Here’re the ten simplified meanings of Machine Learning Key Terms. Frankly, you need these terms memorized. Whether you work as a professional or search for a job, these terms matter. ML provides a framework for solving real-life problems and offers valuable insights.
Machine learning key terms
| Word | Definition |
|---|---|
| 1. instance or example | A single object, observation, transaction, or record. |
| 2. target or label | The numerical or categorical attribute of interest. A model predicts this variable for each new instance. |
| 3. features | The input attributes that a model uses to predict the target. These attributes may be numerical or categorical. |
| 4. model | A mathematical object that describes the relationship between the features and the target. |
| 5. training data | The set of instances with a known target that developers use to fit an ML model. |
| 6. recall | Using a model to predict a target or label. |
| 7. supervised machine learning | Supervised machine learning uses examples with known output values to infer a function that relates input values to the output. |
| 8. unsupervised machine learning | Unsupervised machine learning analyzes unlabeled examples to find hidden structure instead of relying on labeled examples. |
| 9. ML workflow | The stages in the ML process: data preparation, model building, evaluation, optimization, and prediction. |
| 10. online machine learning | A form of machine learning in which the system makes a prediction and updates the model for each new example. |
Also read: Google Cloud Certification
Supervised Vs. Unsupervised
| Instances (Examples) | M/F (Attribute-1) | Salary (Attribute-2) | Target (Label) |
|---|---|---|---|
| instance-1 | M | 10,000 | Eligible for Bonus |
| Instance-2 | F | 20,000 | Not Eligible for Bonus |
Here, the predictable value (Eligibility) has a predefined value. If you know the Target upfront, you use supervised learning. In unsupervised learning, you do not know the target; instead, you use the data to discover hidden patterns and make predictions.
