artificial_intellegence.pdf

Upload: haque-nawaz-lashari

Post on 02-Mar-2018

219 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/26/2019 Artificial_Intellegence.pdf

    1/49

    Biyani's Think Tank

    Concept based notes

    Artificial IntelligenceMSc-IT

    Ms RashmiDeptt. of IT

    Biyani Girls College, Jaipur

  • 7/26/2019 Artificial_Intellegence.pdf

    2/49

    2

    Published by :

    Think TanksBiyani Group of Colleges

    Concept & Copyright :

    Biyani Shikshan SamitiSector-3, Vidhyadhar Nagar,

    Jaipur-302 023 (Rajasthan)

    Ph : 0141-2338371, 2338591-95 Fax : 0141-2338007E-mail : [email protected]

    Website :www.gurukpo.com; www.biyanicolleges.org

    Edition : 2012

    Leaser Type Setted by :Biyani College Printing Department

    While every effort is taken to avoid errors or omissions in this Publication, any mistake or

    omission that may have crept in is not intentional. It may be taken note of that neither the

    publisher nor the author will be responsible for any damage or loss of any kind arising to

    anyone in any manner on account of such errors and omissions.

  • 7/26/2019 Artificial_Intellegence.pdf

    3/49

    AI 3

    Preface

    I am glad to present this book, especially designed to serve the needs of thestudents. The book has been written keeping in mind the general weakness inunderstanding the fundamental concepts of the topics. The book is self-explanatory andadopts the Teach Yourself style. It is based on question-answer pattern. The languageof book is quite easy and understandable based on scientific approach.

    Any further improvement in the contents of the book by making corrections,omission and inclusion is keen to be achieved based on suggestions from the readers

    for which the author shall be obliged.I acknowledge special thanks to Mr. Rajeev Biyani, Chairman& Dr. Sanjay Biyani,

    Director (Acad.) Biyani Group of Colleges, who are the backbones and main conceptprovider and also have been constant source of motivation throughout this endeavour.They played an active role in coordinating the various stages of this endeavour andspearheaded the publishing work.

    I look forward to receiving valuable suggestions from professors of variouseducational institutions, other faculty members and students for improvement of thequality of the book. The reader may feel free to send in their comments and suggestionsto the under mentioned address.

    Author

  • 7/26/2019 Artificial_Intellegence.pdf

    4/49

    4

    Artificial Intelligence and Expert System

    Prerequisite: System Software, Operating System, Data and File Structure.Introduction of Artificial Intellignce: Simulation of so called intelligent behavior, indifferent areas. Problem solving: Games, natural language, question answering, visualperception, learning, Aim-oriented (heuristic) algorithm versus solution guaranteedalgorithms.

    Understanding Natural Languages: Parsing techniques. Context free andtransformational grammars, transition nets, augmented transition nets, Fillmore'sgrammars. Shank's conceptual dependency, grammar-free analyzers, sentencegeneration, translation.

    Knowledge Representation: First-Order predicate calculus Horn's clauses, TheLanguage PROLOG, semantic nets, Partitioned rules, knowledge base, the inferencesystem, forward and backward deduction.

    Expert Systems: Existing system (DENDRAL MYCIN): Domain exploration, meta-Knowledge, expertise transfer, self-explanining systems machine perception, line

    finding, interpretation semantics and models, object identification, speech recognition.Books

  • 7/26/2019 Artificial_Intellegence.pdf

    5/49

    AI 5

    Unit1

    Introduction

    Q1. What is AI?

    Ans Artificial intelligence is the study of how to make computers do things which, atthe moment, people can do better.Artificial Intelligence is the study of human intelligence such that it can bereplicated artificially.

    Q.2 . What are the different applications of AI?Ans

    a) Game playingb) Speech recognitionc) Understanding natural languaged) Computer visione) Expert system

    f) Fuzzy logic system

    Q.3. What are the different problems involved in AI?

    Ansa) Deduction,

    b) Reasoning,

    c) Problem solving

    d) Knowledge representation

    e) Planning

    f) Learning Natural Language Processing

    g) Motion and manipulation

    h) PerceptionCreativity

    Q.4. Explain Depth-First SearchAns

    1. Set Lto be a list of the initial nodes in the problem.

  • 7/26/2019 Artificial_Intellegence.pdf

    6/49

    6

    2. If Lis empty,failotherwise pick the first node nfrom L

    3.

    If nis a goal state, quit and return path from initial node.

    4. Otherwise remove nfrom Land add to the front of Lall of n's children. Label

    each child with its path from initial node. Return to 2.

    Q.5. Explain Breadth-First Search .Ans

    1. Set L to be a list of the initial nodes in the problem.

    2. If L is empty, fail otherwise pick the first node n from L

    3.

    If n is a goal state, quit and return path from initial node.

    4. Otherwise remove n from L and add to the end of L all of n's children. Label

    each child with its path from initial node. Return to 2.

    Q6 . What is meant by heuristic search technique?What are different heuristicsearch techniques.

    Ans It is a search technique that relies on the estimate provided by the heuristic

    function. Heuristic search techniques make use of domain specific information

    1.

    Generate and Test Search :-In the Generate and Test Search

    A. Generate a possible solution with generating a path from Start state

    B. Then test to see if this is actually a solution by comparing the chosen point orthe endpoint of the chosen path to the set of acceptable goal states.

    C. If the solution has found then quit. Otherwise, return to step A

    The Generate and Test algorithm is a depth-first-serach procedure sincecomplete solution must be generate before they can be tested . It also operate by

    generating solution randomly, but there is no guarantee that a solution will everbe found

    2. Hill Climbing Search :-

  • 7/26/2019 Artificial_Intellegence.pdf

    7/49

  • 7/26/2019 Artificial_Intellegence.pdf

    8/49

    8

    As we already know that heuristics always has information about the problemand its parameter.In the Best-First Search it follow a single path at a time but it has capability toswitch path whenever some competing path looks more promising than thecurrent one does.

    The Best-First-Search is an instance of the general Tree-Search or Graph-Searchalgorithm in which a Node is selected for expansion based on an evaluationfunction f(n). The evaluation function measures distance to the Goal and theNode with the lowest evaluation is selected for expansion first.

    The Key component in Best-First algorithm is a heuristic function h(n) , which is

    the estimated cost of his cheapest path from n to a goal node.

    The heuristic function h(n) are the most common form in which additionknowledge of the problem is imparted to the search algorithm

    At the each step of the Best-First Search process, we select the most promising ofthe nodes we have generated so far. This is done by applying an appropriateheuristic function to each of them. We then expand the chosen node by using therules to generate its successor. If one of them is a solution, we can quit. If not, allthose new nodes are added to the set of new nodes generated so far. Again the

    most promising node is selected and the process continue, if the solution notfound, then that branch will start to look less promising then one of the top-levelbranches that has been ignored. At this point, the now more promising,previously ignored branch will be explored. But the old branch not forgotten.

    To implement it we will need to use two list of nodes

    OPEN nodes that have been generated and have had the heuristic functionapplied to them but which have yet to be examined. OPEN is actually a priorityqueue in which the elements with the highest priority are those with the most

    promising value of the heuristic function.

    CLOSE nodes that have already been examined. We need to keep these nodesin memory to check whenever a node is generated it already examined or not.

  • 7/26/2019 Artificial_Intellegence.pdf

    9/49

    AI 9

    OPEN = initial statewhile OPEN != nulldo1. Pick the best node on open.2. Create open's successors3. For each successor do:

    a. If it has not been generated before: evaluate it,add it to OPEN, and record its parent

    b. Otherwise: change the parent if this new path isbetter than previous one.

    done

    Lets consider the following example to find the path from Start to Goal in thegraph below

  • 7/26/2019 Artificial_Intellegence.pdf

    10/49

    10

    >>> First add the Start node to the fringe

    >>> Visit the Start Node and add its neighbors to the fringe

  • 7/26/2019 Artificial_Intellegence.pdf

    11/49

    AI 11

    >>> Visit the Node A and add its neighbors to the fringe

    Now there is a choice on which node to visit next. Because we are using greedybest-first search, the node with the lowest heuristic is used. If this solution was

  • 7/26/2019 Artificial_Intellegence.pdf

    12/49

    12

    implemented, a priority queue would be used for the fringe, so it would alwaysreturn the node with the lowest heuristic. Since node Dhas the lowest heuristicvalue, we visit at that node and add its neighbors to the fringe.

    Now, since node Ehas the lowest heuristic in the fringe, it is visited at and itsneighbors are added to the fringe.

  • 7/26/2019 Artificial_Intellegence.pdf

    13/49

    AI 13

    Finally, since the Goal is in the priority queue with a heuristic of 0, it is visitedand a path to the goal is found. The path found from Start to Goal is: Start -> A ->D -> E -> Goal. In this case, it was the optimal path, but only because the

    heuristic values were fairly accurate.

    Q7. Explain Best First Search.Ans The best first search allows us to switch between paths thus gaining the benefit

    of both approaches. At each step the most promising node is chosen. If one of thenodes chosen generates nodes that are less promising it is possible to chooseanother at the same level and in effect the search changes from depth to breadth.If on analysis these are no better then this previously unexpanded node andbranch is not forgotten and the search method reverts to the descendants of thefirst choice and proceeds, backtracking as it were.

    Best First Search Algorithm:

    1. Start with OPEN holding the initial state2. Pick the best node on OPEN

  • 7/26/2019 Artificial_Intellegence.pdf

    14/49

    14

    3. Generate its successors4. For each successor Do

    o If it has not been generated before evaluate it add it to OPEN and recordits parent

    o If it has been generated before change the parent if this new path is betterand in that case update the cost of getting to any successor nodes

    5. If a goal is found or no more nodes left in OPEN, quit, else return to 2.

    Q.8. What are the achievements of AI?Ans. The achievements of AI are as follows: -

    1. Deep thought is an international grand master chess player.2. Sphinx can recognize continuous speech without training for each speaker. It

    operates in nearreal time using a vocabulary of 1000 words and has 94% wordaccuracy.

    3. Navlab is a truck that can drive along a road at 55 KMPH in normal traffic.4. Carlton and United Breweries use an AI planning system to plan production

    of their beer.5. Robots are used regularly in manufacturing.6. Natural language interface to databases can be obtained on a PC.

    7.

    Machine Learning methods have been used to build expert systems.8. Expert systems are used regularly in finance, medicine, manufacturing, and

    agriculture

    Q9 . Comment on Best-First is a combination of depth first and breadth firstsearches.

    Ans. Depth first is good because a solution can be found without computing all nods

    and breadth firstis good because it does not get trapped in dead ends. The bestfirst search allows us to switchbetween paths thus gaining the benefit of bothapproaches. At each step the most promising node ischosen. If one of the nodeschosen generates nodes that are less promising it is possible to chooseanother atthe same level and in effect the search changes from depth to breadth. If onanalysis theseare no better than this previously unexpanded node and branch isnot forgotten and the searchmethod reverts to the descendants of the first choiceand proceeds, backtracking as it were.

  • 7/26/2019 Artificial_Intellegence.pdf

    15/49

    AI 15

    Q.10 . Write goals of A.I?Ans Traditionally there are four possible Goal of A.I. and have been followed and

    the approaches are

    1. Cognitive Science Approach :- Cognitive means Process of Understanding andIn this approach we aspect that system should think like a human beings. It justnot focus on behavior and I/O, but to produce a sequence of steps of thereasoning process, similar to the steps followed by a human in solving the sametask.

    2. Low of thought Approach :- It focus on logical thought rather than emotions tomake decisions. It inference mechanisms that are probably correct and guaranteean optimal solution. As it tells about the system of logical rules and procedure

    for final decision.3. Turning Test Approach :- The art of creating machines that perform

    tesk/function utilizing the same intelligence when they perform by the peoplehence it is the study of, how to make computer to things in the same manner ashuman beings do better. It focus on action not on intel ligent behavior. Its dontbother on how to get result be focus on result should be similar to what humanresult are.

    4. Rational agent Approach :- An Agent is one who act upon and a rational agent isone who that act so as to achieve best outcome or if there is uncertainty in theresult then achieve at least best expected outcome. Moreover it concern with that

    system should at least produce sufficient outcome if could not produce optimumoutput in all case.

    Q.11. Explain problem solving.

    Ans Problem Solving:- Problem solving is the process of generating solution from

    observed or given data (Inputs). It is however not always possible to use directmethods. Instead, problem solving often use indirect or model based methods. Inthe A.I. most of real word problems can be solved by searching for a solution. Tobuild a system to solve a particular problem, we need to

    Define the problem precisely find input situation as well as final situation for

    acceptable solution to the problem.Analyze the problem find few important feature that may have impact on the

    appropriateness of various possible technique for solving the problem.

    Isolate and represent task knowledge necessary to solve the problem.

  • 7/26/2019 Artificial_Intellegence.pdf

    16/49

    16

    Choose the best problem solving technique(s) and apply to the particularproblem

    Q12. What is Natural Language Processing?

    Ans Natural Language Processing (NLP) is an area of research and application thatexplores how computers can be used to understand and manipulate naturallanguage text or speech to do useful things. NLP researchers aim to gatherknowledge on how human beings understand and use language so thatappropriate tools and techniques can be developed to make computer systemsunderstand and manipulate natural languages to perform the desired tasks. Thefoundations of NLP lie in a number of disciplines, viz. computer and informationsciences, linguistics, mathematics, electrical and electronic engineering, artificial

    intelligence and robotics, psychology, etc. Applications of NLP include a numberof fields of studies, such as machine translation, natural language text processingand summarization, user interfaces, multilingual and cross language informationretrieval (CLIR), speech recognition, artificial intelligence and expert systems,and so on.

    Q13. What is the Turing test?Ans Alan Turing's 1950 article Computing Machinery and Intelligence discussed

    conditions for considering a machine to be intelligent. He argued that if themachine could successfully pretend to be human to a knowledgeable observer

    then you certainly should consider it intelligent. This test would satisfy mostpeople but not all philosophers. The observer could interact with the machineand a human by teletype (to avoid requiring that the machine imitate theappearance or voice of the person), and the human would try to persuade theobserver that it was human and the machine would try to fool the observer.

    The Turing test is a one-sided test. A machine that passes the test shouldcertainly be considered intelligent, but a machine could still be consideredintelligent without knowing enough about humans to imitate a human.

    Q.14. Isn't AI about simulating human intelligence?

    Ans Sometimes but not always or even usually. On the one hand, we can learnsomething about how to make machines solve problems by observing otherpeople or just by observing our own methods. On the other hand, most work inAI involves studying the problems the world presents to intelligence rather than

  • 7/26/2019 Artificial_Intellegence.pdf

    17/49

    AI 17

    studying people or animals. AI researchers are free to use methods that are notobserved in people or that involve much more computing than people can do.

    Q.15. Are computers the right kind of machine to be made intelligent?

    Ans Computers can be programmed to simulate any kind of machine.

    Many researchers invented non-computer machines, hoping that they would beintelligent in different ways than the computer programs could be. However,they usually simulate their invented machines on a computer and come to doubtthat the new machine is worth building. Because many billions of dollars thathave been spent in making computers faster and faster, another kind of machinewould have to be very fast to perform better than a program on a computer

    simulating the machine.

    Q.16. Explain term simulation?

    Ans Simulation

    A simulation is a system that is constructed to work, in some ways, analogouslyto another system of interest. The constructed system is usually made simplerthan the original system so that only the aspects of interest are mirrored.Simulations are commonly used to learn more about the behavior of the originalsystem, when the original system is not available for manipulation. It may not beavailable because of cost or safety reasons, or it may not be built yet and the

    purpose of learning about it is to design it better. If the purpose of learning is totrain novices, then cost, safety, or convenience are likely to be the reasons towork on a simulated system. The simulation may be a computer simulation(perhaps a realistic one of a nuclear power station's control room, or amathematical one such as a spreadsheet for "what-if" analysis of a company'sbusiness); or it may be a small-scale physical model (such as a small-scale bridge,or a pilot chemical plant).

    Q17. Explain Cognitive Science.Ans Artificial intelligence can be defined as the mimicking of human thought to

    perform useful tasks, such as solving complex problems. This creation of newparadigms, algorithms, and techniques requires continued involvement in thehuman mind, the inspiration of AI. To that end, AI software designers team with

  • 7/26/2019 Artificial_Intellegence.pdf

    18/49

    18

    cognitive psychologists and use cognitive science concepts, especially inknowledge elicitation and system design

    Multiple Choice Questions

    1. Weak A.I. is the

    a) Study of mental fecilities through the use of mental methodsimplemented on a computer

    b) Set of computer programc) All of the above

    d)

    None of the aboveAns: a

    2. Cognitive science is a

    a) Combination of AI and Phychologyb) Combination of AI and medicinec) Combination of AI and Sociologyd) None of the aboveAns: a

    3. A.I can be defined as

    a)

    A branch of computer science concerned with creating computersystems exhibiting intelligence

    b) A branch of computer science dealing with graph theoryc) A branch of computer science used for creating a databasesd) None of the aboveAns :a

    4. What is the term used for describing the judgmental or commonsense

    part of problem solving?

    a) Heuristic

    b)

    Criticalc) Value basedd) None of the aboveAns: a

  • 7/26/2019 Artificial_Intellegence.pdf

    19/49

    AI 19

    5. Which kind of planning consists of successive representations ofdifferent levels of a plan?

    a)

    Hierachial planningb) Non-hierachial planningc) All of the above andd) None of the aboveAns: a

    6. What was originally called the "imitation game" by its creator?

    a) LISPb) Turning testc) Cybernetics

    d)

    None of the aboveAns: b

    7. If a robot can alter its own trajectory in response to external conditions

    it is considered to be

    a) Intelligientb) Mobilec) Open loopd) NoneAns: a

    8. An A.I technique that allows computers to understand association andrelationship between objects and events called

    a) Heuristic processingb) Cogiitive sciencec) Pattern matchingd) noneAns: c

    9. The field that investigate the machanics of human intelligence is

    a) Coginitive scienceb)

    Sociologyc) Psychologyd) Sociologye) NoneAns: a

  • 7/26/2019 Artificial_Intellegence.pdf

    20/49

    20

    10. Natural language processing is divided in to subfields of

    a)

    Symbolic and numericb) Time and motionc) Understanding and generationd) NoneAns:c

    11. Father of A.I

    a) Alan Turningb) Fisher Adac) Allen Newell

    d)

    noneAns:a

    12. Area of A.I. that investigate methods of facilitating communicationbetween people and computers is

    a) Natural language processingb) Symbolic processingc) Roboticsd) noneAns: a

    13. Which approach to speech regonition avoids the problem caused by the

    differences in the way words words are prounced accordibng to context

    a) isolated word recognitionb) continous speech regonitionc) speaker-dependentd) noneAns:a

    14. Field of A.I. that covers finger print identification , handwritingrecognition, wheather forecasting is

    a)

    pattern recognitionb) image processingc) fuzzy logicd) parsing programAns:a

  • 7/26/2019 Artificial_Intellegence.pdf

    21/49

    AI 21

    15. Which is true regarding BFS?

    (a) BFS will get trapped exploring a single path(b) The entire tree so far been generated must be stored in BFS(c) BFS is not guaranteed to find a solution, if exists(d) BFS is nothing but Binary First Search(e) BFS is one type of sorting.Ans : (b)

    16. What is a heuristic function?

    (a) A function to solve mathematical problems(b) A function which takes parameters of type string and returns an

    integer value

    (c) A function whose return type is nothing(d) A function which returns an object(e) A function that maps from problem state descriptions to measures of

    desirability.Ans : (e)

    17. The traveling salesman problem involves n cities with paths connecting

    the cities. The time taken for traversing through all the cities, withoutknowing in advance the length of a minimum tour, is

    (a) O(n)

    (b) O(n2)(c) O(n!)(d) O(n/2)(e) O(2n).

    Ans : (c)

  • 7/26/2019 Artificial_Intellegence.pdf

    22/49

    22

    Unit 2

    Knowledge Representation

    Q.1. What is Knowledge?Explain natural level processing.

    Ans Knowledge is a general term. Knowledge is a progression that starts with datawhich is of limited utility.

    knowledge--1. In artificial intelligence, symbolic information used by a domainexpert to solve problems. 2. Facts and relationships used to solve problems.

    Natural Language Processing

    English is an example of a natural language, a computer language. For acomputer to process a natural language, it would have to mimic what a humandoes. That is, the computer would have to recognize the sequence of wordsspoken by a person or another computer, understand the syntax or grammar of

    the words (i.e., do a syntactical analysis), and then extract the meaning of thewords. A limited amount of meaning can be derived from a sequence of wordstaken out of context (i.e., by semantic analysis); but much more of the meaningdepends on the context in which the words are spoken (e.g., who spoke them,under what circumstances, with what tone, and what else was said, particularlybefore the words), which would require a pragmatic analysis to extract. To date,natural language processing is poorly developed and computers are not yet ableto even approach the ability of humans to extract meaning from naturallanguages; yet there are already valuable practical applications of thetechnology.

    Q.2 What is Knowledge Representation and Reasoning (KR or KRR):

    Ans A subarea of Artificial Intelligence concerned with understanding, designing,

    and implementing ways of representing information in computers, and usingthat information to derive new information based on it. KR is more concerned

  • 7/26/2019 Artificial_Intellegence.pdf

    23/49

    AI 23

    with belief than knowledge". Given that an agent (human or computer) hascertain beliefs, what else is reasonable for it to believe, and how is it reasonablefor it to act, regardless of whether those beliefs are true and justified.

    Q.3. What is logic? Explain different types of logic used in knowledgerepresention..

    Ans Logic is the study of correct reasoning.It is a collection of rules we use when

    doing logical reasoning. Human reasoning has been observed over centuriesfrom at least the times of Greeks, and patterns appearing in reasoning have beenextracted, abstracted, and streamlined. The foundation of the logic we are goingto learn here was laid down by a British mathematician George Boole in themiddle of the 19th century, and it was further developed and used in an attempt

    to derive all of mathematics by Gottlob Frege, a German mathematician, towardsthe end of the 19th century. A British philosopher/mathematician, BertrandRussell, found a flaw in basic assumptions in Frege's attempt but he, togetherwith Alfred Whitehead, developed Frege's work further and repaired thedamage. The logic we study today is more or less along this line.In logic we are interested in true or false of statements, and how thetruth/falsehood of a statement can be determined from other statements.However, instead of dealing with individual specific statements, we are going touse symbols to represent arbitrary statements so that the results can be used inmany similar but different cases. The formalization also promotes the clarity of

    thought and eliminates mistakes.There are various types of logic such as logic of sentences (propositional logic),logic of objects (predicate logic), logic involving uncertainties, logic dealing withfuzziness, temporal logic etc. Here we are going to be concerned withpropositional logic and predicate logic, which are fundamental to all types oflogic.Propositional logic is a logic at the sentential level. The smallest unit we dealwith in propositional logic is a sentence. We do not go inside individualsentences and analyze or discuss their meanings. We are going to be interestedonly in true or false of sentences, and major concern is whether or not the truth

    or falsehood of a certain sentence follows from those of a set of sentences, and ifso, how. Thus sentences considered in this logic are not arbitrary sentences butare the ones that are true or false. This kind of sentences are called propositions.

  • 7/26/2019 Artificial_Intellegence.pdf

    24/49

    24

    Q.4. Explain Horn Clause.

    Ans A Horn clause is a clause containing at most one positive literal.

    Examples of a Horn Clause

    [Child, Mail,Boy]

    Q5. Explain the process of Skolemization. How is this accomplished? Givesuitable examples insupport of your answer.

    Ans. Skolemization is the process of removing existential quantifiers by elimination.In the simpletranslate into P (A), where A is a constant that does not appearelsewhere in the KB. But thereis the added complication that some of the

    existential quantifiers, even though move left, may still benested inside auniversal quantifier.Skolemization is accomplished as follows: -

    If the first (leftmost) quantifier in an expression is an existential quantifier,replace alloccurrences of the variable it quantifier with an arbitrary constant notappearingelsewhere and delete the quantifier. The same procedure should befollowed for all otherexistential quantifiers not preceded by a universalquantifier, in each case, using differentconstant symbols in the substitution.

    For each existential quantifier that is preceded by one or more universalquantifiers (iswithin the scope of one or more universal quantifiers) replace all

    occurrences of theexistentially quantified variable by a function symbol notappearing elsewhere in theexpression. The argument assigned to the functionshould match all the variablesappearing in each universal quantifier whichpreceded the existential quantifier. Thisexistential quantifier should then bedeleted. The same procedure should be repeated foreach remaining existentialquantifier using a different function symbol and choosingfunction argumentsthat correspond to all universally quantified variables that precedetheexistentially quantified variable being replaced.Example of Skolemization

    Consider Everyone has a heart:

    Has (x, y)

    If we just replaced y with a constant, H, we would get,

  • 7/26/2019 Artificial_Intellegence.pdf

    25/49

    AI 25

    Has (x, H)Which says that everyone has the same heart H.? We need to say that the heartthey have is notnecessarily shared, that is, it can be found by applying to eachperson a function that maps from personto heart:

    Has (x, F(x))

    Where F is a function name that does not appear elsewhere in the KB. F is called

    a SkolemFunction. In general, the existentially quantified variable is replaced by

    a term that consists of a SkolemFunction applied to all the variables universally

    quantified outside the existential quantifier in question.Skolemization eliminates

    all existentially quantified variables, so we are now free to drop the

    universalquantifiers, because any variable must be universally quantified.

    Q.6. Comment on Heuristics are fallible.Ans. Heuristics are fallible because they rely on limited information, they may lead to

    a suboptimalsolution or to a dead end.Heuristics is a rule of thumb orjudgmental technique that leads to a solution some of the timebut provides noguarantee of success. It may in fact end in failure. Heuristics plays an importantrolein search strategies because of the exponential nature of most problems. Theyhelp to reduce thenumber of alternatives from an exponential number to apolynomial number and, thereby, obtain asolution to a tolerable amount of time.

    When exhaustive search is impractical, it is necessary tocompromise for aconstrained search which eliminates many paths but offers the promise ofsuccesssome of the time. Here, success may be considered to be finding anoptimal solution a fair proportionof the time or just finding good solutions muchof the time.

    Q7. Requirements of KR languages

    Ans 1. At the implementational level, the main concern is efficiency (space and time).

    2. At the logical level, one is concerned with two things: the syntax (Fregescompositionality principle, which states that the meaning of a compoundexpression should be derivable from the meanings of its parts) and thesoundness of the inference rules.

  • 7/26/2019 Artificial_Intellegence.pdf

    26/49

    26

    3. At the epistemological level, there are four main concerns: If there exists anatural way of organising knowledge, the KR language has to respect this; theKR language has to be modular; attention to the granularity (=size of theinformation chunks) is important; and the language should support the actualprimitives of the conceptual level.

    4. At the conceptual level, one is concerned with how concisely one canrepresent particular pieces of knowledge.

    Q8. What are semantic nets?Ans There are two types of primitives: nodes and links. Nodes correspond to objects,

    or classes of objects, in the world, whereas links correspond to relationshipsbetween these objects. They are often labeled to indicate their meaning. No

    information is stored in a node as such; this is done with the links associatingnodes.Quillians original work was taken up by psychologists, but also work done inlinguistics was important for the development of the semantic net approach.

    Representing information in semantic nets

    Quillian distinguished between type and token nodes, corresponding to the twoways in which a word can occur in a dictionary (as entry or as part of anexplanation).Quillian distinguished furthermore between five link types:

    A is 1. subclass of B2. B modifies A3. Conjunction of a number of nodes4. Disjunction of a number of nodes5. B is subject of the relation A which has as object C(see p. 118 for a figure with an example)Quillian did not have problems with the circularity caused by the fact that he didnot set up primitives. Others took this primitive word approach (see further).A second important source was the development of Case Grammar (Fillmore).He distinguished between a surface structure and a deep structure in each

    sentence, related by a number of transformations. The deep structure has amodality (tense, mood) and a proposition (a verb with a number of cases, i.e.,roles to be filled in by other parts of the sentence).

  • 7/26/2019 Artificial_Intellegence.pdf

    27/49

    AI 27

    Inferencing in semantic nets The basic inference mechanism consists of following links between nodes. Thereare two types of strategies: Spreading of activation (or intersection search) Inheritance.The first type tries to find a concept that is related to the words to be related bythe inference. Each node can have an activation tag during this process. When anew node N is reached following a link from A or from a certain C that can bereached from A, three things can happen:1. There is already an activation tag mentioning word B associated with N:

    in this case, N is a basis for comparison.2. There is already an activation tag mentioning word A: no new tag is

    created3. There is no tag: an activation tag is created.Also here, the problem of control returns. In Quillians approach, this i s solvedby bidirectional search (first from A, then B, then A and so on). But there areother possibilities.Not all associative network representations use spreading of activation: inframes, for example, this is never used. But inheritance is. Inheritance consists inderiving properties of subclasses from superclasses.

    Some researchers do not distinguish between Inst links between tokens and theirtypes and the Sub link between classes and their superclasses. This is a mistake;you have to.

    Q9. What is Prolog ?Ans Prolog (programming in logic) is a logic-based programming language:

    programs correspond to sets of logical formulas and the Prolog interpreter useslogical methods to resolve queries. Prolog is a declarative language: you specify what problem you want to solverather than how to solve it.

    Prolog is very useful in some problem areas, such as artificial intelligence,natural language processing, databases, . . . , but pretty useless in others, such asfor instance graphics or numerical algorithms.

  • 7/26/2019 Artificial_Intellegence.pdf

    28/49

    28

    Q10. Explain Backtracking.Ans Backtracking: During goal execution Prolog keeps track of choicepoints. If a

    particular path turns out to be a failure, it jumps back to the most recentchoicepoint and tries the next alternative. This process is known as backtracking.

    Q11. Explain Traveling Salesman problem?

    Ans The traveling salesman problem is too complex to be solved via exhaustivesearch for large values of N. The nearest neighbor heuristicwork well most of thetime, but with some arrangements of cities it does not find the shortest path.

    Consider the following two graphs. In the first, nearest neighbor will find theshortest path. In the second it does not:

  • 7/26/2019 Artificial_Intellegence.pdf

    29/49

    AI 29

    Example:Consider the game of tic-tac-toe. Even if we use symmetry to reduce

    the search space of redundant moves, the number of possible paths through thesearch space is something like 12 x 7! = 60480. That is a measure of the amount of

    work that would have to be done by a brute-force search.

    Simple heuristic for tic-tac-toe: Move to the square in which X has the mostwinning lines.Using this rule, we can see that a corner square has heuristic

    value of 3, a side square has a heuristic value of 2, but the center square has aheuristic value of 4. So we canprunethe left and right branches of the search tree.This removes 2/3 of the search space on the first move. If we apply the heuristicat each level of the search, we will remove most of the states from considerationthereby greatly improving the efficiency of the search.

    Multiple Choice Questions

    1. Knowledge can be defined as

    a) Combination of state and factsb) Gathering of new informationc) Possession of facts and principles gathered

  • 7/26/2019 Artificial_Intellegence.pdf

    30/49

    30

    d) None of the aboveAns:c

    2. Horn clause is a clause having

    a) At most one positive literalb) All literal as +ve literalc) At most one ve literald) None of the aboveAns: a

    3. A.I. programming language which solves problems with a form of

    symbolic logic

    a)

    PROLOGb) Cc) VBd) JAVAAns: a

    4. In a rule based system, procedural domain knowledge in the form of

    a) Production ruleb) Rule interpretersc) Meta rule

    d)

    Common ruleAns: a

    5. The simulation approach in regard to A.I includes all of the followingproblem EXCEPT:

    Among the following which is not a horn clause?

    a) P

    b) p V qc) p q

    d) p qe) All the above.

    Ans : (d)

    6. Knowledge may beI. Declarative.II. Procedural.

  • 7/26/2019 Artificial_Intellegence.pdf

    31/49

    AI 31

    III. Non-procedural.

    (a) Only (I) above(b) Only (II) above(c) Only (III) above(d) Both (I) and (II) above

    Ans : (a)Reason : Idempotency Law is P V P = P

    7. What is the goal of artificial intelligence?

    (a) To solve real-world problems(b) To solve artificial problems(c) To explain various sorts of intelligence

    (d) To extract scientific causes(e) To restrict problems.Ans:c

    8. An algorithm is complete if

    (a) It terminates with a solution when one exists(b) It starts with a solution(c) It does not terminate with a solution(d) It has a loop(e) It has a decision parameter.

    Ans : (a)Reason : An Algorithm is complete if It terminates with a solution

    when one exists.

    9. In language understanding, the levels of knowledge that does notinclude

    (a) Phonological(b) Syntactic(c) Semantic(d) Logical

    (e) Empirical.Ans : (e)Reason : In language understanding, the levels of knowledge that does

    not include empirical knowledge

  • 7/26/2019 Artificial_Intellegence.pdf

    32/49

    32

    10. What is a heuristic function?

    (a) A function to solve mathematical problems(b) A function which takes parameters of type string and returns an

    integer value(c) A function whose return type is nothing(d) A function which returns an object(e) A function that maps from problem state descriptions to measures

    desirability.Ans : (e)

    11. Default reasoning is another type of

    (a) Monotonic reasoning

    (b) Analogical reasoning(c) Bitonic reasoning(d) Non-monotonic reasoning(e) Closed world assumption.

    Ans : (d)

  • 7/26/2019 Artificial_Intellegence.pdf

    33/49

    AI 33

    Unit 3

    Expert System

    Q.1 What is Expert System and its advantage?Ans Expert system is an A.I. program that has expert-level knowledge about a

    particular domain and knows how to use its as knowledge-based systems andknowledge-based expert systems. The experts knowledge about solving thegiven specific problems is called knowledge domain of the expert.

    Advantages of Expert SystemsAvailability:Expert systems are availabe easily due to mass production

    software.

    Cheaper:The cost of providing expertise is not expensive.

    Reduced danger:They can be used in any risky environments where humanscannot work with.

    Permanence:The knowledge will last long indefinitely.

    Multiple expertise:It can be designed to have knowledge of many experts.

    Explanation:They are capable of explaining in detail the reasoning that led to a

    conclusion.

    Fast response:They can respond at great speed due to the inherent adavantages

    of computers over humans.

    Unemotional and repsonse at all times: Unlike humans, they do not get tense,

    fatigue or panic and work steadily during emergency situations.

  • 7/26/2019 Artificial_Intellegence.pdf

    34/49

    34

    Q2. Explain MYCIN?Ans Mycinis a program that diagnoses infectious diseases. It reasons backward from

    its goal of determining the cause of a patient illness. It attempts to solve its goalof recommending a therapy for a particular patient by first finding the cause ofthe patients illness. It uses its production rule4s to reason backward from goalsto clinical observations. To solve the top-level diagnostic goal, it looks for ruleswhose right sides suggest diseases. It then uses the left sides of those rules (thepreconditions) to set up sub goals whose success would enable the rules to beinvoked . these sub goals are again matched against rules, and theirpreconditions are used to set up additional sub goals.

    Mycin is a well known rule based deduction system. Its expertise lies in the

    domain of bacterial Infections. Physicians usually must begin antibiotictreatment for patient who have bacterial infections without knowledge exactlywhich organism is the culprit. There is no time to wait for definitive laboratoryculture evidence, which accumulates too slowly. For the desperately sick,therapy must begin at once not 2 days from can either prescribe a broad spectrum drug that covers all possibilities , or she can prescribed a better, diseasespecific drug.

    Mycin helps the physician to prescribe disease specific drugs. Mycin in-forms itself about particular cases by requesting information from the physician about a

    patients symptoms, general condition. History, and laboratory test results thatcan be obtained easily and quickly. At each point, the question mycin asks isdetermined by Mycins current hypothesis and the answers to all previousquestions.

    Q.3. Explain DENDRAL.Ans DENDRAL

    DENDRAL is a program that analyses organic compounds to determine theirstructure. It is one of the early example of a successful AI program . It uses astrategy called plan-generate-test in which a planning process that used

    constraint-satisfaction techniques, creates lists of recommended andcontraindicated substructures.

    Q.4. Explain Parameter required in building of any Expert System:

  • 7/26/2019 Artificial_Intellegence.pdf

    35/49

    AI 35

    Ans (1) Meta-knowledge: Meta-knowledge is knowledge about knowledge. or we

    can say thatmeta-knowledge is systematic problem and domain independentknowledge which performs or enables operation in another more or less specificdomain dependentknowledge in different domain of human activities.Metaknowledge can be considered as a fundamental conceptual instrument insuchresearch and scientific domains as, knowledge engineering, knowledgemanagement andothers dealing with study & operations on knowledge that weknow.

    (2) Expertise Transfer: the objective of expertise transfer is to transfer expertise

    from one expert system to a computer system. these process evolved fouractivities:(a)Knowledge acquisition( from expert or other sources)(b)Knowledge

    representation( in the computer)(c)Knowledge inferencing.(d)Knowledgetransfer to user.

    (3) Domain Exploration: in general Domain knowledge in the knowledge which

    is validand directly used for preselected domain of human or autonomouscomputer activity.different specialist and expert use and develop their owndomain knowledge. Domainrefer to the knowledge that is part of the world thesystem knows about .this includeobject description, relationship and otherrelevant concepts.

    Q.5. Explain Self Explaining System. Ans The more interesting feature of expert system is their ability toexplain

    themselves, is known as self explaining system. Most system has the selfexplainingfacility that means why it asked certain question, how it arrived itsanswers. Most of theseanswers are provided by explanation module. It providesthe used with an explanation of reasoning process when requested.Expert systemcontains many modules to make it operate (Ex: memory). One of the moduleisself explaining module. This module is very much required in medical expertsystems. Themodule explains how a conclusion is arrived about a patient andwhat its basics assumptions for deriving that conclusion.

    Q6. Characterestic of expert systemAns Characteristics of Expert system (comparison between the Expert systemand

    conventional computer system)

  • 7/26/2019 Artificial_Intellegence.pdf

    36/49

    36

    1. Expert system use knowledge rather than data to control the solutionprocess. Much of the knowledge used in heuristic in nature rather thanalgorithm.

    2. The knowledge is encoded and manipulate as an entity separate fromthe control programsuch as if not compiled together with the controlprogram itself. In some cases, it is possible to use different knowledgebases with the same control program to producedifferent types ofexpert system such system are known as Expert system shells.

    3. Expert systems are capable of explaining how a particular conclusionwas drawn and whyrequested information is needed during a

    conclusion.

    4. Expert system use symbolic representation for knowledge (rules,networks or frames) and perform their inference through symboliccomputations that closely resemble manipulateof manual language.

    Q7. Advantage of Expert system:Ans 1. Expert systems do not forget, but human expert may forget.

    2. An expert system needs the symbolic representation.3. It reduces the risk of doing business.

    4. It provides the permanent documentation of the decision process.5.AnExpert system can review all the transactions but human expert systemcan onlyreview a sample.6.Expert system able to deal with uncertainty.

    5. Expert systems are not focus on abstract. It delivers the answer to goaloriented of the interview

    Q.9 Disadvantage of Expert system (or limitations of Expert system):Ans 1. Expert system has no any common sense.

    2. Expert system cannot respond creatively to unusual situation.3. Expert system must be explicitly updates while any changes in

    environments.4. Expert system is currently dependent on symbolic input.5. Measuring the performance of an expert system is difficult because we do

    not know howto quantify the use of knowledge.6.An Expert system hasaccess to highly specific domain knowledge.

  • 7/26/2019 Artificial_Intellegence.pdf

    37/49

    AI 37

    Multiple Choice Questions

    1. Mycin developed for

    a) Blood pressureb) Underground mineralsc)

    Meningitis and infectious blood deseased) None of the aboveAns: c

    2.

    PROLOG is an AI programming language which solves problems with aform of symbolic logic known as predicate calculus. It was developed in

    1972 at the University of Marseilles by a team of specialists. Can you namethe person who headed this team?

    a) Alain Colmerauer

    b)

    Nicklaus Wirthc) Seymour Papertd) None of the aboveAns: a

    3. Travelling salesman problem can be solved using

    a) Hill climbingb) Means and analysisc) Constraint satisfactiond) None of the above

    Ans: b

    4. An expert system differs from a database program in that only an expertsystem:

  • 7/26/2019 Artificial_Intellegence.pdf

    38/49

    38

    a) Contains declarative knowledgeb) Contains procedural knowledgec)

    Features the retrieval of stored informationd) Expects users to draw their own conclusionAns: b

    5. Which one is NOT the advantage of Neural Network

    a) Excellent for pattern recognitionb) Excellent classifiersc) Handles noisy data welld) None of the givenAns: d

    6. Decision trees give us disjunctions of conjunctions, that is, they havethe form: (A AND B) _______ (C AND D).

    a) ORb) ANDc) XORd) None of the givenAns: a

    7. Default reasoning is another type of(a) Monotonic reasoning(b) Analogical reasoning(c) Bitonic reasoning(d) Non-monotonic reasoning(e) Closed world assumption.Ans : (d)

    8. Consider a good system for the representation of knowledge in aparticular domain. What property should it possess?

    (a) Representational Adequacy(b) Inferential Adequacy(c) Inferential Efficiency(d) Acquisitional Efficiency(e) All the above.

  • 7/26/2019 Artificial_Intellegence.pdf

    39/49

    AI 39

    Ans: (e)

    9.

    Perception involves

    a) Hittingb) Boxingc) Dancingd) Sights,rounds,smell and touchAns : d

    10.

    Name of the computer programm that contains expertise in particular

    domain is called an

    a) Expert system

    b)

    Personel informationc) Human logicd) NoneAns:a

    11.

    Component of expert system

    a) Inference engineb) User interfacec) Knowledge based) None of the above

    e)

    All of the aboveAns:e

  • 7/26/2019 Artificial_Intellegence.pdf

    40/49

    40

    Key Terms

    User Interface: Provides the means for dialog between the user and system.

    Explanation facility: Provides the user with Explanations of how a conclusion was

    reached or why a piece of knowledge is needed. They also need to be convinced that thesolution isappropriate and applicable in their circumstances.

    Inference Engine

    accepts user input quarries and response to questions through the user interface and

    uses this dynamic information together with the static knowledge (the rules andfacts)stored in the knowledge base.

    The inference processis carried out recursively in three stages (I) match (II) select (III)

    execute.During the match stage, the contents of working memory are compared to factsand rulescontained in the knowledge base.

    Knowledge base contains facts and rules about some specialized knowledge domain.

    Learning module implies that an organize or machine must be able to adapt to new

    situations.The job of Knowledge engineer is to extract the knowledge from the expert

    and other sourceslike book, journals, article etc.

    Adaptive Interface A computer interface that automatically and dynamically adapts to

    the needs and competence of each individual user of the software.

    Agents Agents are software programs that are capable of autonomous, flexible,

    purposeful and reasoning action in pursuit of one or more goals. They are designed totake timely action in response to external stimuli from their environment on behalf of ahuman. When multiple agents are being used together in a system, individual agentsare expected to interact together as appropriate to achieve the goals of the overall

    system. Also called autonomous agents, assistants, brokers, bots, droids, intelligentagents, software agents.

    AI Languages and Tools: AI software has different requirements from other,

    conventional software. Therefore, specific languages for AI software have been

  • 7/26/2019 Artificial_Intellegence.pdf

    41/49

    AI 41

    developed. These include LISP, Prolog, and Smalltalk. While these languages oftenreduce the time to develop an artificial intelligence application, they can lengthen thetime to execute the application. Therefore, much AI software is now written inlanguages such as C++ and Java, which typically increases development time, butshortens execution time. Also, to reduce the cost of AI software, a range of commercialsoftware development tools have also been developed. Stottler Henke has developed itsown proprietary tools for some of the specialized applications it is experienced increating.

    Algorithm: An algorithm is a set of instructions that explain how to solve a problem. It

    is usually first stated in English and arithmetic, and from this, a programmer cantranslate it into executable code (that is, code to be run on a computer).

    Applications of Artificial Intelligence:

    The actual and potential applications are virtually endless. Reviewing Stottler Henke'swork will give you some idea of the range. In general, AI applications are used toincrease the productivity of knowledge workers by intelligently automating their tasks;or to make technical products of all kinds easier to use for both workers and consumersby intelligent automation of different aspects of the functionality of complex products.

    Associative Memories:

    Associative memories work by recalling information in response to an information cue.

    Associative memories can be autoassociative or heteroassociative. Autoassociativememories recall the same information that is used as a cue, which can be useful tocomplete a partial pattern. Heteroassociative memories are useful as a memory.Human long-term memory is thought to be associative because of the way in which onethought retrieved from it leads to another. When we want to store a new item ofinformation in our long term memory it typically takes us 8 seconds to store an itemthat can't be associated with a pre-stored item, but only one or two seconds, if there isan existed information structure with which to associate the new item.

    Autonomous AgentsA piece of AI software that automatically performs a task on a human's behalf, or evenon the behalf of another piece of AI software, so together they accomplish a useful taskfor a person somewhere. They are capable of independent action in dynamic,unpredictable environments. "Autonomous agent" is a trendy term that is sometimes

  • 7/26/2019 Artificial_Intellegence.pdf

    42/49

    42

    reserved for AI software used in conjunction with the Internet (for example, AI softwarethat acts as your assistance in intelligently managing your e-mail).

    Autonomous agents present the best hope from gaining additional utility fromcomputing facilities. Over the past few years the term "agent" has been used veryloosely. Our definition of a software agent is: "an intelligent software application withthe authorization and capability to sense its environment and work in a goal directedmanner." Generally, the term "agent" implies "intelligence", meaning the level ofcomplexity of the tasks involved approaches that which would previously haverequired human intervention.

    BacktrackingA control method used to search backwards for solutions

    ClausesEither a Prolog fact or rule.

    Cognitive Science : Cognitive Science, as a discipline, is concerned with learning how

    animals (and machines) acquire knowledge, represent that knowledge, and how theymanipulate those representations..

    Computer Vision :Making sense of what we see is usually easy for humans, but veryhard for computers. Practical vision systems to date are limited to working in tightlycontrolled environments. Synonym: machine vision

    Domain: An overworked word for AI people. "Domain" can mean a variety of thingsincluding a subject area, field of knowledge, an industry, a specific job, an area ofactivity, a sphere of influence, or a range of interest, e.g., chemistry, medical diagnosis,putting out fires, operating a nuclear power plant, planning a wedding, diagnosingfaults in a car. Generally, a domain is a system in which a particular set of rules, facts, orassumptions operates. Humans can usually easily figure out what's meant from thecontext in which "domain" is used; computers could probably not figure out what ahuman means when he or she says "domain."

    Domain Expert

    The person who knows how to perform an activity within the domain, and whoseknowledge is to be the subject of an expert system. This person's or persons' knowledgeand method of work are observed, recorded, and entered into a knowledge base for use

  • 7/26/2019 Artificial_Intellegence.pdf

    43/49

    AI 43

    by an expert system. The domain expert's knowledge may be supplemented by writtenknowledge contained in operating manuals, standards, specifications, computerprograms, etc., that are used by the experts. Synonym: subject-matter expert (SME).

    Extension LanguageA general-purpose programming language accessible to the users

    of the application created with that language. LISP dialects (including Scheme) are oftensuitable extension languages.

    Expert System

    An expert system encapsulates the specialist knowledge gained from a human expert(such as a bond trader or a loan underwriter) and applies that knowledge automaticallyto make decisions. For example, the knowledge of doctors about how to diagnose a

    disease can be encapsulated in software. The process of acquiring the knowledge fromthe experts and their documentation and successfully incorporating it in the software iscalled knowledge engineering, and requires considerable skill to perform successfully.Applications include customer service and helpdesk support, computer or networktroubleshooting, regulatory tracking, autocorrect features in word processors,document generation such as tax forms, and scheduling.

    Game TheoryGame theory is a branch of mathematics that seeks to model decision making in conflictsituations.

    HeuristicsA term describing an exploratory method of attacking a problem in which the solutionis obtained by successive evaluations of progress toward the final results.

    Inference Engine

    The part of an expert system responsible for drawing new conclusions from the currentdata and rules. The inference engine is a portion of the reusable part of an expert system(along with the user interface, a knowledge base editor, and an explanation system),that will work with different sets of case-specific data and knowledge bases.Knowledge-based Representations

    The form or structure of databases and knowledge bases for expert and other intelligentsystems, so that the information and solutions provided by a system are both accurateand complete. Usually involves a logically-based language capable of both syntacticand semantic representation of time, events, actions, processes, and entities. Knowledge

  • 7/26/2019 Artificial_Intellegence.pdf

    44/49

    44

    representation languages include Lisp, Prolog, Smalltalk, OPS-5, and KL-ONE.Structures include rules, scripts, frames, endorsements, and semantic networks.Knowledge

    knowledge--1. In artificial intelligence, symbolic information used by a domain expertto solve problems. 2. Facts and relationships used to solve problems.

    Knowledge-based Systems

    Usually a synonym for expert system, though some think of expert systems asknowledge-based systems that are designed to work on practical, real-world problems.

    Knowledge Elicitation

    Synonym: knowledge acquisition.

    Knowledge EngineeringKnowledge engineering is the process of collecting knowledge

    from human experts in a form suitable for designing and implementing an expertsystem. The person conducting knowledge engineering is called a knowledge engineer.

    Knowledge Representation

    Knowledge representation is one of the two basic techniques of artificial intelligence,the other is the capability to search for end points from a starting point. The way in

    which knowledge is represented has a powerful effect on the prospects for a computeror person to draw conclusions or make inferences from that knowledge. Consider therepresentation of numbers that we wish to add. Which is easier, adding 10 + 50 inArabic numerals, or adding X plus L in Roman numerals? Consider also the use ofalgebraic symbols in solving problems for unknown numerical quantities, comparedwith trying to do the same problems just with words and numbers.

    LISP

    LISP (short for list processing language), a computer language, was invented by JohnMcCarthy, one of the pioneers of artificial intelligence. The language is ideal for

    representing knowledge (e.g., If a fire alarm is ringing, then there is a fire) from whichinferences are to be drawn.

    Machine Learning:

  • 7/26/2019 Artificial_Intellegence.pdf

    45/49

    AI 45

    Machine learning refers to the ability of computers to automatically acquire newknowledge, learning from, for example, past cases or experience, from the computer'sown experiences, or from exploration.

    machine code--An operation code that a machine is designed to recognize.

    Natural Language Processing

    The study of strategies for computer programs to recognize and understand languagein spoken and written form.

    Neural NetworksNeural networks are an approach to machine learning which

    developed out of attempts to model the processing that occurs within the neurons of thebrain.

    Procedural Language The traditional programming that is based on algorithms or a

    logical step-by-step process for solving a problem.PropositionAn expression about an object which can have either a true or false value.Propositional CalculusThe formal logic system used to define the true or false values

    of objects.

    Pattern Recognition

    1.The recognition of forms, shapes, or configurations by automatic means. A subfield ofartificial intelligence. 2. The use of a computer to identify patterns.. 3. The use of

    statistical techniques and templates to process and classify patterns of data

    Plan RecognitionThe goal of plan recognition is to interpret an agent's intentions by ascribing goals andplans to it based on partial observation of its behavior up to the current time. Diviningthe agent's underlying plan can be useful for many purposes including: interpreting theagent's past behavior, predicting the agent's future behavior, or acting to collaboratewith (or thwart) the agent.

    Roboticsis the branch of technology that deals with the design, construction, operation,

    structural disposition, manufacture and application of robots.RuleA clause that defines the relationship or relationships between facts and objects.

  • 7/26/2019 Artificial_Intellegence.pdf

    46/49

    46

    Relevance Feedback

    Relevance feedback methods are used in information retrieval systems to improve theresults produced from a particular query by modifying the query based on the user'sreaction to the initial retrieved documents. Specifically, the user's judgments of therelevance or non-relevance of some of the documents retrieved are used to add newterms to the query and to reweight query terms. For example, if all the documents thatthe user judges as relevant contain a particular term, then that term may be a good oneto add to the original query.

    Rule-based System

    An expert system based on IF-THEN rules for representing knowledge.

    Scheme Langauge A LISP dialect often used within computer science curricula and

    programming language research.

    Speech Recognition The ability of a computer to understand spoken words for the

    purpose of receiving commands and data input from the speaker.

    Source CodeSymbolic coding in its original form before being processed by a computer.

    SimulationA simulation is a system that is constructed to work, in some ways, analogously toanother system of interest. The constructed system is usually made simpler than theoriginal system so that only the aspects of interest are mirrored. Simulations arecommonly used to learn more about the behavior of the original system, when theoriginal system is not available for manipulation. It may not be available because of costor safety reasons, or it may not be built yet and the purpose of learning about it is todesign it better. If the purpose of learning is to train novices, then cost, safety, orconvenience are likely to be the reasons to work on a simulated system. The simulationmay be a computer simulation (perhaps a realistic one of a nuclear power station's

    control room, or a mathematical one such as a spreadsheet for "what-if" analysis of acompany's business); or it may be a small-scale physical model (such as a small-scalebridge, or a pilot chemical plant).Turing Test--A game to determine whether a computer might be considered to possess

    intelligence, developed by British mathematician Alan Turing. Participants include two

  • 7/26/2019 Artificial_Intellegence.pdf

    47/49

    AI 47

    respondents (a computer and a human) and a human examiner who tries to determinewhich of the unseen respondents is the human. According to this test, intelligence andthe ability to think would be demonstrated by the computer's success in fooling theexaminer.UnificationThe pattern matching technique used by Prolog to match goals and sub-

    goals in a program.

  • 7/26/2019 Artificial_Intellegence.pdf

    48/49

    48

    Abbreviations

    AI : Artificial IntelligenceQA: Question AnsweringIR: Information RetrievalIE: Information ExtractionNLP: Natural Language ProcessingXML: Extensible Markup LanguageAIML: Artificial Intelligence Markup LanguageALICE: Artificial Linguistic Internet Computer Entity

    PNAMBIC: Pay No Attention to that Man Behind the Curtain

  • 7/26/2019 Artificial_Intellegence.pdf

    49/49

    AI 49

    Bibliography1. Charniak, E.: Introcuction of Artificial Intelligence, Narosa Publishing House.2. Winton. P.H. : LISP, Narosa Publishing House.3. Marcellus: Expert System Programming in TURBO PROLOG Prentice-

    Hall Inc. 1989.4. Clark, K. L. & McCabe, F.G.: Micro-Prolog Prentice-Hall Inc. 19875. Elaine rich & Kevin Knight: Artificial Intelligence and Expert System, PHI.