Artificial Intelligence (AI)

Q1. a)  Describe what is meant by  Artificial Intelligence (AI). 

⇒  Artificial intelligence refers to the simulation of human intelligence in computer system that are programmed to think like humans and mimic their actions.

⇒  AI is based on cognitive functions of human brain; the mental process of acquiring knowledge and understanding through thought, experience and the perception of five senses.

⇒  The goals of artificial intelligence include computer-enhanced learning, reasoning, self-correction, perception, and problem solving.

Note : Perception means the ability to see, hear, or become aware of something through the senses.

b)  What are the three types/categories of artificial intelligence?

1. Artificial narrow intelligence (ANI), designed to perform single or very few tasks with superior capability than human.

Example : Speech recognition, Facial recognition, image classification, expert systems, etc.

2. Artificial general intelligence (AGI), designed to perform any intellectual task that a human can, but they are not superior to a human.
3. Artificial super intelligence (ASI), designed to perform any intellectal task with capabilities superior than humans.

Note : AGI and ASI system is the subject of current research in AI and can’t be achieved for now, as we don’t have complete knowledge of our brain.

c)  Give three examples of artificial intelligence.

1. Self-driving cars.
2. Smart assistants.
3. Healthcare management.
4. Manufacturing robots.
5. Web-crawlers or online search engines.
6. Marketing chatbots.

d)  Give three examples of artificial intelligence that you may find in a home.

1. Virtual Reality gaming.
2. Smart Home devices capable of voice interaction, (such as Amazon Alexa, Google Now, Apple Siri and Microsoft Cortana) which could play music, control several smart devices, providing real-time information such as news, weather, traffic, etc.
3. Robot Vacuum Cleaner.
4. Security camera that use facial recognition.
5. Smart Thermostat, automatically adjusts temperature in home for optimal performance.

Q2.  Describe four main  characteristics  of artificial intelligence.

Artificial intelligence focuses on four cognitive skills: learning, reasoning, self-correction and perception of senses.

1. Learning and adaptation : capable of learning from experience, improve its performance over time and adapting to its surroundings.
2. Reasoning and problem-solving : use algorithms that can analyze data, make predictions, and recommend actions with reason.
3. Self-correction : ability to continuously detect the right approach from new set of data, fine-tune algorithms to improve the results each time.
4. Perception of sensing : ability to perceive and sense their environment using sensors. Interpret visual and auditory information, recognize patterns, and understand natural language.

Note : Perception means, the ability to see, hear, or become aware of something through the senses.

Perceive means, ability to realize or understand something through consciousness.

Q3. a)  Describe what is meant by  Machine learning. 

⇒  Machine learning is an application of artificial intelligence that enables a machine to learn through data acquisition, make predictions, and improve themselves through past experience without programming explicitly.

⇒  It could offer quick decisions and predictions due to very powerful processing capability.

⇒  Capable to manage and analyse considerable volumes of complex data which humans would take years to complete.

Example :

• Medical Diagnosis
• Virtual Assistants (like, Alexa, Siri, etc.)
• Autonomous Vehicles.
• Voice Recongnition.
• Image Recognition.
• Online Chat-bots.
• Online Search-engine.

b)  Give difference between Artificial Intelligence and Machine learning.

✬  Artificial intelligence is a technology that enables a machine to think like humans and mimic their actions, whereas Machine learning is a technology or algorithms that enable a machine to learn through data acquisition, make predictions, and improve themselves through past experience without programming explicitly.

c)  Give three examples of Machine learning.

1. Search engine
2. Categorising email as spam.
3. Recognising user online shopping behaviour.
4. Detection of fraudulent activity.
5. Voice recognition.
6. Speech recognition.
7. Image recognition.
8. Self-driving cars.

Q4.  Describe  how a search engine might use machine learning  to determine the most appropriate results based on a user's search criteria.

⇒  The Search engine uses "Search bots", a machine-learning algorithm, that crawls through all websites to search, locate and display the list of websites that matches with the user's search criteria.

⇒  If the user selects one of the websites found on page-1, then search-engine classes this as a success and the same machine learning algorithm can now be used for similar search criteria.

⇒  If the user has to go to page-2, 3 or 4 to find the information they are looking for, then Search-engine classes as a failure.

⇒  The machine learning model is then fully tested with known data (search criteria) and known outcomes.

⇒  The system is modified if it hasn't met its criteria to search for appropriate website.

Q5.  An Expert system is a type of artificial intelligence that is able to provide a level of expertise about a certain subject.

a)  Describe what is meant by  Expert system. 

⇒  Expert system is an application of artificial intelligence that enables a machine to mimic the decision-making ability of human or organisation with expert's knowledge and experience.

⇒  It collects and stores expert's knowledge, facts and rules in a knowledge-base and integrate them with inference engine.

⇒  The Inference engine would ask a series of questions to the user and applies the set of 'If-then' rules over knowledge-base to provide solution to problems or answer questions.

b)  Describe the following parts of expert system.

1)   User Interface : 

⇒  method by which the expert system interacts with a user.

⇒  interaction can be through dialogue boxes, command prompts or other input methods.

⇒  the questions being asked usually only have "Yes/No" answers and are based on the responses to previous questions.

2)   Knowledge base : 

⇒  an organized collection of facts and knowledge about a particular subject obtained from different human experts.

⇒  it is basically a collection of objects (an item) and their attributes (defines the objects).

3)   Inference engine : 

⇒  the main processing element of expert system that act like a search engine examining the knowledge base for information that matches the queries.

⇒  it is responsible for gathering information from the user by asking a series of questions and applying responses where necessary; each question being asked is based on the previous responses.

⇒  it make use of inference rules stored in the rule base to find the information from the knowledge-base and provide solutions to the user.

4)   Rules base : 

⇒  the rules base is a set of inference rules.

⇒  inference rules are used by the inference engine to draw conclusions.

⇒  they follow logical thinking; usually involving a series of 'IF' statements, for example:

IF Continent = "South America" AND Language = "Portuguese" THEN Country = "Brazil"

c)  Give three  applications of expert system. 

1. Diagnosis of a patient's illness.
2. Oil and mineral prospecting.
3. Fault diagnostics in mechanical and electronic equipment.
4. Tax and financial calculations.
5. Identification of plants, animals, and chemical compounds.
6. Strategy games, such as chess.
7. Logistics (efficient routing of parcel deliveries).

Q6.  Describe  the steps in setting up an expert system. 

1. Information needs to be gathered from human experts or from written sources or the internet.
2. Create the knowledge base and store the gathered information.
3. Create a rule base with a series of inference rules so that inference engine can draw conclusions.
4. Set up the inference engine which is the main processing element that makes reasoned conclusions from data in the knowledge base.
5. The user-interface need to be developed to allow the user and the expert system to communicate.
6. Once the system is set up, it needs to be fully tested; this is done by running the system with known outcomes so that results can be compared and any changes to the expert system is made.

Q7. a)  Give three  advantages  of  Expert system. 

1. They offer a high level of expertise.
2. They offer high accuracy.
3. The results are consistent.
4. Very fast response time; quicker than a human expert.
5. They provide the percentage of accuracy of any suggested solution.
6. They have the ability to store vast amount of ideas and facts.

b)  Give three  disadvantages  of  Expert system. 

1. Users of expert system need considerable training.
2. The set up and maintenance costs are very high.
3. They tend to give very 'cold' response (not affectionate) that may not be appropriate in certain medical situations.
4. They are only as good as the information/facts entered into the system.

Q8.  A robot is designed and programmed using Artificial Intelligence (AI) to find its way through different puzzles. Each puzzle has a series of paths that the robot needs to follow to find its way to the end of the puzzle. The puzzle contains dead ends and obstacles, so the robot needs to decide which way to go.

a)  Describe the characteristics of Artificial Intelligence (AI).

• Collecting data
• Stores rules for using the data
• The ability to reason
• The ability to learn
• The ability to adapt
• The ability to change its own rules
• The ability to change its own data

b)  Explain how the program will use Artificial Intelligence (AI).

⇒  Program need to use machine learning algorithms that learns and adapt its behaviour based on new data and rules by itself.

⇒  Collect data about different routes, objects, obstacles or problems while moving around and store it in its knowledge base.

⇒  Create new rules or update its rules at its rule-base to perform its task successfully.

⇒  Store the successful actions, so that it would know what is most likely to work next time.