Machine Learning For People In Hurry
A guide for executives and senior leaders
--
Introduction: What is machine learning
Machine learning is the science of enabling computers to find or suggest solutions without being programmed for every possible scenario. In other words, it is a branch of artificial intelligence (AI), which uses statistical techniques and computational methods which enable the machine to learn from past data sets and use the knowledge to predict the possible outcome for new sets of data and help in making intelligent decisions.
Why should you care about machine learning now anyway?
In research papers, AI and machine learning are mentioned for many decades. Due to the exponential increase in computational power, software capabilities and storage space, machine learning has become ubiquitous. Machine Learning is now everywhere, from your smartphone to your car. Machine learning myriad use cases cover a broad range from image recognition, outcome prediction, fraud detection, computer vision and autonomous driving, chatbots with semantic understanding and many more. It is disrupting every individual and business in a positive sense by discovering new intelligent ways to do things.
What are the different types of machine learning?
Machine learning is categorised into three main types supervised learning, unsupervised learning and reinforcement learning.
Supervised Learning — A large dataset of input variables and results is provided for the machine to learn the relationship between input and the outcome. In the training phase, input parameters and the result are pre-labelled for the machine. A trained model can then predict the results for new input data.
Popular algorithms:- Linear regression, Logistics regression, Random forest, Support Vector Machine, Naive Bayes.
Use Case:- Predictive maintenance of production line in factories, forecast raw materials requirement in manufacturing company, predict diseases based on initial symptoms and test results etc.
Unsupervised Learning — Unlike supervised learning, in the unsupervised learning…
