence the microstructure and the mechanical properties of the samples and instead produced a remarkable increase in the corrosion resistance. The crevice corrosion, present in the sample without the sealing, was avoided with the treatment in boiling water.     

Cachexia,a Systemic Disease beyond Muscle Atrophy

Cachexia is a complication of dismal prognosis, which often represents the last step of several chronic diseases. For this reason, the comprehension of the molecular drivers of such a condition is crucial for the development of management approaches. Importantly, cachexia is a syndrome affecting various organs, which often results in systemic complications. To date, the majority of the research on cachexia has been focused on skeletal muscle, muscle atrophy being a pivotal cause of weight loss and the major feature associated with the steep reduction in quality of life. Nevertheless, defining the impact of cachexia on other organs is essential to properly comprehend the complexity of such a condition and potentially develop novel therapeutic approaches.     

Using normative ethics for building a good evaluation of research practices: towards the assessment of researcher’s virtues

This paper is an attempt of using co-citation analysis to sort out and to analyze the development and evolution of a latest hot area, open innovation from the perspective of network embedding. A dataset of 1437 records published between 1990 and 2019 is collected from Web of Science database. The empirical results show the latest hot topics in the open innovation study focus on innovation performance and value creation. In addition, we make a new interpretation of open innovation from four aspects: innovation and entrepreneurship, resource acquisition, knowledge sharing and innovation performance, then combines the importance of network embedding to the innovation and development of enterprises, and proposes the future research direction of open innovation. Our research in this paper is helpful to systematically sort out the knowledge context of open innovation, which is of great significance to the construction and development of open innovation knowledge system. The conclusions and implications in this paper will be particularly illuminating for both academic research and enterprises’ practice application.

     

Bibliometric analysis and mapping of publications on brewing science from 1940 to 2018

Over the last 80 years, malting and brewing has evolved through increased scientific understanding, driven by numerous factors including efficiency improvement, cost reduction, quality enhancement, risk minimisation, brand marketing and image enhancement. The aim of this work was to review the historical trends in global brewing research in order to identify current opportunities and emerging trends, as well as uncover future key drivers. A bibliometric analysis was performed with English language peer reviewed literature from the Scopus database together with mapping using the VOSviewer clustering software. The bibliometric results were structured to identify which affiliations, authors, and journals have been involved in brewing research, as well as the average number of citations per paper. The bibliometric mapping yielded ‘research term maps’, where the words occurring more frequently in the research papers were grouped in coloured clusters by area of interest, and their relationships indicated. The year maps revealed which average terms tended to emerge in chronological order. The estimated average citation number per paper was about nine. Countries, including Spain, Czech Republic, Italy and Belgium displayed higher average citation scores than the world average. In the period 1940-79, brewing publications were mainly process oriented whereas between 2008-18, there was a shift in focus to the consumer (health and social responsibility) and to environmental sustainability. © 2020 The Institute of Brewing & Distilling     

The Unbearable Shallow Understanding of Deep Learning

This paper analyzes the rapid and unexpected rise of deep learning within Artificial Intelligence and its applications. It tackles the possible reasons for this remarkable success, providing candidate paths towards a satisfactory explanation of why it works so well, at least in some domains. A historical account is given for the ups and downs, which have characterized neural networks research and its evolution from “shallow” to “deep” learning architectures. A precise account of “success” is given, in order to sieve out aspects pertaining to marketing or sociology of research, and the remaining aspects seem to certify a genuine value of deep learning, calling for explanation. The alleged two main propelling factors for deep learning, namely computing hardware performance and neuroscience findings, are scrutinized, and evaluated as relevant but insufficient for a comprehensive explanation. We review various attempts that have been made to provide mathematical foundations able to justify the efficiency of deep learning, and we deem this is the most promising road to follow, even if the current achievements are too scattered and relevant for very limited classes of deep neural models. The authors’ take is that most of what can explain the very nature of why deep learning works at all and even very well across so many domains of application is still to be understood and further research, which addresses the theoretical foundation of artificial learning, is still very much needed.