Created by Steve Hiscock
about 11 years ago
|
||
Question | Answer |
Supervised Learning | Where the teacher helps the system to construct a model by defining classes (values). To predict or estimate a continuous target using past data |
Unsupervised Learning | Learning from observation and discovery. The system is supplied with objects but no classes. It looks for patterns in the data |
Classification | classifying credit applicants as low, medium, or high risk. Putting things into well defined groups |
Estimation | Guessing the number using past outcomes |
Prediction | Predicted value or estimated future value. Can only be tested at a later time. |
Affinity Grouping / Rule Association | Determining things that go together. People who buy A are likely to buy B with probability (x) |
Market Basket Analysis | Association of transactional data grouped together as a single observation. (one record) A single probability determined by a group of actions |
Clustering | Clustering is the task of segmenting a diverse population into a number of subgroups or clusters which contain people or items which are similar |
Want to create your own Flashcards for free with GoConqr? Learn more.