Understanding how computers learn from data to make decisions.
What is AI?
AI systems perform tasks that usually require human intelligence.
Rule-Based
Follows strict instructions.
IF x THEN y
Machine Learning
Learns patterns from data.
Data -> Model
The AI Workflow
- Data: Gather examples (images, text).
- Training: The computer looks for patterns.
- Model: The "brain" created by training.
- Confidence: A score (%) of how sure the AI is.
Bias & Ethics
Bias
If training data is one-sided (e.g., only photos of apples), the AI will fail to recognize other things (e.g., pears).
Impact: Unfair decisions in hiring, facial recognition, or news.
AI Essentials
| Term | Definition | Why it Matters |
|---|---|---|
| Machine Learning | Computers learning from data rather than rules. | Solves complex problems like voice recognition. |
| Training Data | Examples used to teach an AI model. | Quality determines how "smart" the AI becomes. |
| Confidence Score | Percentage showing certainty of prediction. | Helps users decide whether to trust the output. |
| AI Bias | Unfair decisions due to one-sided data. | Can lead to discrimination in society. |