Intelligent Decision Support Methods From the book -Intelligent Decision Support Methods by Vasant...

Intelligent Decision Support MethodsIntelligent Decision Support Methods

From the bookFrom the book-Intelligent Decision Support Methods -Intelligent Decision Support Methods by Vasant Dhar and Roger Steinby Vasant Dhar and Roger Stein

Intelligent DSS by HCH 2

Information SystemsInformation Systems

““I know of no commodity more valuable than I know of no commodity more valuable than information.”information.”

Management Information System (MIS)Management Information System (MIS)• Transaction Processing SystemsTransaction Processing Systems

– Accurate Record KeepingAccurate Record Keeping

• Decision Support Systems (DSS)Decision Support Systems (DSS)– Model-Driven DSSModel-Driven DSS

– Data-Driven DSSData-Driven DSS


Intelligence DensityIntelligence Density

DEF: A Metric for Knowledge Work Productivity.DEF: A Metric for Knowledge Work Productivity.

Knowledge IntensiveKnowledge Intensive organizations organizations transformtransform raw data raw data into something useful-knowledge-and into something useful-knowledge-and deliverdeliver the the knowledge to the part of the organization where it can be knowledge to the part of the organization where it can be used most effectively.used most effectively.

Intelligence Density: How quickly can you get the essence Intelligence Density: How quickly can you get the essence of the underlying data from the output?of the underlying data from the output?


The Vocabulary of Intelligence DensityThe Vocabulary of Intelligence Density

Quality of ModelQuality of Model• Accuracy, Explainability, Speed, Reliability..Accuracy, Explainability, Speed, Reliability..

Engineering DimensionEngineering Dimension• Flexibility, Scalability, Ease of Use,...Flexibility, Scalability, Ease of Use,...

Quality of Available ResourceQuality of Available Resource• Learning Curve, Tolerances for Noise, Complexity,...Learning Curve, Tolerances for Noise, Complexity,...

Logistical ConstraintsLogistical Constraints• Independence from Experts, Computational Ease, Development Time,..Independence from Experts, Computational Ease, Development Time,..


Dimensions of Problems and SolutionsDimensions of Problems and Solutions

Intelligence Density Dimensions: Quality of SystemsIntelligence Density Dimensions: Quality of Systems

How Well is the System Engineered?How Well is the System Engineered?

Quality of Available ResourcesQuality of Available Resources

Logistical ConstraintsLogistical Constraints


Intelligence Density Dimensions:Intelligence Density Dimensions:Quality of Systems (1/2)Quality of Systems (1/2)

Accuracy Accuracy • measures how dose the outputs of a system are to the correct or best decisimeasures how dose the outputs of a system are to the correct or best decisi

on. Can you be confident that the errors(results that are not accurate)are non. Can you be confident that the errors(results that are not accurate)are not so severe as to make the sys-tem too costly or dangerous to use?ot so severe as to make the sys-tem too costly or dangerous to use?

Explainabilitv Explainabilitv • is the description of the process by which a conclusion was reached. Statiis the description of the process by which a conclusion was reached. Stati

stical models explain the output to some degree in the sense that each indestical models explain the output to some degree in the sense that each independent variable influences or ‘explains’ the dependent variable in that it pendent variable influences or ‘explains’ the dependent variable in that it accounts for some portion of the variance of the dependent variable. accounts for some portion of the variance of the dependent variable.


• Other systems, where rule-based reasoning is involved, show exp1icitly how Other systems, where rule-based reasoning is involved, show exp1icitly how conclusions are derived, yet others, such as neural networks, generate conclusions are derived, yet others, such as neural networks, generate opaque mathematical formulas. These are sometimes referred to as 'black opaque mathematical formulas. These are sometimes referred to as 'black boxes’, because for the user they are the mathematical equivalent of the boxes’, because for the user they are the mathematical equivalent of the magician's black box: Data go in at one end and results come out the other, magician's black box: Data go in at one end and results come out the other, but you cannot (easily) see the rationale behind the conclusion.but you cannot (easily) see the rationale behind the conclusion.

Response speed Response speed • is the time it takes for a system to complete analysis at the desired level of is the time it takes for a system to complete analysis at the desired level of

accuracy. The flip side to this dimension is confidence in the sense that you accuracy. The flip side to this dimension is confidence in the sense that you can ask how confident you are that a certain period of time, within which the can ask how confident you are that a certain period of time, within which the system must provide an answer, will be sufficient to perform the analysis. system must provide an answer, will be sufficient to perform the analysis. In applications that require that results be produced within a specified In applications that require that results be produced within a specified timeframe, missing that time frame means that no matter how accurate and timeframe, missing that time frame means that no matter how accurate and otherwise desirable the results are, they will be useless in practice.otherwise desirable the results are, they will be useless in practice.

Intelligence Density Dimensions:Intelligence Density Dimensions:Quality of Systems (2/2)Quality of Systems (2/2)


How Well is the System Engineered? (1/3)How Well is the System Engineered? (1/3)

Scalability Scalability • involves adding more variables to the problem or increasing the range of vinvolves adding more variables to the problem or increasing the range of v

alues that variables can take. For example, scalability is a major issue whalues that variables can take. For example, scalability is a major issue when you're interested in going from a prototype system involving 10 variablen you're interested in going from a prototype system involving 10 variables to one with 30 variables. Scalability can be a real problem when the ines to one with 30 variables. Scalability can be a real problem when the interactions among variables increase rapidly in unpredictable ways with thteractions among variables increase rapidly in unpredictable ways with the introduction of additional variables(making the system brittle)or where te introduction of additional variables(making the system brittle)or where the computational complexity increases rapidly.he computational complexity increases rapidly.

Compactness Compactness • refers to how small (literally, the number of bytes) the system can be madrefers to how small (literally, the number of bytes) the system can be mad

e.Once a system has been developed and tested, it needs to be put into the e.Once a system has been developed and tested, it needs to be put into the hands of the decision makers within an organization. It must be taken out hands of the decision makers within an organization. It must be taken out into the field, be that the shop floor, the trading floor, or the ocean floor.into the field, be that the shop floor, the trading floor, or the ocean floor.



FlexibilityFlexibility• is the ease with which the relationships among the variables or their domains can be is the ease with which the relationships among the variables or their domains can be

changed, or the goals of the system modified. Most systems are not designed to be changed, or the goals of the system modified. Most systems are not designed to be used once and then thrown away. Instead they must be robust enough to perform wused once and then thrown away. Instead they must be robust enough to perform well as additional functionality is added over time. In addition, many of the business ell as additional functionality is added over time. In addition, many of the business processes that you might model are not static (i.e., they change over time). As a resuprocesses that you might model are not static (i.e., they change over time). As a result, the ability to update a system or to have the system adapt itself to new phenomenlt, the ability to update a system or to have the system adapt itself to new phenomena important.a important.

EmbeddabilityEmbeddability• refers to the ease with which a system can be coupled with or incorporated into the irefers to the ease with which a system can be coupled with or incorporated into the i

nfrastructure of an organization. In some situations, systems will be components of nfrastructure of an organization. In some situations, systems will be components of larger systems or other databases. If this is the case, systems must be able to commlarger systems or other databases. If this is the case, systems must be able to communicate well and mesh smoothly with the other components of the organization infraunicate well and mesh smoothly with the other components of the organization infrastructure. A system that requires proprietary software engineer,or specific hardware structure. A system that requires proprietary software engineer,or specific hardware will not necessarily be able to integrate itself into this infrastructure.will not necessarily be able to integrate itself into this infrastructure.



Ease of useEase of use• describes how complicated the system is to use for the businesspeople describes how complicated the system is to use for the businesspeople

who will be using it on a daily basis. Is it an application that requires a lot who will be using it on a daily basis. Is it an application that requires a lot of expertise or training, or is it something a user can apply right out of the of expertise or training, or is it something a user can apply right out of the box?box?


Quality of Available ResourcesQuality of Available Resources

Tolerance for noise in dataTolerance for noise in data• the degree to which the quality of a system, most notably its accuracy, is affthe degree to which the quality of a system, most notably its accuracy, is aff

ected by noise in the electronic data.ected by noise in the electronic data. Tolerance for data sparsenessTolerance for data sparseness

• is the degree to which the quality of a system is affected by incompleteness is the degree to which the quality of a system is affected by incompleteness or lack of data. or lack of data.

Tolerance for complexity Tolerance for complexity • is the degree to which the quality of a system is affected by interactions amis the degree to which the quality of a system is affected by interactions am

ong the various components of the process being modeled or in the knowledong the various components of the process being modeled or in the knowledge used to model a process. ge used to model a process.

Learning curve requirements Learning curve requirements • indicate the degree to which the organization needs to experiment in order tindicate the degree to which the organization needs to experiment in order t

o become sufficiently competent at solving a problem or using a technique.o become sufficiently competent at solving a problem or using a technique.


Logistical ConstraintsLogistical Constraints

Independence from expertsIndependence from experts• is the degree to which the system can be designed, built, and tested is the degree to which the system can be designed, built, and tested

without experts. While expertise is valuable, access to experts within an without experts. While expertise is valuable, access to experts within an organization can be a logistical nightmare and can be very expensive.organization can be a logistical nightmare and can be very expensive.

Computational easeComputational ease• is the degree to which a system can be implemented without requiring is the degree to which a system can be implemented without requiring

special-purpose hardware or software. special-purpose hardware or software.

Development speed Development speed • is the time that the organization can afford to develop a system or, is the time that the organization can afford to develop a system or,

conversely, the time a modeling technology would require to develop a conversely, the time a modeling technology would require to develop a system.system.


TopicsTopics

Data-Driven Decision SupportData-Driven Decision Support

Evolving Solutions: Genetic AlgorithmsEvolving Solutions: Genetic Algorithms

Neural NetworksNeural Networks

Rule-Based Expert SystemsRule-Based Expert Systems

Fuzzy LogicFuzzy Logic

Case-Based ReasoningCase-Based Reasoning

Machine LearningMachine Learning


Data-Driven Decision SupportData-Driven Decision Support

OLTP: On-Line Transaction ProcessingOLTP: On-Line Transaction Processing ISAM: Indexed Sequential Access Method, early DBMSISAM: Indexed Sequential Access Method, early DBMS RDBMS: Relational Database Management SystemsRDBMS: Relational Database Management Systems

• Data NormalizationData Normalization

• SQL: Sequential/Structured Query LanguageSQL: Sequential/Structured Query Language

EIS: Executive Information SystemsEIS: Executive Information Systems• Friendly & Intelligent User-interface Friendly & Intelligent User-interface

Data Warehousing and OLAP: On-Line Analytical ProcessData Warehousing and OLAP: On-Line Analytical Process• LAN: Local Area NetworkLAN: Local Area Network

• Data Loader->Converter->Scrubber->Transformer->Warehouse->OLAPData Loader->Converter->Scrubber->Transformer->Warehouse->OLAP


Evolving SolutionsEvolving Solutions - - Genetic Algorithms (I)Genetic Algorithms (I)

Optimization Problems:Optimization Problems:• A set of problem A set of problem variablesvariables

• A set of A set of constraintsconstraints

• A set of A set of objectivesobjectives

Example:Example:• ACME Transport, Inc., a shipping firm, needs to plan a delivery route that ACME Transport, Inc., a shipping firm, needs to plan a delivery route that

will minimize the time and cost of the shipping, but at the same time , will minimize the time and cost of the shipping, but at the same time , make delivers to all 10 of its overseas clients.make delivers to all 10 of its overseas clients.

• Exhaustive Search: evaluate all possible 10! = 3,628,800 routes.Exhaustive Search: evaluate all possible 10! = 3,628,800 routes.– Problem: If the number of clients increase to 25, then there are 25! = 1.55*10Problem: If the number of clients increase to 25, then there are 25! = 1.55*10 2525 possible possible

route. Therefore it will take a very fast computer (evaluate a million route per second) to route. Therefore it will take a very fast computer (evaluate a million route per second) to evaluate only 0.23% possible route in 4 billion years.evaluate only 0.23% possible route in 4 billion years.

• Often not a LP problemOften not a LP problem


The ExampleThe Example - - Genetic Algorithms (II)Genetic Algorithms (II)

Possible constraints to the ACME problemPossible constraints to the ACME problem• Shipping costs must be less than 70% of fee charged.Shipping costs must be less than 70% of fee charged.

• Customer waiting time must be less than 90 days.Customer waiting time must be less than 90 days.

• If a customer does more than $x of business with ACME then waiting If a customer does more than $x of business with ACME then waiting time must be less than 60 days.time must be less than 60 days.

Possible objectives to the ACME problemPossible objectives to the ACME problem• Overall delivery time is minimized.Overall delivery time is minimized.

• Overall profit is maximized.Overall profit is maximized.

• Ship fleet wear is minimized.Ship fleet wear is minimized.

• Number of repeated country visits is minimized.Number of repeated country visits is minimized.


The OriginThe Origin - - Genetic Algorithms (III)Genetic Algorithms (III)

GAs were originally developed by computer scientist John GAs were originally developed by computer scientist John Holland in the 1970s as experiments to see if computer Holland in the 1970s as experiments to see if computer programs could evolve in a Darwinian sense.programs could evolve in a Darwinian sense.

GAs are very useful for solving classes of problems that GAs are very useful for solving classes of problems that were previously computationally prohibitive, especially in were previously computationally prohibitive, especially in the area of optimization.the area of optimization.

GAs is a heuristic techniques that GAs is a heuristic techniques that cannot guaranteecannot guarantee optimal optimal solutions. Only solutions. Only near optimalnear optimal solutions can be expected solutions can be expected


The Theory of EvolutionThe Theory of Evolution - - Genetic Algorithms (IV)Genetic Algorithms (IV)

Basic Concept:Basic Concept:

• Natural Selection, i.e., SurvivalNatural Selection, i.e., Survival

Different Different kromeskromes will survive based on the compatibility of will survive based on the compatibility of their attributes with their environment. They are hunted by their attributes with their environment. They are hunted by their predators at night.their predators at night.

Each type of Each type of kromekrome represents one solution to represents one solution to the survival the survival problemproblem. . KromesKromes with better attributes have higher with better attributes have higher probability to survive and therefore reproduce,probability to survive and therefore reproduce,


IntroductionIntroduction - - Genetic Algorithms (V)Genetic Algorithms (V)

The smallest unit of a GA is called a The smallest unit of a GA is called a genegene, representing a , representing a unit of information in your problem domain.unit of information in your problem domain.

A series of these genes, or a A series of these genes, or a chromosomechromosome, represents one , represents one possible complete solution to the problem.possible complete solution to the problem.

A A decoderdecoder converts the chromosome into a solution to the converts the chromosome into a solution to the problem. (or interprets the meaning of a chromosome)problem. (or interprets the meaning of a chromosome)

A A fitness functionfitness function then is used to determines which then is used to determines which chromosome solutions are good and which are not very chromosome solutions are good and which are not very good.good.


IntroductionIntroduction - - Genetic Algorithms (VI)Genetic Algorithms (VI)

A GA randomly creates an initial population of A GA randomly creates an initial population of chromosomes and evaluates their fitness.chromosomes and evaluates their fitness.

A new A new generationgeneration (new population of chromosomes) is (new population of chromosomes) is created by created by combining and refiningcombining and refining the information in the the information in the chromosome using chromosome using • SelectionSelection

• CrossoverCrossover

• MutationMutation

The process is repeated until a satisfactory solution is The process is repeated until a satisfactory solution is found.found.


NotesNotes - - Genetic Algorithms (VII)Genetic Algorithms (VII)

Do not guarantee an optimal solution.Do not guarantee an optimal solution.

You can use a GA to solve problems that you don’t even You can use a GA to solve problems that you don’t even know hoe to solveknow hoe to solve. All you need to be able to do is . All you need to be able to do is describe a good solution and provides a fitness function describe a good solution and provides a fitness function that can rate a given chromosome.that can rate a given chromosome.

How good a solution provided by a GA is determined by How good a solution provided by a GA is determined by how good the problem is formulated.how good the problem is formulated.


Simulating the Brain to Solve Problems Simulating the Brain to Solve Problems - Neural Networks (I)- Neural Networks (I)

Learning preserves the errors of the past, as well as its Learning preserves the errors of the past, as well as its wisdom.wisdom.

The Learning Process: The Learning Process: InductionInduction• DataData• GeneralizationGeneralization• ModelModel

The Example:The Example:• Over the years, you must have a very good idea how much time you need to Over the years, you must have a very good idea how much time you need to

spend on and how to prepare a quiz to get certain grade.spend on and how to prepare a quiz to get certain grade.• That is, you build mental models based on the past experiences (data) by That is, you build mental models based on the past experiences (data) by

generalization. generalization.


The OriginThe Origin - Neural Networks (II)- Neural Networks (II)

Neural networks were first theorized as early as the 1940’s Neural networks were first theorized as early as the 1940’s by two scientists at the University of Chicago (McColloch by two scientists at the University of Chicago (McColloch and Pitts). Works was done in the mid-1950s as well and Pitts). Works was done in the mid-1950s as well (McCarthy 1956; Rosenblatt 1957) when researchers (McCarthy 1956; Rosenblatt 1957) when researchers developed simple neural nets in attempts to simulate the developed simple neural nets in attempts to simulate the brain’s cognitive learning processes.brain’s cognitive learning processes.

ANNs are simple computer programs that build models ANNs are simple computer programs that build models from data by trial and error.from data by trial and error.

Very useful in modeling complex poorly understood Very useful in modeling complex poorly understood problems for which sufficient data can be collected.problems for which sufficient data can be collected.


Nervous SystemsNervous Systems - Neural Networks (III)- Neural Networks (III)

Our nervous systems consist of a network of individual but Our nervous systems consist of a network of individual but interconnected nerve cells called interconnected nerve cells called neuronsneurons..

Neurons can receive information (Neurons can receive information (stimulistimuli) from the outside ) from the outside world at various points in the network.world at various points in the network.

The information travels through the network by generating The information travels through the network by generating new internal signals that are passed from neuron to neuron. new internal signals that are passed from neuron to neuron. These new signals ultimately produce a These new signals ultimately produce a responseresponse..

A neuron passes information on to neighbor neurons by A neuron passes information on to neighbor neurons by firingfiring or releasing chemicals called or releasing chemicals called neurotransmittersneurotransmitters. .


Nervous SystemsNervous Systems - Neural Networks (IV)- Neural Networks (IV)

The connections between neurons at which information The connections between neurons at which information transfers are called transfers are called synapsessynapses..

Information can either Information can either exciteexcite or or inhibitinhibit neurons. neurons.

Synaptic connection can be strengthened (Synaptic connection can be strengthened (learninglearning) or ) or weakened (weakened (forgettingforgetting) over time with experience.) over time with experience.

With repeated learning, one can generalize his/her With repeated learning, one can generalize his/her experience, modifying the response to stimuli, and thus experience, modifying the response to stimuli, and thus ultimately reach the level of reflexes.ultimately reach the level of reflexes.


IntroductionIntroduction - Neural Networks (V)- Neural Networks (V)

ANN involves a system of ANN involves a system of neuronsneurons (or (or nodesnodes) and ) and weighted weighted connectionsconnections (the equivalent of synapses) inside the memory (the equivalent of synapses) inside the memory of a computer.of a computer.

Nodes are arranged in Nodes are arranged in layerslayers::• Input layerInput layer• Hidden layerHidden layer• Output layerOutput layer

Through Through learninglearning (trial and error, propagating, other (trial and error, propagating, other algorithms), ANN algorithms), ANN adjusts the weightsadjusts the weights on each connections to on each connections to match the desired response match the desired response (minimize the amount of error).(minimize the amount of error).


Training Steps of a Neural Network (1/2)Training Steps of a Neural Network (1/2)

Step l:Step l: The network makes a guess based on its current weights and the input The network makes a guess based on its current weights and the input data.data.

Step 2:Step 2: The net calculates the error associated with the output (at the out,put n The net calculates the error associated with the output (at the out,put node). For example, if the desired output were 1, but the network output were 0, ode). For example, if the desired output were 1, but the network output were 0, the error would be +1, based on the difference between l and 0.the error would be +1, based on the difference between l and 0.

Step 3:Step 3: The net determines by how much and in what direction each of the we The net determines by how much and in what direction each of the weights leading in to this node needs to be adjusted. How? ights leading in to this node needs to be adjusted. How?

This is accomplished by calculating how much each of the individual weighteThis is accomplished by calculating how much each of the individual weighted inputs to the node contributed to the error,given the particular input value. Sd inputs to the node contributed to the error,given the particular input value. So, for example, if a node's output were too small, the net might need to conceno, for example, if a node's output were too small, the net might need to concentrate on (that is,increase) small or negative weights that lead up to that node. Itrate on (that is,increase) small or negative weights that lead up to that node. In essence, the network feeds back the information about how well it's doing to n essence, the network feeds back the information about how well it's doing to the neurodes in the net, and where possible problems might be.the neurodes in the net, and where possible problems might be.


Training Steps of a Neural Network (2/2)Training Steps of a Neural Network (2/2)

Step4:Step4: The net adjusts the weights of each node in the layer according to the a The net adjusts the weights of each node in the layer according to the analysis in the previous step. For example, in the case where thc output was tonalysis in the previous step. For example, in the case where thc output was too small, the neural network will try to increase the values of the positive weigo small, the neural network will try to increase the values of the positive weights since that would make the weighted sum larger. This would bring the outphts since that would make the weighted sum larger. This would bring the output closer to 1, which is what you want in this case. Similarly, the neural net sut closer to 1, which is what you want in this case. Similarly, the neural net should also try to decrease the size of the negative weights (or even make them hould also try to decrease the size of the negative weights (or even make them positive).positive).

Step5:Step5:The net repeats the process by performing a similar set of calculations The net repeats the process by performing a similar set of calculations (Step l-Step3)for-each node in the hidden layer below it. But since you canno(Step l-Step3)for-each node in the hidden layer below it. But since you cannot tell the net what the desired output of each of the hidden nodes should be (thet tell the net what the desired output of each of the hidden nodes should be (they are internal and hidden), the neural network does a kind of sensitivity analysy are internal and hidden), the neural network does a kind of sensitivity analysis to determine how large the error of each of these nodes is..is to determine how large the error of each of these nodes is..


NoteNote - Neural Networks (VI)- Neural Networks (VI)

No domain experts are needed, unlike Rule-based Systems No domain experts are needed, unlike Rule-based Systems or Fuzzy Systems.or Fuzzy Systems.

Excel at mapping relationships on to data that are noisy Excel at mapping relationships on to data that are noisy and incomplete.and incomplete.

Need adequate learning rate step size.Need adequate learning rate step size.

Avoid over-training. (may accidentally learn from noise)Avoid over-training. (may accidentally learn from noise)


Putting Expert Reasoning in a BoxPutting Expert Reasoning in a Box- Rule-Based Systems (I)- Rule-Based Systems (I)

Learn to reason forward and backward on both sides of a Learn to reason forward and backward on both sides of a question.question.

You can view much of problem solving as consisting of You can view much of problem solving as consisting of rules.rules.• Automobile/Car RepairAutomobile/Car Repair• Medical CareMedical Care• Accounting and Tax PracticeAccounting and Tax Practice• Quality ControlQuality Control

The most famous of RBS, XCON, developed in 1979 by The most famous of RBS, XCON, developed in 1979 by Digital Equipment Corp.Digital Equipment Corp.


The Basic ConceptThe Basic Concept- Rule-Based Systems (II)- Rule-Based Systems (II)

CreditBank loan application exampleCreditBank loan application example

IFIF employment stability is very low employment stability is very low

ANDAND credit history is very low credit history is very low

THENTHEN credit risk is very high credit risk is very high

The region that each rule applied as in Fig. 7.1 is called The region that each rule applied as in Fig. 7.1 is called prproblem space.oblem space. Each cube is essentially a rule. In other wor Each cube is essentially a rule. In other words, a rule “samples” a region of the problem space.ds, a rule “samples” a region of the problem space.


The Basic ConceptThe Basic Concept- Rule-Based Systems (III)- Rule-Based Systems (III)

The part before the “then” is referred to as the The part before the “then” is referred to as the condition condition part of the rule or the left-hand side (LHS), and the part part of the rule or the left-hand side (LHS), and the part after the “then” as the after the “then” as the actionaction part, or the right hand side part, or the right hand side (RHS).(RHS).

Forward ChainingForward Chaining

HypothesizeHypothesize

Backward ChainingBackward Chaining


The Basic ConceptThe Basic Concept- Rule-Based Systems (IV)- Rule-Based Systems (IV)

How the rules are used is flexible and is referred as the How the rules are used is flexible and is referred as the control strategy.control strategy.

Three basic componentsThree basic components• a rule basea rule base• working memoryworking memory• a rule interpretera rule interpreter

Steps Steps (Recognize-Act Cycle)(Recognize-Act Cycle)• Rules are matched against the dataRules are matched against the data• The interpreter selects one instantiated ruleThe interpreter selects one instantiated rule• The selected rule is firedThe selected rule is fired


The Basic ConceptThe Basic Concept- Rule-Based Systems (V)- Rule-Based Systems (V)

Differences between RBS and Decision TreeDifferences between RBS and Decision Tree• How well you understand the problem at that timeHow well you understand the problem at that time

• ModificationModification

• RBS tells nothing about how to do thingsRBS tells nothing about how to do things

• One-direction Vs. Multiple directionsOne-direction Vs. Multiple directions

The order in which rules are processed affects the resultsThe order in which rules are processed affects the results• Meta RulesMeta Rules


NotesNotes- Rule-Based Systems - Rule-Based Systems

60% to 70% of the time taken to develop rule-based syste60% to 70% of the time taken to develop rule-based systems is spent on knowledge acquisition.ms is spent on knowledge acquisition.

Only worth considering when you have experts available.Only worth considering when you have experts available.

What-if analysis using dependency network.What-if analysis using dependency network.

The difficulty: making the right rules to fire at the right timThe difficulty: making the right rules to fire at the right time.e.

Intelligent Decision Support Methods From the book -Intelligent Decision Support Methods by Vasant...

Documents

Transcript of Intelligent Decision Support Methods From the book -Intelligent Decision Support Methods by Vasant...