Machine learning

Machine learning is one of the most intriguing and exciting areas of data science. It conjures all forms of images around artificial intelligence which includes Neural Networks, Support Vector Machines (SVMs), and so on.

Fundamentally, we can describe the term machine learning as a method of training a computer to make or improve predictions or behaviors based on data or, specifically, relationships within that data. Continuing, machine learning is a process by which predictions are made based upon recognized patterns identified within data, and additionally, it is the ability to continuously learn from the data's patterns, therefore continuingly making better predictions.

It is not uncommon for someone to mistake the process of machine learning for data mining, but data mining focuses more on exploratory data analysis and is known as unsupervised learning.

Machine learning can be used to learn and establish baseline behavioral profiles for various entities and then to find meaningful anomalies.

Here is the exciting part: the process of machine learning (using data relationships to make predictions) is known as predictive analytics.

Predictive analytics allow the data scientists to produce reliable, repeatable decisions and results and uncover hidden insights through learning from historical relationships and trends in the data.