Humans learn by reading or by making observations in and around them. Since our early days we have been learning in the same fashion. Humans use their knowledge for improvising human life by calling upon various revolutions most significant of them being the technological revolution. This technological revolution has got a huge, impressive history and has been a journey for all humans. In this technological evolution that the man-kind has experienced, the idea of inculcating learning abilities into machines remains to be novel. The ability of a machine to learn from it’s experience and give an output accordingly accounts for what is called as machine learning. Experience here means that the machine learns from the training dataset given to it. Machine learning is a subset of artificial intelligence and deep learning is a small component of machine learning. It is said that machine learning is the next internet.
The word machine learning came into being in 1959. This word was coined by Arthur Samuel from America, who was from IBM and also pioneered in the computer gaming and artificial intelligence field. In the 1960’s, Nilsson published a book which was a representative book for machine learning which consisted of machine learning for pattern classification. “A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E” was rightly quoted by Tom M. Mitchell which is a more direct definition of algorithms which is specifically studied in machine learning.
It has wide applications in varied fields such as:
Medical science and health care
- Corona detection: Nowadays the pandemic disease Corona or also known as Covid-19 has created chaos all over the world. There was a panic situation between doctors. At the initial stage of this disease spread, no device was available for the doctors to test the patients. In such a situation many apps and instruments were introduced to people. While making the instruments or apps, specific types of algorithms were inculcated in them which belonged to classification techniques. There is a recent example of Israel devising a corona testing app that tells you to count from 1 to 20 and you have to cough 5 times and within 30 seconds it gives the report specifying whether you have been tested positive or negative for the novel Coronavirus. This has yet not been released out for public usage as far as India is concerned. Scientists are still looking out for the accuracy of the results provided by this app.
- Cancer detection: Cancer is characterised as a heterogenous disease. In early days there was no machine as such to detect a cancer patient. This led to many biomedical and bioinformatics research teams to work in the field of machine learning so as to classify patients into high or low risk groups. They devised a system that could predict whether the person is prone to cancer based on various inputs that are provided to the system.
- Genetics (gene editing): Genome comprises four nitrogen bases namely; adenine (A), guanine (G), cytosine (C), thymine (T). There is a specific way of making alterations in DNA, which is formed by the above mentioned bases, known as Gene editing; it is done on cellular or organismic level. CRISPR is a technology which offers an effective way to conduct the editing of genes. Desktop genetics is a London based software company at the convergence of AI and machine learning.
Predicting disasters
ML is also used in predicting disasters on a large scale. By studying the nature of a specific area and the harms faced, algorithms fed into systems of machines give out predictions and alert the areas that are at higher risks.
Cyber security
Cyber security has a lot to do with machine learning like for face detection, spam detection, social network analysis, email spam, malware filtering, online fraud detection, self driving cars, market segmentation, e.t.c.
In conclusion, machine learning is a leading subject as technology evolves. ML is not only used in the medical field or in cybersecurity but also is used in all fields, in minor or major form.