Artificial Intelligence | O Level Computer Science 2210 & IGCSE Computer Science 0478 | Detailed Free Notes To Score An A Star (A*)
Introduction and Characteristics of AI
- Artificial Intelligence (AI)
- Deals with intelligent human behavior by a computer
- Cognitive functions of human brain
- Mental processes that acquire knowledge, understand things through thought, experience and five human senses.
- Can be done by machines
- Measured against human capability level to determine the benchmark
- Such as reasoning, speech and sight.
- Characteristics of AI
- Collection of rules and data
- Has the ability to learn, reason and adapt to external stimuli
- 3 main categories
- Narrow AI
- Machine has superior performance to human when doing one specific task
- General AI
- Similar but not superior in its performance of a specific task to humans
- Strong AI
- Machine is superior in performance of many different tasks compared to humans
- Narrow AI
- Reasoning
- The ability to draw reasoned conclusions from given data/ situations
- Deductive reasoning
- Number of correct facts are built up to form a set of rules
- These rules can be applied to other problems
- By a sequence of steps
- AI can learn
- Know how to do the task more effectively the next time
- Apply the learning to new and novel situations
- AI system is capable of learning and adapting to its surroundings
- AI can quickly discern patterns
- Sometimes, even humans cannot do this.
- Make predictions by adapting the new data.
- Examples of AI
- New generation of live news feeds
- Smart home devices
- Interacts with humans recognizing different verbal commands
- Learns from the environment and the data it receives
- Becomes increasingly sophisticated in its responses
- Automated repetitive learning occurs
- Use of chatbots that interact through instant messaging
- Artificially replicating patterns of human interactions suing AI to respond to typed or voice messages.
- The chatbot responds to using information known at the time
- Autonomous cars
- Factual expression recognition
- Algorithms identify key facial landmarks
- A combination of these landmarks are used to map emotions
AI Systems
- Expert system
- Computer system that mimics the decision making ability of humans
- Simulate judgement
- Simulate behavior of organizations or humans with expert knowledge and experience.
- Solve problems or answer questions that would normally require a human expert
- For example, an expert system performing a diagnosis
- Provides a percentage of accuracy of its conclusions
- Advantages
- High level of expertise and accuracy
- Consistent Results and the ability to store vast amounts of ideas and facts
- Traceable logical solutions and diagnostics
- Multiple expertise
- Fast response times, much faster than human experts
- Unbiased reporting and analysis of facts
- Probability of suggested solution being correct is provided
- Disadvantages
- Machine learning
- Training computers with sample data
- Can make predictions about new unseen data
- Without the need to program them for any new data
- Machine learning can be supervised or unsupervised
- Supervised means that user is telling the program what its data means
- Unsupervised means that data is input and then the program learns from the data
- For example, clustering
- Plotted on a graph
- Program identifies which items of data are close to each other.
- Image recognition in shape recognition
- Without human interaction
