Avel-COVID19-Comp Sc-HND3 COURSE- ARITIFICIAL INTELLIGENCE & EXPERT SYSTEM
COVID – 19 SIT@HOME
INTEL. DEV. CLASS
HND COMPUTER SCIENCE
ARITIFICIAL INTELLIGENCE & EXPERT SYSTEM
STEP BY STEP BREAKDOWN
Committee of Intellectuals [CoI]
INTELLECTUAL DEVELOPMENT CONFERENCE CENTER [INTEL CENTER]
Artificial Intelligence is a field of science and technology dependent on control and based on discipline such as science, psychology, and engineering.
QUESTION: AI is an aspect of computer science that deals with intelligent behaviour, learning and adaptation in the machine. Discuss.
AI is the science and engineering of intelligent machines, especially computer programs which is used to perform specific task in which the problem is encapsulated. AI is application as differing as medical diagnosis, computer, search engine, voice and handwriting recognition. AI is depicted as a digital machine or computerized control robot which is utilized to perform task generally connected with an intelligence machines.
QUESTION: A rule-based system is a relatively simple model that can be adopted to any number of problems. Hence, you are required to:
(i) Highlight how to create a rule-based system for a given problem
– The set of fact to represent the initial working rule
– A set of rules and this include all action that should be taken between the scope of the problem
(ii) Discuss briefly the essential components of a fully functional rule-based system
- Working algorithms: this is a temporary storage location that holds the fact
produced during processing and possibly awaiting other processes produced by inference engine during it activities. It should be noted that the working algorithm contains only facts and these facts are those produced during searching processing.
- Knowledge Based: It is a storage location where the knowledge in a particular domain is kept, i.e. it stores information about the object domain. However, this gives beyond passing collection of record in a database rather it contains symbolic representation of expert knowledge which include definition of terms, inter-connectivities of component entities and cost effect relationship between this component.
- Inference Engine: This is the core of the expert system. T is the part of the expert system that manipulate knowledge based to produce new fact in order to solve a given problem.
A typical rule-based system has four basic components
- A list of rules or rule base, which is specific type of knowledge base.
- An inference engine or semantic reasoned, which infers information or takes action based on the interaction of input and the rule base. The interpreter executes a production system program by performing the following match-resolve-act cycle.
- Match: In this first phase, the left-hand sides of all productions are matched against the contents of working memory. As a result a conflict set is obtained, which consists of instantiations of all satisfied productions. An instantiation of a production is an ordered list of working memory elements that satisfies the left-hand side of the production.
- Conflict-Resolution: In this second phase, one of the production instantiations in the conflict set is chosen for execution. If no productions are satisfied, the interpreter halts.
- Act: In this third phase, the actions of the production selected in the conflict-resolution phase are executed. These actions may change the contents of working memory. At the end of this phase, execution returns to the first phase.
- Temporary working memory
- A user interface or other connection to the outside world through which input and output signals are received and sent.
QUESTION: Briefly describe the two (2) basic goals of neural network research.
(i) Biological Viewpoint: This can be used to replicate and simulate components of the human (or animal brain), thereby giving us insight into natural information processing.
(ii) Technical Viewpoint: Some problems such as character recognition or the prediction of future states of a system require massively parallel and adaptive reasoning.
QUESTION: Briefly explain any five (5) features of an intelligent agent.
(i) Agents are situated: These agents are sensitive to its own surrounding environment and has no knowledge of the full domain of either agents.
(ii) Agents are autonomous or semi autonomous: This means that each agent has:
(i) Certain responsibility to problem-solving
(ii) Little or no knowledge of either what other agents does or how they do it.
(iii) Each agent does his own independent piece or the problem-solving.
(iv) Produces a result itself or generate report back to other agent in the community.
(iii) Society of Agent is Structure: Each individual agent though having its own unique environment and skill set with coordinate with other agent in the overall problem-solving.
(iv) Agents are interactive: Agent formed a collection of individual that co-operate in a particular task, hence they may be seen as society.
(v) Intelligence is emergent: The overall co-operative result can be viewed as bigger as that of some of the individual contributors therefore intelligent is seen as a phenomenon resident in an emergent from a society and not a property of an individual agent.
QUESTION: Consider the following graph.
Starting from the initial state A, to the goal state G, show the order in which the nodes are expanded using a Depth First Search. Assume alphabetically smaller nodes are expanded first to break ties.
Given the search space below, assume the initial state is A and the goal state is G,
i. Draw the state space for the search space below
ii. Using BFS, create a search tree to find a path from the initial state to the goal state, show which node is being expanded at each step, the content of the fringe, and the final solution found.
iii. Using DFS, create a search tree to find a path from the initial state to the goal state, show which node is being expanded at each step, the content of the fringe, and the final solution found
ii. Path is A B C D E F G
iii. Path is A B D E G
QUESTION: Define Artificial Intelligence
This is the intelligence displayed by an Artificial entity with intelligence behaviour, learning and adaptations in machines such a system is generally thought to be a computer
QUESTION: Define Natural Processing
Natural Language processing is a set of program that allow (a certain types) a human computer dialogue, day – to – day language (Natural Language like English, French or Dutch)
QUESTION: Define knowledge representation
A knowledge representation is the representation expected for processing a modern computer. It entails/having the ability to recover, retrieve, accumulation, collection of an organization knowledge.
- Metal knowledge
QUESTION: Define an intelligent agent. Give two examples
An intelligent agent is a system that perceives its environment and takes action that will boost ormaximize its chances of progress. E.g. agent that is utilized to predict share prices and agent that is programmed to detect smoke i.e. smoke detector.
QUESTION: Define the term Artificial Intelligence
Artificial Intelligence is the science of making machines do things that would require intelligence if done by man
- The capacity of human made machine (an automation) to imitate or simulate human methods for the deductive acquisition & application of knowledge & reason.
- Artificial intelligence is the field that reviews the synthesis andexamination of computational agents.
QUESTION: Define intelligence and state the difference approaches for defining artificial intelligence?
Artificial Intelligence can be defined in two ways namely.
- Thought of processes and reasoning
For Thought of Processes and Reasoning
- System that think like human
- System that think rationally
- System that act like human
- System that get rationally
For system that Think like human
Artificial Intelligence can be defined as the automation of activities which is associated with human thinking, such as Decision making, problem solving and learning which was introduced by Bellahm in 1978.
For system that Think Rationally
Artificially Intelligence can be defined as the study of mental faculty through the use of Rational model. It was introduced by Chammak and MC Demoll in 1985.
- System that act like human
- System that act rationally
For system that act like human
Artificial Intelligence can be defined as the act of creating machines that perform a specific or particular function which requires an intelligence when performed by people. It was introduced by Kurweiz in 1990.
For system that act rationally
Artificially Intelligence can be defined as the study and design of an Intelligence agent by role and also concerned with the study of an intelligent behavior in artifact.
Approaches for Artificial Intelligence
- Turning Test Approach: It deals with acting humanly
- Congnetive Modeling Approach: It deals with thinking humanly
- The law of thought approach: Which deals with thinking rationally
- Rational agent (acting rationally) approach
To pass a Turning Test Approach
- Machine learning
- Automated learning
- Natural language representation
- Computer vision
- Knowledge representation
Machine Learning: An intelligent machine must be able to adapt to a new circumstances and detect extrapolate pattern e.g. learning.
Automated Learning: An intelligent machine must be able to use what it stores to answer any question.
Natural Language Representation: An intelligent machine does not need any artificial language.
Computer Vision: An intelligent machine must be able to process virtual imputes or signal.
Robotic: Ability of a machine to store and manipulate physical object.
Knowledge Representation: An intelligent machine must be able to store what it hears and learns.
Goals of Artificial intelligence
- To create an expert system
- To implement human intelligence to machine
(i) Congnetive Modeling Approach
It deals with thinking human ability to build a machine that acts like human.
(ii) The law of thought approach:
It deals with thinking rationally to provide an argument structure that must be followed to get an actual result at a given premises.
Application of Artificial Intelligence
- Natural language processing
- Speech recognition
- Vision system
- Expert system
- Handwriting recognition
Artificial Intelligence Techniques
- Goal creation and action processing
- Meural Network
- Path finding
- Expert system
- Finite state of a machine
QUESTION: Define state space of a problem
It can be defined as the process that is used in the field of Computer Science, as well as AI in which states of an instant are considered with a goal of finding a goal state with the desired property.
QUESTION: Differentiate between Inductive Reasoning and Deductive Reasoning
Inductive Reasoning: this form of inference produce proposition about an observed or types either specifically or generally based on previous observations. In this form of reasoning, the truth of the premises does not generally guaranty the truth of the conclusion, WHILE
Deductive Reasoning: this form of reasoning, a conclusion follows necessarily from a stated premises, i.e. an inference by reasoning from the general to the specific.
QUESTION: Describe in detail the following types of feedback that determine the main types of learning.
(i) Unsupervised Learning: In this learning method, the target output is not embedded in the network. It is assume that there is no teacher presenting the desired pattern. Hence, the system learns in its own by discovering and adaptive to the structural feature of the input pattern.
(ii) Supervised Learning: Every input pattern that is used to train the network is associated with an output pattern which is the target or designed pattern. A teacher is assumed to be present during t he learning process. When a comparison is made between the networks computed outputs and the correct expected output to determine the error, thereafter, the error can then be used to change the network parameters which can result in performance improvement.
(iii) Reinforced Learning: In this method, though a teacher may be available but does not present in the expected answer but only in the case if the computed output is correct or incorrect. The information provided helps the network in its learning process. A reward is given to a correct answer computed and penalty for the wrong answer.
QUESTION: Describe the known types of reasoning.
(i) Analogical Reasoning: This is reasoning from the particular to the particular. It can be viewed as a form of inductive reasoning, since the truth of the premises cannot guarantee the truth of conclusion.
(ii) Inductive Reasoning: This is a form of inference producing proposition about an observed or types either specifically or generally based on previous observation to formulate general statement. In this type of reasoning, the truth of the premises does not guarantee the truth of the conclusion.
(iii) Abductive Reasoning: In this type of reasoning, the conclusion does not follow with certainly from its premises and concerns something un-observed.
(iv) Deductive Reasoning: This is a form of reasoning in which a conclusion follows necessarily from a state premises. It is generally an inference by reasoning from the general to the specific.
QUESTION: Describe the Turn’s experiment and outline its contribution towards the development of intelligent System
Turning Test Approach
- Test proposed by Alan Turning 1950
- The Computer is asked question by human interrogation
The computer pass the test if a human interrogator after posing some written questions cannot tell whether the written responses come from a person or not, programming a computer to pass the computer need to possess
- Natural Language processing
- Knowledge reproduction
- Automated reasoning
QUESTION: Describe (i) rational agent (ii) Goal base agent (iii) Problem solving agent
An agent can be defined as anything that has the ability to perceives its environment through sensor and act uponds that environment through actuation.
Sensor: Is an agent percept sequence to date
Actuator: It determines the action performed by an agent.
Rational Agent: Ration Agent can be defined as an agent whereby for each possible percept sequence, the rational agent must be able to select an action which is being expected to maximize its performance measure and given the evidence provided by the percept sequence.
Rational agent is an agent that maximizes it performance measures – one that does the right thing
Goal base agent is an agent that combines the current state description with information about the results of possible actions and some goal information that describes which of this actions are desirable
Problem solving agent is an agent that formulates a goal based on the current situation and its performance measures and finds the sequence of actions that will maximize its performance measures.
To have a successful Agent
- Performance Measure: It defines the criteria area for success
- Environment: It is a place whereby the agent work on
- Actuator: It determines the action performed by an agent
- A Sensor: Sensor is an agent percept sequence to date
States to be considered when given a goal
Search: The search algorithm takes the problem as input and returns a salution in a form of action sequence.
Execute:The execute the problem using search tree to generate a search strategies
QUESTION: Describe the salient features of an agent
An agent exists in an environment. It has the following features:
- Some sensor: to detect and collect percept from the environment
- An agent program: to process and interpret the percepts and decide the agent’s action
- Some Actuator: with which the agent manipulates the environments based on what comes out from the agent program
The salient features of an agent
- Percept sequence
- Agent function
- Agent program
- Percept: Percept is an agent perceptual input at any given instant
- Percept Sequence: It completes the history percept of anything that the agent has ever perceive
- Agent Function: It makes from a percept history to an action. f(a) = x3 + 2
- Agent Program: It runs on a physical Architecture to produce f
- Goal: Goal is what the agent needs to achieve at a given time
Characteristics of an Agent
- It must be autonomous
- It must be able to adapt to change
- It must be able to perceive its environment
- It must be able to perceive for a very long time
- It must be able to act for another because not all computer agent are rationally thinker
QUESTION: Describe the basic components of biological neurons
A biological neutron has three (3) types of components that are of particular interest
- Dendrites: Many dendrites receive signals from other neurons. The signal are electric impulses that are Manumitted across a synaptic gap by means of a chemical process
- Soma: This is a site integration of incoming signals. The soma or cell body sums the incoming signals. When sufficient input is received, the cell fires, that is it transmits a signal over its axon to other cells.
- Axon: This is the output pathway forming synapses on target new ions. The transmission of the signal from a particular neuron is accomplished by an action potential resulting from differential concentrations of ions on either side of the neuron’s axon sheath.
QUESTION: Describe four agent architecture
- Simple – reflex Agent
- Model base reflex Agent
- Goal base Agent
- Utility base Agent
- Learning Agent
Simple – Reflex Agent Model base reflex Agent
Goal based agent Utility based agent
Properties of Environment
- Fully observable and partially observable
- Deterministic and stochastic
- Static and Dynamic
- Discrete and continuous
- Episode and sequential
- Fully observable and partially observable
Is fully observable the sensor access all relevant information on its environment while partially the sensor does not.
- Deterministic and Stochastic
In deterministic, the environment totally pendent on the current state of its environment while stochastic is independent
- Static and Dynamic
Static, the agent ask question when deciding on which action to implement while Dynamic the agent does not ask question.
- Human Agent
- Software Agent
- Robotic Agent
- It consists of eyes, nose, ear, tongue and other sensory organ
- It consists of mouth, hands, legs and other part of the body as actuator as sensor
- It consists of keystrokes, file content, received network package as sensor
- It consist of Display screen, file, gent network package as actuator
- It consist of camera and infrared range finder as a sensor
- It consist of various motors as actuator
QUESTION: Describe the components of an expert system
Components of an expert system are
- Inference engine: This is the core of an expert system. It is part of the expert system that manipulates the knowledge based to produce new facts in order to solve a given problem
- Knowledge base
- User interface
- Inference Engine: This is the generic control mechanism that applies the axiomatic knowledge to the task specific data to arrive at some conclusion
- Knowledge base: This contain the domain knowledge which is used by the inference engine to draw conclusions
QUESTION: Differential between uniformed search and heuristic search. State any two search strategies for each one
An uninformed search begins off with definitely no thought of the location of its goal as if one exists at all.
A heuristic search has some thought regarding the whereabout of its goal within the state space and or even if it exists.
QUESTION: Differentiate between knowledge processing and Data processing (Diagram necessary)
In conventional data processing, the decision maker obtains the information created and performs an explicit investigation and analysis of the information before making his/her decision.
In an expert system knowledge is prepared by utilizing accessible data as processing fuel. Conclusions are reached & recommendation are derived implicitly.
The expert system offers the recommendation to the decision maker who makes the final decision & implements it as verifiably.
QUESTION: Differentiate between Supervised and Unsupervised Learning
teacher is available amid learning process and promote expected output.
- Every input design is utilized to prepare the Network
- Learning process depends on correlation and comparison, between Networkcomputed output and connect expected output generating error.
- The error generated is utilized to change network promoters that result in process performance
expected output is absent to the Network
- The system learns of its own by disclosure and versatile to basic andstructural features in the input pattern.
QUESTION: Distinguish between the traditional approach and AI approach of problem solving
|Traditional approach||AI approach|
|This methodology make utilization of both program and data structure||It makes use of data structure only.|
QUESTION: Distinguish between AI, ES and Cybernetic
|1.||It is a PC controlled robot in order to emulate human intelligent||Is an application that is delivered to act in place of human being||This is the study of communication and control involving significant feed back used by living organism in machine and organization|
|2.||It used to build robot||It is utilized to create software||It is utilized to compete favourably by organization.|
QUESTION: Differentiate between agent function and agent program. State why function approach is not advisable agent implementation
Agent Functions maps from a percept history to an action while Agent program runs on a physical architectural to form or product F.
An agent function is a table that maps all possible percepts sequences of an agent to actions.
An agent program. It is an external abstract mathematical description
An agent program is a concrete implementation of the agent function that runs on the agent architecture
Function Approach is not advisable on agent implementation because it deals with the mapping of an history which is being generated from the percept sequence.
Explain the following reasoning strategies:
(i) Forward Chaining Strategy
(ii) Backward Chaining Strategy
(i) Forward Chaining Strategy (or Forward Reasoning) is one of the main methods of reasoning when using an inference engine and can be described logically as repeated application of modus ponens. Forward chaining is a popular implementation strategy for expert systems, business and production rule systems. The opposite of forward chaining is backward chaining. Forward chaining starts with the available data and uses inference rules to extract more data (from an end user, for example) until a goal is reached. An inference engine using forward chaining searches the inference rules until it finds one where the antecedent (If clause) is known to be true. When such a rule is found, the engine can conclude, or infer, the consequent (Then clause), resulting in the addition of new information to its data. Inference engines will iterate through this process until a goal is reached.
(ii) Backward Chaining Strategy (or Backward Reasoning) is an inference method that can be described colloquially as working backward from the goal(s). It is used in automated theorem rovers, inference engines, proof assistants and other artificial intelligence applications. In game theory, its application to (simpler) subgames in order to find a solution to the game is called backward induction. In chess, it is called retrograde analysis, and it is used to generate tablebases for chess endgames for computer chess. Backward chaining is implemented in logic programming by SLD resolution. Both rules are based on the modus ponens inference rule. It is one of the two most commonly used methods as Lisp machines and personal computers. As a result, much effort in the later stages of expert system tool development was focused on integrating with legacy environments such as COBOL and large database systems, and on porting to more standard platforms. These issues were resolved mainly by the client-server paradigm shift, as PCs were gradually accepted in the IT environment as a legitimate platform for serious business system development and as affordable microcomputer servers provided the processing power needed for AI applications
Explain the following:
- Morphological segmentation
- Part-of-speech tagging
- Word segmentation
- Terminology extraction
- Lexical semantics
- Machine translation
- Natural language generation
- Morphological segmentation
Separate words into individual morphemes and identify the class of the morphemes. The difficulty of this task depends on the complexity of the morphology (i.e. the structure of words) of the language being considered. English has fairly simple morphology, especially inflectional
the analyst about the performance of the new installation. This feedback often results in enhancements to meet the user’s requirements.
The environment is the “supersystem” within which an organization operates. It is the source of external elements that impinge on the system. In fact, it often determines how a system must function. For example, the organization’s environment, consisting of vendors, competitors, and others, may provide constraints and, consequently, influence the actual performance of the business.
(vi) Boundaries and Interface
A system should be defined by its boundaries – the limits that identify its components, process and interrelationship when it interfaces with another system. For example, a teller system in a commercial bank is restricted to the deposits, withdrawals and related activities of customers checking and saving accounts. It may exclude mortgage foreclosures, trust activities, and the like. Each system has boundaries that determine its sphere of influence and control. For example, in an integrated banking – wide computer system design, a customer who has a mortgage and a checking account with the same bank may write a check through the “teller system” to pay the premium that is later processed by the “mortgage loan system.” Recently, system design has been successful in allowing the automatic transfer of funds form a bank account to pay bills and other obligations to creditors, regardless of distance or location. This means that in systems analysis, knowledge of the boundaries of a given system is crucial in determining the nature of its interface with other systems for successful design.
QUESTION: Explain the following reasoning strategies.
(i) Forward chaining: This involves applying inference rules to the known fact to generate new facts.
When all inferences have been completed query answering cum proceed quickly.
Greater initialization cost.
High space of memory usage, especially when the number of inferred facts are large
(ii) Backward chaining: This involves starting with a fact to be proved or a query to be answered.
Known inference cost at start up.
There is minimal space requirement
For complex search graphs, this can be computationally expensive and slow.
Inference must be done each and every time a query is answered
QUESTION: Enumerate any four (4) categories of an intelligent known to you.
(i) Rote Agent: i.e. Model based reflex agent. These are agents that capture and communicate pieces of information. It can handle a partially observable environment. Its current state is stored inside the agent maintaining some kind of structure which cannot be seen.
(ii) Decision Agent: These are agents that can be task and come to conclusion with the line of limited information and processing
(iii) Search Agent: i.e. Goal based agent. These are agents that examine multiple pieces of information and return some chosen bit of it. It further expand or the capabilities of the model-based agent by using goal information
(iv) Learning Agent: i.e. Utility based agent. These are agents that examine collection of information and based on that they form concept of generalization to the collected information. It distinguish between the goal states and non-goal states
(v) Coordination agent: i.e. Simple reflex agent. This act only on the basis of the current percept, ignoring therest of the percept history.
QUESTION: Enumerate the properties of acknowledge representation system.
(i) Acquisitional Efficiency: The ability to acquire new knowledge using automatic methods wherever possible rather than reliance human intervention.
(ii) Representational Adequacy: The ability to represent the required knowledge.
(iii) Inferential Adequacy: The ability to manipulate the knowledge represented to produce new knowledge corresponding to that inferred from the original.
(iv) Inferential Efficiency: The ability to direct the inferential mechanisms into the most productive directions by storing appropriate guides.
QUESTION: Enumerate two practical examples each where AI is better than HI and vice versa.
AI is programmed into the computer,that is it is gives consistent answer to the recurrence of processes, task etc
AI can be utilized by several user and can be used in the area where human being cannot be.
HI: has the ability to think and give logical reasoning when solving problems which does not occur normally.
HI: has the ability to solve problem on their own without being programmed while AI need to be programmed
QUESTION: Enumerate any five (5) fundamental principles of knowledge representation
(i) Simple Relational Knowledge: This is the simplest way of storing facts which used relational methods where each facts about a set of object is set out systematically in columns.
(ii) Inheritable knowledge: This exist beyond that of simple relational knowledge b allowing inference mechanisms and this involves property inheritance, element inherit values from the members of the class in such a way t hat data are organize in hierarchical form.
(iii) Inferential knowledge: This is where knowledge is represented as formal knowledge, a logic defines the meaning of semantic of the knowledge.
The semantics defines the truth with respect to each possible word. In standard logic, every sentence of statement must be either me or false so there can be no inbetween.
(i) This approach is popular in A.I. System
(ii) There is a set of strict rules that must be followed
(iii) Procedural knowledge: The basic idea in procedural knowledge is that knowledge are encoded in some procedure, e.g. a parser in a natural language has the knowledge that are noun phrase may contain articles, adjectives and nouns so the parser generates a call routines to call the articles adjectives and nouns.
(i) Extended knowledge inferences can be facilitated.
(ii) The domain specific knowledge can be represented.
(i) Modularity is sacrifice
(ii) Not all the reduction can be corrected, i.e. it cannot be consistent.
QUESTION: Explain five (5) benefits of neural networks in problem-solving
(i) Uniformity of analysis and design
(ii) Very large scale integrated implementability
(iii) Non-linearity: An artificial neuron can be linear or non-linear. A neural network made up of an interconnection of non-linear neurons
(iv) Fault tolerance: A neural network implemented no hardware form has the potential to be inherently fault tolerant or capable of robust computation
(v) Adaptivity Neural networks have a built-in capability to adapt t heir synaptic weights to changes in the surrounding environment.
QUESTION: Explain the following terms:
(i) Context free grammar (ii) Knowledge Engineer (iii) Domain Experts
(iv) Knowledge Respiration Scheme (v) Software Engineer (iv) Sensors
(i) Content Free Grammar: This is aarrangement of rewrite rules, with non terminal symbols changing into a succession of terminal & non terminal symbols.
(ii) Knowledge Engineer: These are people that plan , design, build, assemble & debus the knowledge base on consultation with domain expert
(iii) Domain Expert: These are people who have the suitable prior knowledge about the domain. They think about the area and know about the domain, but commonly they know nothing about the particular case that may be under consideration.
(iv) Knowledge Representation Scheme: This is the form of the knowledge that is used in an agent. A representation of some piece of knowledge is the internal representation of knowledge. A representation scheme specifies the form of knowledge
(v) Software Engineer: Build the inference engineer & user interface.
(vi) Sensor: A sensor provide information about the environment.
(vii) Agent: An agent is something that demostrate in a domain.
QUESTION: Explain any seven (7) branches of Artificial Intelligence that you know
The branches of Artificial intelligence are
(i) Robotics (ii) Perception Systems (iii) Expert System (iv) Fuzzy Logic
(v) Neural Network (vi) Genetic algorithm (vii) Knowledge Engineering
(viii) Natural language processing
- Robotics is the field that endeavor to create machines that can perform work normally done by people, can be classified into manipulations, mobile Robot & humanoid robot.
- Perception Systems are sensing devices that imitates human capabilities of sight, hearing, touch & smell.
- Expert System is an interactive computer based decision tool that uses both facts & heuristics to solve difficult decision problem based on knowledge acquired by the Expert.
- Fuzzy Logic: This is a technique for managingimprecise data & uncertainty with problems that have many answers rather than one.
- Neural Network: Neural Networks utilize physical electronic devices or software to mimic the neurological structure of human brain.
- Genetic Algorithm: A generic algorithm is a program that uses Darwinian principles of random mutation to improve itself.
- Natural Language Processing: This is the study of courses for PCs to perceive and comprehend human language, whether in spoken or written form.
QUESTION: Explain Computational Agent as used in Artificial Agent as used in Artificial Intelligence
Computational Agent is an agent whose choices about its activity can be clarified regarding computation. That is, decision can be broken down into primitives operation that can be implemented in a physical device.
QUESTION: Explain five (5) benefits of neural networks in problem solving
- Non-Linearity: An artificial neuron can be linear or non-linear. A neural network made up of an interconnection of non-linear neurons.
- Input/output mapping: A popular paradigon called supervised learning involves the modification of the synaptic weights of a neural network by applying a set of fairing samples.
- Adaptivity: Neural networks have a built-in capability to adapt their synaptic weights to changes in the surrounding environment
- Contextual Information: Knowledge is represented by the vary structure and activation state of a neural Every neuron in the network is potentially affected network by the global activities of all other neutron in the network
- Fault tolerance: A neural network implemented in hardware form, has the potential to be inherently fault tolerant or capable of robust computation
- Very large scale integrated implementability
- Uniformity of analysis and Design
QUESTION: Explain the difference between forward chaining and backward chaining
Forward Chaining looks at the IF part of a rule first. When the majority of the conductors are met then the appropriate rule is chosen. While Backward chainingbegins with the end i.e. conclusion, at that point identifier the IF conditions necessary to make that conclusion true.
QUESTION: Explain what is meant by Expert System
An Expert is a PC program that endeavors to mirror, imitate and mimic human-experts by the system’s ability to render advice and execute intelligent task.
QUESTION: Explain the three (3) major categories or artificial intelligence
QUESTION: Explain the term Neural Network
Neural Networks: Neural networks utilize physical electronic devices or software to imitate the neurological structure of the human cerebrum i.e. brain.
QUESTION: Give two reasons why natural language is important in AI.
i. The end users of Artificial Intelligence in human beings hence the natural language is critical.
ii. Before data is converted into a machine readable form, it is written in natural language and that is what is being coded.
QUESTION: Given an n-queen problem, define any two ways it may be formulated as a state space search problem
An n-queen problem may be formulated as
- Inevmental formulation – which starts with an empty space and augments the state desenplain with each action.
For an n-queen this means starting off with an empty board and adding a queen to the board with each action.
- Complete State Formulation – Starts with all 8-queen on board and mines them around until the anial is needed
QUESTION: Give and explain the four processes an problem solving agent adopts in its task
Goal Formulation – This helps to organize an agents behavior by limiting its objective. Goals are products of the agents performance measures and its current state.
Problem Formulation – Given a goal, problem formulation decide what states to consider and the actions that transits the agent between this states. This is because not all possible states and actions are reasonable.
Search – The process of looking for a sequence of actions that gives a path from initial scale to goal scale is a search. A search is done by an algorithm based on a sebalgy
Execution – A search action a path or a sequence of action to the goal. The carrying out of the action is excention
QUESTION: Give a brief description of any five (5) application areas of AI
(i) Robotics: Use of Vacuum cleaners as well as development of devices to handle hazardous materials and clear exclusively
(ii) Game Playing: It involves the development of human hyper computer matches using human computing interaction technologies
(iii) Machine Translation: This is the use of statistics and machine learning algorithm to translate from one natural language to another.
(iv) Speech Recognition: Travelers booking a flight can have the entire conversation guided by an automated system.
(v) Logistic Planning: The development of dynamic analysis or re-planning tools to do automated logistic planning and scheduling for transportation
(vi) Robotics Vehicles: Some vehicles are developed with the following properties such as Radar, Lazer range, finder to sense environment, outfitted cameras, driverless
QUESTION: Highlight any eight (8) branches of AI
Branched of A.I.
- Speech recognition
- Bayesian network
- Engineering or representation
- Automatic programming
- Machine learning
- Virtual pattern recognition
- Natural lang. processing
QUESTION: Highlight any eight (8) branches of AI.
(i) Engineering or representation
(ii) Virtual pattern recognition
(iii) Speech recognition
(iv) Bayesian network
(v) Machine learning
(vi) Natural Language Processing
(vii) Automatic programming
QUESTION: List and explain six (6) elements of propositional logic?
Simple sentences which are true or false are basic proportions. Larger and more complex sentences are constructed from basic propositions by combining them with connections. This propositional and connection are basic elements are propositional logic.
NOT, AND, OR IF _ THEN (or IMPLY), IF _ AND _ ONLY _ IF
They are also denoted by the symbol
¬, Λ, V, →, ↔, respectively
QUESTION: List eight (8) application areas of Hidden Marker Model
(i) Speech synthesis
(ii) Gene prediction
(iii) Protein folding
(iv) Time series analysis
(v) Partial discharge
(vi) Handwriting recognition
(vii) Computational finance
(viii) Part of speech tagging
QUESTION: List three (3) examples each of informed and uninformed search algorithms
Informed Search Algorithms
(ii) Greedy Best first search
(iii) A* Search
Uninformed Search Algorithms
(i) Depth-first search
(ii) Uniform cost search
(iii) Breadth-first search
List the factors that constitute the environment of a business system.
- Economic Indication
- Government policies
List four (4) examples of real world application areas of neural networks
- Stock market prediction
- Character recognition
- Image compression
- Loan application
QUESTION: List any four (4) examples of programming language used in developing expert system applications.
(ii) Small talk
(iv) C ++
QUESTION: List six (6) examples of AI programming language known to you.
(iv) Small talk
QUESTION: List and briefly explain the basic components of biological neurons
= } Soma: this is the site of integration of incoming signals. It is called cell body, it sums the incoming signals. When sufficient input is received the cell fires, that is, it transmits a signal over its axon to other cells.
= } Axon: This is the output pathway forming synopses in target neurons. The transmission of the signal from a particular neuron is accomplished by an action potential resulting from differential concentrations of ions in either side of the neuron’s axon sheath.
QUESTION: List four (4) examples of real world application of artificial neural networks.
(i) Prediction/Forecasting: It is used for stock market prediction and weather forecasting.
(ii) Medicine: This is known as instant physician, it is just to store a large number of medical records such as symptoms
(iii) Signal processing: The commercial application is to suppress noise in a telephone line.
(iv) Speech production: Learning to read English test aloud is a difficult task, because the correct phonetic pronunciation of a letter depends on the context in which the letter appears.
QUESTION: List any three (3) examples of the following
(i) Informed search algorithm (ii) Uninformed search algorithm
- Informed search algorithms: To solve large problems with large number of possible states, problem-specific knowledge needs to be added to increase the efficiency of search algorithms.
- Greedy Best first search
- A* search
- Memory—bounded heuristic search
- Uninformed search algorithms: This means that they have no information about the number of steps or the path cost from the current state to the goal. All they can do is distinguish a goal state a non-goal state.
- Breadth-first search
- Depth-first search
- Uniform cost search
QUESTION: List any four (4) features of Expert Systems
- Adaptive Learning ability/reliable
- Explanation facility/high performance
- Easy modification/high responsive
QUESTION: List the Limitations of Expert System
- Lack of flexibility and robustness
- Inability to give deep explanation
- Learning from experience
- Difficulties in verification
- High development cost
QUESTION: Mention any method used by machines in carrying out the following:
(i) Acting humanly: Speech recognition approach
(ii) Thinking humanly: Speech recognition approach
(iii) Thinking rationally: Logic and syllogism
QUESTION: Mention any three (3) tools for building intelligent systems by AI researchers
(i) Automated Reasoning System
(ii) Dynamic Bayesian Network
(iii) Hidden Markoy Model
QUESTION: Mention five (5) limitations of expert system.
High development cost
Most export do not have an independence means to results
Some of knowledge is not always readily available
Learning from experience
Lack of flexibility and robustness
Users have cognitive limit
Inability to give deep explanation
There are frequently multiple correct assessments
Difficulties in verification
Knowledge transform is subject to brases
It can be difficult to extract expertise from
Vocabulary is offer limited and difficult to understand
QUESTION: Mention any six (6) reasons why you study Artificial Intelligence.
(i) To take healthcare and life sciences
(ii) To work with neat and appropriate technology
(iii) To gain insight in human intelligence by considering computational model of intelligence
(iv) To gain the ability to create programs that performs function normally thought to require intelligence
(v) To improve our own problem-solving by taking to heart lessons learned in AI
(vi) To find solutions to specific modern problems such as education, providing online services in medicine, etc.
QUESTION: Show with the aid of a diagram, the structural representation of the application path for Artificial Intelligence and Expert System
QUESTION: State five (5) Application areas of Expert System
- Decision management
- Design / configuration
- Selection / classification
- Process monitoring / control
QUESTION: State the users of expert systems.
- It is utilized in medical diagnosis (the knowledge base would contain medical information, the symptoms of the patient would be used as the query)
- Used in playing strategy games like chess
- Providing financial advice
- Helping to distinguish items such as plants, animals, rock etc.
- Help to discover locations to drill for water, oil
- Help to analyze and diagnose car engine issues
QUESTION: State five distinct roles played by Kr in the development of ES
- KR – enables expert to represent knowledge that is required in a given domain
- It enables expert to acquire more organization
- It enables expert to control and manipulate representation structures
QUESTION: State any six (6) computational intelligence methods used in problem solving
(i) Evolutionary computation
(ii) Genetic Algorithm
(iii) Neutral Network
(iv) Fuzzy System
(v) Hybrid Intelligent System
(vi) Intelligent and Application
QUESTION: State any six (6) commercial application areas of AI.
(i) Speech recognitions
(ii) Game playing
(iii) Machine translation
(iv) Logistic planning
(v) Robotic vehicles
(vi) Automated reasoning and theorem proving
QUESTION: State any six (6) examples of AI programming lang.
- Lisp (List processor)
- PROLOG (programming in Logic)
- Objective C
QUESTION: State the architecture you would recommend for the following agents and justify your answer(i) Internet shopping agent (ii) Intelligent taxi agent (iii) A time tabling agent
(i) Internet shopping agent –
Performance measure: Quality, efficiency, cost
Environment: Vendors, shoppers, current & future website
Actuator: Display for user to see, fill the form
Sensor: Web page
Utility based agent recommended. Shopping involved a goal which is to acquire a product or service at reasonable cost among several competing alternatives and within some limited budget.
A utility based agent will incorporate the goal of purchase, and will also be able to assess how satisfied it will be with a particular product.
(ii) Intelligent Taxi Agent
Performance Measure: Ability to maximize profit, ability to maximize fuel consumption, ability to minimize tears & wears, ability to minimize trip lost & time.
Environment: Ability to work on road e.g. rural, urban
- Pedestrian crossing, zebra crossing and construction
- Traffic light, portholes, cars e.t.c.
Learning agent architecture recommended. A learning agent architecture allows an agent to operate in an initially unknown environment and to become more competent than its initial knowledge alone might allow. This implies that learning agent can learn then environment then environment, can adapt to changing environments and have propensity of getting better over time, and also evaluate their performance. All this are guided to an automated taxi that will operate in new, dynamic environment within performance evidence of cost, speed, safely e.t.c.
(iii) A time tabling agent
Simple reflex agent recommended. A time labeling agent generally operate in an environment that is fixed, decormination and fully observable.
QUESTION: State the types of expert systems known to you
- Inference engine
- Knowledge base
- User interface
QUESTION: State the uses of expert systems
(i) Help to diagnose car engine problems
(ii) Providing financial advice
(iii) Help to discover locations to drill for water, oil
(iv) Used in playing strategies games like chess
(v) Helping to identify items such as plants, animals, etc.
QUESTION: State any five (5) representing guidelines known to you.
- Simple relational knowledge
- Inheritable knowledge
- Inferential knowledge
- Procedural knowledge
- Heuristic knowledge
QUESTION: State any five (5) application areas of Hidden Markov model
- Computational finance
- Single molecule kinetic analysis
- Speech recognition
- Speech synthesis
- Part of speech tagging
- Machine translation
- Partial discharge
- Handwriting recognition
- Gene prediction
10) Time series analysis
- Activity recognition
- Protein folding
QUESTION: What do you understand by uniformed search?
It simply means that they have no information about the number of steps or the path cost from the current state to the goal. All they can do is to distinguish a goal state from a non-goal state, e.g. BFS, UCS, DFS.
QUESTION: Write the algorithm for Depth First Search (DFS)
(i) Start by putting any one of the graph’s vertices on top of a stack
(ii) Take the top item of the stack and add it to the visited list
(iii) Create a list of that vertex’s adjacent nodes. Add the one’s which are not in the visited list to the top of stack list to the top of stack
(iv) Keep repeating step 2 and 3 until the stack is empty
QUESTION: What is Artificial Neural Network?
It is referred to as a massively parallel distributed processor made up of simple processing units which has a natural propensity for storing experiential knowledge and making it available for use.
QUESTION: What is a state space problem? Explain the characteristics that define a state space problem
State refers to the vertices of an underlining graph which/is being searched to generate an actual node. The set of all reachable states from the initial state defines a state space problem
State Space Problem: State space problem is being represented by the no attacked.
QUESTION: What is an Intelligent Agent?
It can be referred to as anything that can be viewed through sensor as perceiving its environment and affect that environment either directly or through the co-operation of other agents. They sense and act upon that environment through actuators.
It can also be referred as any element of society that can perceive aspect of its environment and affect that environment either directly or through co-operation with other agents. It is also referred to as anything that can be viewed as perceiving with environment through a sensors and acting upon that environment through actuator. This is illustrated as follows:
QUESTION: What is an Expert System?
This is referred to as the program t hat attempt to perform the duty of expert in problem domain in which its defined. It is also referred to as knowledge-based computer program that uses AI to provide solution to complete program that ordinarily would have require the human expert or expertise, i.e. they are computer programs that has been constructed with assistance of human expert in such a way that they are capable of fine-tuning at the standard. It is sometimes at a higher standard than human expert in a given field. This also involves domain specific knowledge and a combination of problem solving rules. This is a knowledge based computer program that uses AI to provide solution to complex program that ordinarily would have require the human expert or expertise. This is also includes domain specific knowledge and a combination of theoretical understanding of the problem and a collection of problem solving rules.
QUESTION: What is AI?
This can be referred to as the intelligence exhibited by an artificial entity with intelligence exhibited by an artificial entity with intelligence behavior, learning and adaptations in machines, such a system is generally assumed to be a computer. It covers a variety of ranging from the general to the specific such as playing chess, proving mathematical theorem, writing poetry, driving a car on a crowded street, etc. It is relevant to any intelligent task.
What is Artificial Neural Networks?
(a) Artificial Neural Networks (ANNs) or connectionist systems are computing systems inspired by the biological neural networks that constitute animal brains. Such systems learn (progressively improve performance on) tasks by considering examples, generally without task-specific programming. For example, in image recognition, they might learn to identify images that contain morphology, and thus it is often possible to ignore this task entirely and simply all possible forms of a
Giving a sentence, determine the part of speech for each word. Many words, especially common ones, can serve as multiple parts of speech.
Determine the parse tree (grammatical analysis) of a given sentence. The grammar for natural languages in ambiguous and typical sentences have multiple possible analyses. In fact, perhaps surprisingly, for a typical sentence there may be thousand of potential parses (most of which will seem completely nonsensical to a human).
Separate a chunk of continuous text into separate words. For a language like English, this is fairly trivial, since words are usually separated by spaces. However, some written languages
The goal of terminology extraction is to automatically extract relevant terms from a given corpus.
What is the computational meaning of individual words in context?
Automatically translate text from one human language to another. This is one of the most difficult problems, and is a member of a class of problems colloquially termed “Al-complete”, i.e. requiring all of the different types of knowledge that humans possess (grammar, semantics, facts about the real world, etc.) in order to solve properly.
Natural language generation
Convert information from computer databases or semantic intents into readable human language.
cats by analyzing example images that have been manually labeled as “cat” or “no cat” and using the results to identify cats in other images. They do this without any a prior knowledge about cats, e.g., they have fur, tails, whiskers and cat-like faces. Instead, they evolve their own set of relevant characteristics from the learning material that they process.
QUESTION: What is “reasoning” as used in AI?
Reasoning is the capacity of consciously making sense of thing, applying logic for establishing and verifying facts and changing practicing institution and belief base of new or existing information. It is closely associated with such characteristics human activities as philosophy, science, language, mathematics and art. The concept of reasoning is associated with thinking, cognition and intellect.
Types of Reasoning:
- Inductive Reasoning
- Deductive Reasoning
- Abductive Reasoning
- Analogical Reasoning
- Inductive Reasoning: This is a form of inference producing preposition about an observed or types either specifically or generally based on previous observation to formulate general statement. In this type of reasoning, the truth of the premises does not guarantee the truth of the conclusion.
Premise: The sun rises from the east each morning.
Conclusion: The sun will rise from the east tomorrow
- Deductive Reasoning: This is a form of reasoning in which a conclusion follows necessarily from a stated premises. It is generally an inference by reasoning from the general to the specific.
Premise 1: All Nigerian are billionaire
Premise 2: Obasanjo is a Nigerian
Conclusion: Obansanjo is a billionaire
- Analogical Reasoning: This is reasoning from the particular to the particular. It can be viewed as a form of inductive reasoning, since the truth of the premises cannot guarantee the truth of conclusion.
- Abductive Reasoning: In this type of reasoning, the conclusion does not follow with certainly from its premises and concerns something un-observed. It is an attempt to favour one conclusion above others.
QUESTION: What is Representation and Reasoning System?
This composed of a language to communicate with a computer in such a way of assigning meaning to the language and procedure to computer answers given input in the language.
QUESTION: What is a Decision Table?
Decision Table: This is compact and precise methods for displayingcomplicated logic C., such as that which may be use in a PC program.
ii) Reasons for using a Decision Table
- It give a general method of stating complex business rules
- It aid advancement process with developers to do a better job
- It is utilized to display and model complicated logic
- It is a remarkable methods utilized in both testing and requirements management
QUESTION: What are the distinguishing characteristics that make PROLOG and LISP programming languages useful for Expert System?
Characteristics of Expert System Programming Language
- Efficient blend of integer number and real variable
- Good memory management methods
- Extensive data manipulation schedules
- Incremental compilation
- Tagged memory architecture
- Efficient search procedures
- Optimization of the systems environment
QUESTION: What is knowledge representation according to the five (5) distinct roles that is plays?
Knowledge representation is most fundamentally a substitute for the thing useful, used to unable an entity to determine by thinking rather than acting.
- It is a set of ontological limitations
- It is a dividing theory of intelligent name and expression
- It is medium for pragmatically effective calculation
- It is a medium of human expression