Artificial neural networks for data mining in animal sciences

摘要

Abstract Background With the advancement in technology the amount of data generated, in almost every sphere of life, is increasing exponentially. This enormous amount of data needs new powerful tools for analysis and inference drawing. One such process is data mining which is the automated extraction of hidden, previously unknown, and useful knowledge from big data. Data mining is crucial as conventional strategies cannot keep up with the analysis of rapidly accumulating data and they are also inflexible in the wake of new challenges. Animal sciences are no exception to the changing scenario, especially when animal farms are quickly becoming more data intensive.Main body of the abstract The amount of data generated on the farms is also growing exponentially as farms become more intensive and mechanized. There is thus a need to utilize the knowledge of multidisciplinary fields like advanced statistics, artificial intelligence, machine learning, and database management, for revamping animal sciences. Artificial neural networks (ANNs) offer a lot of promise in this direction since they are motivated by the distributed, massively parallel computation in the brain. ANNs are powerful machine learning tools that offer multiple advantages for data mining over traditional techniques in being fast, accurate, self-organizing, robust, and highly accepting of noisy and imprecise data. Neural networks are being applied successfully for a myriad of supervised and unsupervised learning applications to draw useful hitherto unknown inferences, patterns, and relationships. Neural networks have been used successfully for pattern recognition, clustering, forecasting, prediction, and classification in animal sciences due to their capacity to learn from data, their nonparametric nature, and their ability to generalize well. Today ANN computing is a major element within any data mining tool kit. Popular methods used for neural network computing include feed-forward networks, feedback networks, and self-organization networks. ANN also offers powerful and distributed computing architecture, especially under a scenario where the data are readily available in significant quantity.Short conclusion This paper gives an overview of ANN and their applications in animal sciences and reviews major research conducted in this new and exciting area of artificial intelligence. Research in many aspects of ANN in Animal Sciences has been conducted globally although there is scope for more research in aspects of animal health, monitoring, breeding as well as nutrition .

Artificial neural networks for data mining in animal sciences

摘要

著录项

相关主题

期刊订阅