feature selection algorithm

Data mining explores hidden and invisible relationships and patterns in the vast amount of data without which these relationships may never be revealed. Determining the factors affecting dystocia in dairy cattle can help to find the weaknesses of the management of the dairy cattle industry and the resulting problems. Therefore, it is necessary to identify and manage the most important of the various genetic and non-genetic factors that affecting dystocia. The feature selection algorithm is one of the data mining methods that can be useful in this field. This study aimed to determine the most important factors affecting the dystocia of Holstein dairy cows using the feature selection algorithm. The total records were 413205, along with 14 features related to reproductive and production records. Data analyzed with four important methods of feature selection algorithm (Best-First, Greedy-Stepwise Genetic-search, and Ranker) and nine different models (CFS-Subset-Eval, Gain-Ratio-Attribute-Eval, etc.) to determine the most important factors. The results showed that the genetic-search method using Naive Bayes Tree classification of the feature selection was the most appropriate method for selecting the factors affecting dystocia with the lowest error rate (AMS = 0.011, ARAE = 0.001). The most important factors affecting dystocia were milk production, dam age at parturition, parity, gestational length, corrected milk based on 4% fat, calf sex, calf weight, the season of calving, type of calving, fat-to-protein percentage ratio, and open days. It is expected that by considering these factors, dystocia can be better managed in dairy cattle enterprises.

Understanding the regulation of milk fat biosynthesis is vital to the development of nutritional strategies to increase the nutritional value of milk, reduce milk-energy out-put and improve the energy balance of lactating cows. The objective of this study was to determine the most important fatty acids affecting milk fat of Holstein cows using feature selection algorithm. Feature selection algorithm is one of the data mining methods which is used to determine the most effective and best-known parameters predicting targeted trait. A total of 3072 raw data and three important methods of feature selection with six models was used. Data was analyzed using the WEKA Software (v. 3.8). Results indicated that Ranker method of feature selection algorithm is the most appropriate method to select the most important fatty acids affecting milk fat using LMT classification with minimized error. Accordingly, n-3 C20:3 (eicosatrienoic acid), n-6 C20:2 (eicosadienoic acid), total trans C16:1, trans-10 C16:1, trans-6, 7, 8 C16:1, total trans monounsaturated fatty acids, trans-11, cis-15 C18:2, total trans C18:1, n-6 C20:4, and cis-12 C18:1 were determined as the most important fatty acids affecting milk fat. Feature selection algorithm showed that fatty acids other than trans-10, cis-12 conjugated linoleic acid and trans-10 C18:1 is associated with milk fat biosynthesis which is necessary to be further characterized their biological importance in the next future studies.

Monitoring the milk components of a herd can help assess the health and nutritional status of lactating cows. Fat is one of the most valuable components of milk. Milk fat (proportion and total amount) is typically the most variable component in milk which is influenced by various physiological and environmental factors. A herd milk fat test below 0.3% of breed average can indicate a problem situation. Therefore, diets that allow for an improvement in milk fat output would potentially be economically advantageous. In order to maintain milk fat, it is necessary to identify, monitor, and measure the most important factors influencing milk fat. Feature selection algorithm is one of data mining methods and to the best of our knowledge no available literature to use this algorithm to refine most contributing factors affecting milk fat in Holstein dairy cows. Objective of this study was to determine the most important factors affecting milk fat of Holstein cows using feature selection algorithm.
In the current study, a total of 2112 raw data obtained from 66 factors (which might potentially able to affect milk fat) was used. Data was evaluated by the three important methods (Best-First, Greedy-Stepwise, and Ranker methods) and six models (CFS-Subset-Eval, Info-Gain-Attribute-Eval, Gain-Ratio-Attribute-Eval, Symmetricer-Attribute-Eval, RelifeF-Attribute-Eval, Principal-Components) of feature selection algorithm. Data was analyzed using the WEKA Software (v. 3.8).
Results indicated that both Best-First and Gready-Stepwise methods of feature selection algorithm are the most appropriate methods to select efficient factors affecting milk fat using Naive Bayes classification with minimized error. Accordingly, blood non-esterified fatty acid, total volatile fatty acids of rumen fluid, ether-extract intake, time spent rumination per kilogram of neutral detergent fiber intake, time spent rumination per kilogram of forage neutral detergent fiber intake, total chewing time per kilogram of neutral detergent fiber intake, body weight, body condition score, back fat thickness, and blood glucose were the most important factors affecting milk fat.
It is possible to monitor and measure of some limited factors selected by feature selection to efficiently maintain milk fat on-farm and save time and cost.