Machine Learning
Machine Learning is ability of computers to learn without being explicitly programmed. Therefore, machine learning algorithm can learn from and make new groups on data, root cause analysis on data, predictions and alternative decision set prescription on data. Instead of depending on manual logic elaboration and encoding that in strictly static program instructions, Data Science/ML/Data Mining helps by making data-driven or automated decisions, through building a model from sample inputs.
Resurging interest in machine learning is due to the same factors that have made artificial intelligence, data mining and Bayesian analysis more popular than ever. Things like growing volumes, velocity and varieties of available data was a challenge, but now computational processing that is cheaper and more powerful and affordable data storage are now making it implementable.
All of these things mean it's possible to quickly and automatically produce models that can analyse bigger, more complex data and deliver faster, more accurate results, even on a very large scale. And by building precise models, an organization has a better chance of identifying profitable opportunities or avoiding unknown risks.
On broader level they can be categorized into four categories:
- Supervised Learning - In this, we provide training data on which machine learn and we examine the accuracy of model on test data (unseen data).
- Unsupervised Learning- In this labelled or train data is not given and model automatically organizes the data.
- Semi-supervised Learning - It is a mix of both supervised and unsupervised in which some data is label but most of the data is not.
- Reinforcement based - The feedback is provided in this kind of ML algorithm.