Please enable JavaScript.
Coggle requires JavaScript to display documents.
Methods - Coggle Diagram
Methods
Heuristics
-
-
-
Examples: smoke suggests fire, bad food color, broken ladder.
-
-
-
Expert systems
Can diagnose illnesses, determine jail sentences, or identify car issues.
Advantages: usable without an expert present, consistent, unbiased decisions, transparent rules in code for fairness.
Examples include a doctor, judge, or engineer.
Disadvantages: experts may struggle to articulate their knowledge, not all situations fit logical tests, errors in expert knowledge lead to algorithm errors.
-
Decision trees
-
To automate a decision tree, convert it into a program.
-
-
Each branch becomes an if...else structure, nested inside each other.
-
Machine learning
Training methods include supervised, unsupervised, and reinforcement learning.
-
-
Supervised training uses labeled data to teach the computer, such as identifying faces.
Unsupervised training lets the computer find patterns in unlabeled data, grouping similar items.
-
-
-
Machine learning is used for face recognition, diagnosing illnesses, and weather forecasting.
-