17Statistical Learning
Learning is not compulsory …neither is survival.
W. Edwards Deming (1900–1993)
A general type of artificial intelligence, called machine learning, refers to techniques that sift through data and find patterns that lead to optimal decision rules, such as classification rules. In a way, these techniques allow computers to “learn” from the data, adapting as trends in the data become more clearly understood with the computer algorithms. Statistical learning pertains to the data analysis in this treatment, but the field of machine learning goes well beyond statistics and into algorithmic complexity of computational methods.
In business and finance, machine learning is used to search through huge amounts of data to find structure and pattern, and this is called data mining. In engineering, these methods are developed for pattern recognition, a term for classifying images into predetermined groups based on the study of statistical classification rules that statisticians refer to as discriminant analysis. In electrical engineering, specifically, the study of signal processing uses statistical learning techniques to analyze signals from sounds, radar, or other monitoring devices and convert them into digital data for easier statistical analysis.
Techniques called neural networks were so named because they were thought to imitate the way the human brain works. Analogous to neurons, connections between processing elements are generated dynamically in a learning ...
Get Nonparametric Statistics with Applications to Science and Engineering with R, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.