Basic Terms of Machine Learning

1. Data: records about the subjects of the study

2. Data set: a collection of records

3. Instance/sample: a description of each event or object

4. Attribute/feature: A matter that reflects the performance or nature of an event or object in some way.

5. Attribute value: the value of the attribute

6. Attribute space/sample space: the space formed by attributes

7. Dimensionality of the sample (dimensionality): The number of attribute descriptions of the sample.

8. Learning/training: Learning a model from data.

9. Training data/training sample/training set: The set of data, samples and samples used in training.

10. Hypothesis: The underlying pattern of the learned model corresponding to the obtained data.

11. Ground-truth: The underlying law itself.

12. Label: Information about the example results.

13. Example: Include an example of markup.

14. Label space/output space (label space): The collection of all labels.

15. Classification: Predicting discrete values.

16. Regression: Predicting continuous values.

17. Testing: After learning the model, use the model to make predictions.

18. Testing sample: The sample being tested.

19. Clustering: Divide the data in the training set into multiple groups.

20. Cluster: A group divided by cluster analysis.

21. Supervised learning: The training data has labeled information.

22. Unsupervised learning: The training data has no labeled information.

23. Generalization: the ability of the learned model to apply to new samples

24. Distribution: The distribution of samples in the sample space