More than 20 years have passed since a computer program defeated the reigning world champion in a game that is thought to require a high level of intelligence to play. Deep Blue, an IBM computer program, defeated world chess champion Gary Kasparov. That was probably the time when the most people paid serious attention to a rapidly evolving field in computer science, or more specifically artificial intelligence — i.e. machine learning (ML).
Machine learning is now a mature technology area with applications in almost every aspect of life. It can recommend toys to toddlers in the same way that it can recommend a technology book to a geek or a rich literary title to a writer. It forecasts the future market to assist novice stock traders in competing with seasoned stock traders. It assists an oncologist in determining whether a tumour is malignant or benign. It aids in the optimization of energy consumption, thereby contributing to the cause of Green Earth. Google has emerged as a front-runner in machine learning and artificial intelligence research, with Google's self-driving car and Google Brain being two of its most ambitious projects in its journey of machine learning innovation. In a nutshell, machine learning has become a way of life, regardless of which aspect of life we examine closely.
In the 18th and 19th centuries, the first machine learning techniques were developed. The earliest relevant work was created in 1763. Two years after his passing, Thomas Bayes' essay "An Essay towards solving a Problem in the Doctrine of Chances" was published that year. This is the fundamental piece of work on which the Bayes Theorem and a number of machine learning algorithms are based. The French mathematician Pierre-Simon Laplace actually formalised the Bayes theorem in 1812. The foundational idea for regressive problem solving, the method of least squares, was formalised in 1805. Andrey Markov developed the idea of Markov chains in 1913.
However, Alan Turing's seminal work from 1950 is thought to mark the real beginning of focused work in the field of machine learning. In his paper 'Computing Machinery and Intelligence' (Mind, New Series, Vol. 59, No. 236, Oct., 1950, pp. 433—460) Turing posed the question 'Can machines think?' or you can say, 'Do machines have intelligence?'.
He was the first to suggest that artificially intelligent machines could "learn" and develop. In 1952, IBM researcher Arthur Samuel began developing checkers-playing computer programmes as a first step in his work on machine learning. Frank Rosenblatt created the first neural network simulation programme in 1957. The development of machine learning has been fascinating over the ensuing 50 years. The nearest neighbour algorithm was developed in 1969, recurrent neural networks in 1982, support vector machines and random forest algorithms in 1995, and many other machine learning algorithms have been developed by various researchers. Google's AlphaGo programme, which used machine learning to defeat a skilled human Go player, is the most recent achievement in machine learning research.
Evolution of Machine Learning
Types of Machine Learning
Difference between Supervised, Unsupervised, and Reinforcement Learning
Machine Learning Activities
Types of Data in Machine Learning
Basics of Feature Engineering
Classification
Classification Learning Steps
k-Nearest Neighbour (kNN)