Machine learning is the process where a machine learns from its experience. A machine learning program works by receiving data and learning patterns represented in the data.
Example: A machine learning program that inputs historic housing data for the city of Washington, DC (learning) can predict the value of an additional home in Washington, DC, not already represented in the data (prediction from pattern recognition).
Advantage: The advantage of machine learning is its ability to recognize patterns with vast amounts of data. This is impossible to do with traditional statistical methods.
Disadvantage: A machine learning program is only as “smart” as the data you give it. If a program receives information not represented in the learning data, it will produce inaccurate results because it has not been presented with the scenario. Continuing the housing example, imagine trying to predict the value of a home in Boston, MA. The machine learning’s prediction will produce an inaccurate prediction since the data used to create the pattern recognition represent homes in Washington, DC.
Machine Learning and Neural Networks
Neural network is another term often used in machine learning and artificial intelligence. A neural network is a method for creating software to support the learning of patterns in machine learning.
A neural network is modeled to mimic the way our brain functions. Neurons are connected to a set of inputs, and the inputs receive data and find patterns in the data. The advantage of neural networks is its ability to recognize patterns from an almost infinite number of combinations of data.
AI is a machine that can perform tasks characteristic of human intelligence, such as learning, planning, and decision making.
An important distinction between artificial intelligence and machine learning is intelligence. A machine learning program does not exhibit abstract thinking or have self-awareness. On the other hand, artificial intelligence takes pattern recognition and applies a level of intelligence to it (characteristic of human intelligence).
There are two categories of artificial intelligence, applied artificial intelligence and general artificial intelligence:
Applied artificial intelligence is used to perform a specific task (such as self-driving cars or playing chess). Whereas general artificial intelligence is having the ability to think more broadly, and outside of its trained knowledge. That is, similarly to humans, the artificial intelligence learns and acts on its own.
The terms machine learning and artificial intelligence are often used interchangeably, particularly in marketing. Just remember, machine learning works on pattern recognition. Artificial intelligence builds on machine learning by adding intelligence to patterns.