Artificial Intelligence (Copy)

Definition:
- Artificial Intelligence (AI) is a branch of computer science focused on creating systems that can perform tasks normally requiring human intelligence.
- It involves the simulation of intelligent behaviour by computers.
Key Idea:
- AI systems are programmed to collect and process data, apply rules, reason logically, and in some cases, learn and adapt from experience.
Examples:
- Voice assistants (e.g., Siri, Alexa, Google Assistant)
- Chatbots for customer service
- Self-driving cars
- Facial recognition software
- Recommendation systems (e.g., Netflix, YouTube)

Data Collection:
- AI requires large amounts of data to operate effectively.
- Data can be structured (databases, spreadsheets) or unstructured (images, videos, audio).
Rules for Using Data:
- AI systems have algorithms that define how collected data is processed and applied.
- Example: A medical diagnosis AI might compare patient symptoms with its database of known illnesses.

AI can make decisions based on available data and programmed rules.
Uses logic-based reasoning to solve problems.
Example: An AI chess program predicts the best move by evaluating potential outcomes.

Some AI systems can improve over time by learning from mistakes and adapting strategies.
This is the basis of machine learning.
Example: A spam filter gets better at detecting junk emails as it processes more examples.

Definition:
- A computer program that mimics the decision-making ability of a human expert in a specific field.
Structure:
1. Knowledge Base – Stores facts and information about a particular subject area.
  - Example: In a medical expert system, this might contain diseases, symptoms, and treatments.
2. Rule Base – Contains rules (IF…THEN statements) that dictate how the knowledge is applied.
  - Example: IF “fever” AND “rash” THEN “possible measles”.
3. Inference Engine – Applies the rules to the knowledge base to deduce new facts or reach conclusions.
  - Example: Works like a reasoning brain that uses the rules to provide advice or diagnoses.
4. User Interface – Allows users to interact with the system (input data and receive output).
  - Example: A question-and-answer form for a medical diagnosis tool.
Advantages:
- Consistent decision-making
- Can store vast knowledge from multiple experts
- Available 24/7
Disadvantages:
- Lacks human intuition and creativity
- Needs regular updating
- Expensive to develop

Definition:
- A subset of AI where systems automatically adapt processes and data handling without explicit reprogramming.
How it Works:
1. Training Phase – The AI is fed with large datasets to identify patterns.
2. Testing Phase – The AI applies learned patterns to new data to make predictions or decisions.
3. Feedback and Adjustment – The AI refines its methods to improve accuracy over time.
Examples:
- Predictive text on smartphones
- Image recognition systems
- Financial fraud detection systems
Advantages:
- Improves over time
- Can process huge amounts of data quickly
- Finds patterns humans might miss
Disadvantages:
- Needs large datasets for accuracy
- Can develop bias if training data is biased
- Complex to understand and troubleshoot

Input:
- Data from sensors, databases, or user input.
Processing:
- AI applies rules (from the rule base) and/or learned patterns to the input.
Output:
- Produces a decision, recommendation, or action.
Feedback Loop (if machine learning is involved):
- Uses results and outcomes to improve future performance.

Medicine:
- AI assists doctors in diagnosing diseases by comparing patient data with medical records.
Transport:
- Autonomous vehicles navigate traffic using AI decision-making.
Agriculture:
- AI-powered drones analyse crop health and recommend treatments.
Banking:
- AI detects suspicious transactions to prevent fraud.

Want To Teach Online?