Nghia Bui, Saigon, Christmas 2016
From the beginning days of computers, people have wondered whether computers can be programmed to learn. If we can make them to learn, the effect would be so amazing. For example, ML is frequently used in cancer diagnosis and detection. Cruz and Wishart [CW06] showed that ML methods can be used to substantially (15-25%) improve the accuracy of predicting cancer susceptibility, recurrence and mortality. Indeed, recently [Ng16], it was reported that IBM’s Watson gave proper diagnosis for Japanese leukemia patient after doctors were stumped for months. The supercomputer, sifted through 20 million cancer research papers, was able to find out the proper diagnosis within 10 minutes, and also suggested a new treatment that has since been more effective.
With an undoubted impression about applications of ML, let our discussion continue with the formal definitions and basic concepts of this subfield of computer science.