There is a lot of controversy about artificial intelligence. I imagine it is because of the movies like Terminator, and movies around robots with AI going rogue. But Artificial intelligence is nothing more than Machine Learning. It is about getting a machine to “learn” what is right or wrong based on a set of data presented to it, using a particular algorithm for it to find what is likely right or wrong.
Don’t worry, this is not about machines taking over the world. This is merely about teaching machines given a set of data how to make a good choice, and reporting that information to you. Remember, machines are but simple tools designed to crunch data and provide results for you to make decisions on. Machines need to be told what to do, and can’t make decisions on their own. Trust me, we are no where near to Skynet using machine learning and sending through time a robot to kill Sarah Connor!
There are a number of algorithms machines use to determine whether or not a piece of data fits into a particular category. Some are better than others, but these are:
- Linear Regression
- Decision Trees
- Logistic Regression
- Neural Networks
- Naive Bayes
- Support Vector Machines
None of these would support a robot going back in time. They are merely a way of interpreting data in a graph, and using these algorithms to analyse the graph data in a way that makes sense, through a number of passes to reduce the error rate based on a set of rules.
For more information, please visit this link for a great explanation.