Search

A back-propagation artificial neural network was employed to detect clinical mastitis using a file of 460,474 test day records. Two data files were created to train the artificial neural networks, containing a relatively large (1:1) ratio and a relatively small (1:10) ratio in the incidence to non-incidence of clinical mastitis. These ratios were applied to each of two input file designs; one comprised variables that are traditional in the modeling of mastitis (e.g., age, stage of lactation and somatic cell count) and a second included additional variables (e.g., season of calving, milk components and conformation class). Results from analyses of relative operating characteristics indicated that artificial neural networks could discriminate between mastitic states with an overall accuracy of 86%. This discriminatory ability was subject to patterns that existed in the training data files but was not affected by differing proportions of mastitic records. However, differing proportions of mastitic records had some effect on the particular purpose of the artificial neural network being developed: training with a higher proportion of mastitic cases increased the ability of an artificial neural network in discriminating positive from negative cases. Similar effects were obtained by modifying the threshold value used to categorize the ANN output, which constitutes a much simpler approach than modifying the proportion of cases in training data sets. Additional variables had little effect on the prediction accuracy, but this lack of effect needs to be verified for optimal artificial neural network configuration, data preprocessing, and new sources of information. more...

The effect of data preprocessing on the learning ability of artificial neural networks was investigated with regard to the impact of distributing the input vectors uniformly with respect to the output categories in the training data set. The analyses were performed for neural networks dedicated to (1) dairy cow culling classification and (2) milk yield prediction. The two types of neural network used for culling classification were backpropagation and learning vector quantization. For yield prediction, backpropagation was used. The study was repeated with several architectures for both types of network. Preprocessing of data did not have a large impact on the general performance of the networks, but did affect the results for each output category. The effects were more pronounced in the categories containing less frequent events, for which the results always improved. For the categories with larger number of records, balancing the data degraded the results. The respective improvements and degradation of the results occurred for both prediction and classification, with the two types of neural networks, and with all architectures tested. However, the magnitude of the effects varied with the type of neural network and with the architecture. The results of this study indicate that, in general, the distribution of outputs influences the learning process of neural networks for both types of application. The results also suggest that the types of output distribution required for the training of neural nets may depend on the specifics of each problem. more...

Since the main reason for disposal of dairy cows is low milk yield, implementation of an optimum selection program requires the prediction of cow performance with regard to production. The prediction of fat and protein content in milk are also rapidly becoming important factors for decisions related to breeding and herd policy. While, on average, traditional lactation models furnish good results, some improvement is possible when predicting the yield for an individual cow early in lactation. Artificial neural networks (ANNs), known to perform well in pattern recognition, may constitute an effective alternative to the traditional models. The objective of this research was to investigate how ANNs might be used to predict total milk, fat, and protein production for individual cows. Results indicated that ANNs generally performed at least as well overall as the model currently used by Canadian milk recording agencies, especially in the first third of lactation. This has important implications for early identification of superior animals. Predictions from both methods were relatively similar for the later stages of lactation. The addition of nontraditional data inputs such as average milk herd production and weight of cow improved the quality of prediction. Three different techniques were used to analyze the sensitivity of the ANN to different inputs, and their relative abilities are discussed. Results illustrate the potential effectiveness of ANNs in predicting milk yield and its composition and appear to justify further pursuit of this research. more...

Machine learning was employed to develop knowledge–based modules to automate the filtering of test–day data for group–average lactation curve analysis. The classification task consisted of identifying an individual cow’s milk–yield test as being either outlier or non–outlier. Experiments were carried out to explore the importance of deriving predictive attributes and choice of machine–learning technique. Data consisted of 1080 milk yield tests, of which 108 had been classified as outliers by a dairy nutrition specialist and 972 cases had been classified as non–outliers. Two different approaches to machine learning were applied to these data: decision–tree induction and naive–Bayes classification. Performance of the classifiers was estimated through ten–fold cross–validation, while relative operating characteristic curves were used to visualize the achieved trade–off between sensitivity and specificity. Use of an initial set of derived attributes significantly improved the performance compared to limiting attributes to those available to the domain specialist. However, adding even more complex attributes did not necessarily improve the classification performance. Overall, the naive–Bayes approach showed significantly better performance than decision–tree induction. For each machine–learning approach, three final classifiers, associated with a low, medium, and high filtering intensity, were generated from the 1080 cases. The expected true positive rate varied from 41% to 78% for false positive rates between 1.1% and 4.6%. However, due to the low prevalence of outlier tests, this performance was associated with a large number of false positives. The domain specialist considered the final classifiers of both approaches as plausible, but found the decision trees easier to understand than the naive–Bayes classifiers. Machine learning was considered to be a promising approach to assist knowledge acquisition. more...