Computational Intelligence is a field of computer science that deals with complex real-world problems that cannot be reduced to mathematical algorithms. Computational Intelligence approximates human intelligence and can handle uncertainty and incomplete information, adapt to new situations, learn from mistakes, and find the optimal way to solve a problem.
Artificial Intelligence is a field of computer science that applies deductive and inductive reasoning to do human things like visual perception, speech recognition, or creative problem-solving. An Expert System can be created that has specific knowledge about its domain and a set of rules or heuristics to enhance its ability, similar to human experts.
Machine learning is a branch of computer science that develops methods for a computer to operate without a detailed plan of action, to learn new approaches to a problem from its actions and data. Machine learning methods are used to discover previously unrecognized patterns in data, find a better path from start to finish (optimization), or to make predictions about system outcomes and behavior.
Modern computers have so much memory that databases (collections of facts) can be difficult to manage. This is compounded when databases are distributed over hundreds or thousands of separate machines. Big data techniques have been developed to help us collect, curate, update, and share such data, and to find important and useful patterns in an otherwise
overwhelming sea of facts.
Optimization refers to computational methods to find better solutions. Optimization draws from machine learning and artificial intelligence techniques to handle logistical problems like scheduling, shipping, workflow, and resource management. These are complex problems that require machines to compare thousands of possible paths to a desired outcome.
Predictive analytics is the use of historical data, statistical modeling, and machine-learning algorithms to predict the likelihood of possible future outcomes.