Machine Learning (ML) is a term used to group the concepts and techniques to teach computers to learn by themselves. The computers learn: a) Based on observation of data ingested and b) Through the feedback on whether the computer has reached the right or wrong conclusion.
An example would be to teach a computer to recognize animals in a picture. The computer will learn from a broad set of known data (in the example, images with known animals) and will be able to arrive at conclusions based on such learning.
Machine Learning is part of a broader concept. Artificial Intelligence (AI), and has three main components
- Large amounts of characterized data. These are known data related to the purpose of the learning. Part of this data is used to train the Machine, and the other part is used to check the results of the training. In the example above, say we have one thousand pictures with and without cats of which we use 800 to train the model to recognize images with cats, and 200 are used to check whether the model can identify those images that have cats.
- An algorithm or model. A simplification would be to say that these are models that similarly classify the learning data that the brain does when learning (as explained below). Each model defines a “reasoning” way of the learning process and the “reasoning” process to conclude.
- Feedback loops that improve the learning process and the model capabilities and precision by receiving feedback on whether the conclusion the model achieved was right or wrong.
Machine Learning (ML) can be applied to teach computers on many of the intuitive, automatic decisions made by the human being such as understanding text or speech, recognizing patterns, play games or engage in a focussed dialog.
There are many deployed applications of ML in the real world, from automatic customer assistance (with text/speech recognition and problem-solving algorithms) to picture content discovery (like Pinterest), improving e-commerce conversion rates or effective countermeasures for cyber attacks.
Machine Learning algorithms such as “Neural Networks” or Deep Learning” frequently appear in many articles and speeches. Machine Learning should not be confused with Expert Systems.
READ ARTIFICIAL INTELLIGENCE (AI) AND EXPERT SYSTEMS DEFINITION IN THESE CARDS
Any “BUZZWORD” you do not find?
Suggest a Buzzword you would like DigitalFullPotential.com to include in this section
Suggest a definition for another “BUZZWORD”
If you want to contribute to the definition of a Buzzword not yet included at DigitalFullPotential.com, you are welcomed.
If you so like it, we will acknowledge your contribution with your name and the link to your LinkedIn/webpage. If you prefer to do it anonymously, it is up to you.