Chapter 11-4: What are the business benefits of using intelligent techniques for knowledge management?

Slide #28

Greetings I’m Ma. Rebecca C. Hamo and I’m gonna be presenting the last topic in this chapter – the benefits of using
intelligent techniques for knowledge management

Slide #29

So, intelligent techniques are used to capture individual and collective knowledge and to extend the company’s
knowledge base. Capturing these knowledge enables the business to make more informed decisions.

There are various intelligent techniques that the company can use:

 First, to capture tacit knowledge, they can use an expert system, case-based reasoning, or fuzzy logic. This would
be further discussed in the succeeding slides.
 Second for knowledge discovery, they can use neural networks or data mining. As brief description since Data
Mining would not be discussed in detail in this chapter, Data Mining is used to capture undiscovered knowledge
residing in large databases; It provides managers with new insight for improving business performance.
 Third, for Generating solutions to complex problem, companies can use Genetic algorithms
 And lastly, for automating tasks, they can use Intelligent agents

Slide # 30

The intelligent techniques I have mentioned earlier are based on Artificial Intelligence (AI) technology, which consists of
computer-based systems that attempt to emulate human behavior.

Slide #31

Expert systems are used to

 Capture tacit knowledge in a very specific and limited domain of human expertise.
 They capture knowledge of skilled employees in the form of a set of rules in a software system that can be used
by others in the organization.
 These expert system can only perform very limited tasks that can be usually performed by professionals in a few
minutes or hours
 These Intelligent technique are typically used in discrete, highly structured decision-making situations. A typical
example would be determining whether to grant credit for a loan

Slide #32

The rules illustrated here are for a simple credit-granting expert systems. An expert system contains a number of rules to
be followed. The rules are interconnected, the number of outcomes is known in advance and is limited, there are
multiple paths to the same outcome, (and the system can consider multiple rules at a single time.) As you can see in the
illustration, if the income is more than 50,000, they should ask about the car payment, and if the car payment is less
than 10% of their income they proceed in asking about the mortgage payment then grant a credit line and so forth.
These shows how the rules are interconnected in an expert system.

Slide #33

How Expert Systems Work

 As you’ve seen in the previous slide, there are set of rules right? This set of rules are formed through human
knowledge. This is because human knowledge must be modeled or represented in a way that a computer can
process. Expert systems model human knowledge as a set of rules that collectively are called the knowledge
base.
  The Knowledge base: is a set of hundreds or thousands of rules; obtained by interviewing experts or
documenting business rules (manuals, reports, etc.); interconnected and nested
 The strategy used to search through the knowledge base is called the Inference engine. The two strategies are
commonly used are forward chaining and backward chaining.
o In forward chaining, the inference engine begins with the information entered by the user and searches
the rule base to arrive at a conclusion.
o In backward chaining, the strategy for searching the rule base starts with a hypothesis and proceeds by
asking the user questions about selected facts until the hypothesis is either confirmed or disproved.

Slide #34

An inference engine works by searching through the rules and “firing” those rules that are triggered by facts gathered
and entered by the user.

In forward chaining, beginning on the left, if the user enters a client’s name with income greater than $100,000, the
engine will fire all rules in sequence from left to right. If the user then enters information indicating that the same client
owns real estate, another pass of the rule base will occur and more rules will fire. Processing continues until no more
rules can be fired.

In backward chaining, it begins with a hypothesis right?, so we can start with a question first so, if the question is…
“Should we add this person to the prospect database?” Begin on the right of the diagram and work toward the left. You
can see that the person should be added to the database if a sales representative is sent, term insurance is granted, or a
financial adviser visits the client.

Slide #35

 Expert systems works best to deal with problems of classification

o That Have relatively few alternative outcomes
o And where the Possible outcomes are known in advance
 But it does have its own disadvantages. Many expert systems require large, lengthy, and expensive development
efforts.
o And sometimes, Hiring or training more experts may be less expensive than building an expert system.
o It must also be continually change
o And Some expert systems, especially large ones, are so complex that in a few years the maintenance
costs equal the development costs.

Slide #36

Case-Based Reasoning

In case-based reasoning (CBR), descriptions of past experiences of human specialists, represented as cases, are
documented and stored in a database for later retrieval when the user encounters a new case with similar parameters.
The system searches for stored cases with problem characteristics similar to the new one, finds the closest fit, and
applies the solutions of the old case to the new case. Successful solutions are tagged to the new case and both are
stored together with the other cases in the knowledge base. Unsuccessful solutions also are appended to the case
database along with explanations as to why the solutions did not work

You’ll find case-based reasoning in diagnostic systems in medicine or customer support where users can retrieve past
cases whose characteristics are similar to the new case.

Slide #37

Case-based reasoning represents knowledge as a database of past cases and their solutions. The system uses a six-step
process to generate solutions to new problems encountered by the user.
Slide #38

Fuzzy logic is a rule-based technology that can represent such imprecision by creating rules that use approximate or
subjective values. It can describe a particular phenomenon or process linguistically and then represent that description
in a small number of flexible rules.

It provides solutions to problems requiring expertise that is difficult to represent in the form of crisp IF-THEN rules such
as:

o When controlling the room temperature automatically

o autofocus device in cameras
o and in selecting companies for potential acquisition

Slide # 39

These are differences between the three intelligent techniques used in capturing tacit knowledge
Expert System Case-Based Reasoning Fuzzy Logic Systems
Capture the tacit knowledge of Capture organization’s collective Capture tacit knowledge
individual experts knowledge and expertise where there is linguistic
ambiguity.
Presented in: IF-THEN-ELSE rules Presented in cases, continuously Represent description in a
format expanded and refined by users small number of flexible rules
(range of values)

Slide # 40

Machine Learning

 The study of how computer programs can improve their performance without explicit programming.
 A machine that learns is a machine that:
o can recognize patterns in data
o change its behavior based on its recognition of patterns, experience, or prior learnings (a database)
 Examples:
o Google Search
o Recommendation systems on Amazon and Netflix

Slide #41

Neural Networks

 Neural Network find patterns and relationships in massive amounts of data that would be too complicated and
difficult for a human being to analyze.
 They “Learn” patterns from large quantities of data by searching for relationships, building models, and
correcting over and over again.
 Humans “train” the network by feeding it data inputs for which outputs are known, to help the neural network
learn solution by example.
 Neural Networks are very sensitive and may not perform well if their training covers too little or too much data.
 It is best used as aids to human decision makers instead of substitutes for them.
 Neural networks are used in the: Examples:
o The financial industry – to discern patterns in large pools of data that might help predict the
performance of equities, corporate bond ratings, or corporate bankruptcies.
o And in detecting credit card fraud (Visa International) – by monitoring all Visa transactions for sudden
changes in the buying patterns of cardholders.
Slide #42

In this illustration, the neural network has been trained to distinguish between valid and fraudulent credit card
purchases. A neural network uses rules it “learns” from patterns in data to construct a hidden layer of logic. The hidden
layer then processes inputs, classifying them based on the experience of the model.

Slide #43

Genetic Algorithms

 Genetic algorithms are useful for finding the optimal solution for a specific problem by examining a very large
number of possible solutions for that problem.
 They are based on techniques inspired by evolutionary biology, such as inheritance, mutation, selection, and
crossover (recombination).
 Genetic algorithms are used to solve problems that are very dynamic and complex, involving hundreds or
thousands of variables or formulas. The problem must be one where the range of possible solutions can be
represented genetically and criteria can be established for evaluating fitness. The problem must be one where
the range of possible solutions can be represented genetically and criteria can be established for evaluating.
 It is typically used in optimization problems (minimization of costs, efficient scheduling, etc.) in which hundreds
or thousands of variable exists.

Slide #44

This example illustrates an initial population of “chromosomes,” each representing a different solution. The genetic
algorithm uses an iterative process to refine the initial solutions so that the better ones, those with the higher fitness,
are more likely to emerge as the best solution.

Slide #45

Intelligent Agents

 Are Software programs that work without direct human intervention to carry out specific, repetitive, and
predictable tasks for user, process, or application, such as deleting junk e-mail and finding the cheapest airfare.
 It helps businesses locate and act on information that is considered important.
 It Uses limited built-in or learned knowledge base
o But some are capable of learning from experience and adjusting their behavior like Siri.
 Chatbots: are software agents designed to simulate a conversation with one or more human users via textual or
auditory methods.
 While Agent-based modeling applications:
o Are Systems of autonomous agents
o Develop to Model the behavior of consumers, stock markets, and supply chains and to predict the
spread of epidemics.

Slide #46

Procter & Gamble (P&G) used agent-based modeling to improve coordination among different members of its supply
chain in response to changing business conditions. It modeled a complex supply chain as a group of semiautonomous
“agents” representing individual supply chain components, such as trucks, production facilities, distributors, and retail
stores. Using intelligent agent models, P&G discovered that trucks should often be dispatched before being fully loaded.
Although transportation costs would be higher using partially loaded trucks, the simulation showed that retail store
stockouts would occur less often, thus reducing the amount of lost sales, which would more than make up for the higher
distribution costs.
Slide #47

The last Intelligent technique are Hybrid AI Systems

 It is when you integrate Genetic algorithms, fuzzy logic, neural networks, and expert systems into a single
application to take advantage of the best features of each.
 Example: Matsushita “neurofuzzy” washing machine that combines fuzzy logic with neural networks.

Slide #48

That’s it for our report. Thank you for listening.

Intelligent Techniques: use to capture individual and collective knowledge and to extend their knowledge base.

 To capture tacit knowledge: Expert systems, case-based reasoning, fuzzy logic

 Knowledge discovery: Neural networks and *data mining
 Generating solutions to complex problems: Genetic algorithms
 Automating tasks: Intelligent agents

The intelligent techniques are based on Artificial Intelligence (AI) technology, which consists of computer-based
systems that attempt to emulate human behavior.

Expert Systems

 Capture tacit knowledge in a very specific and limited domain of human expertise.
 Capture the knowledge of skilled employees in the form of a set of rules in a software system that can be used
by others in the organization.
 Perform very limited tasks that can be performed by professionals in a few minutes or hours (ex: diagnosing a
malfunctioning machine or determining whether to grant credit for a loan)
 Used in discrete, highly structured decision-making situations

How Expert Systems Work

 Expert systems model human knowledge as a set of rules

 Knowledge base: set of hundreds or thousands of rules; obtained by interviewing experts or documenting
business rules (manuals, reports, etc.); interconnected and nested
 Inference engine: strategy used to search through the knowledge base.
o Forward chaining: inference engine begins with the information entered by the user and searches the
rule base to arrive at a conclusion.
o Backward chaining: the strategy for searching the rule base starts with a hypothesis and proceeds by
asking the user questions about selected facts until the hypothesis is either confirmed or disproved.
 Expert systems deal with problems of classification
o Have relatively few alternative outcomes
o Possible outcomes are known in advance
 Many expert systems require large, lengthy, and expensive development efforts.
o Hiring or training more experts may be less expensive than building an expert system.
o Expert system must also continually change
o For large, complex expert systems: maintenance costs = development costs

Case-Based Reasoning
  Descriptions of past experiences of human specialists, represented as cases, are documented and stored in a
database.
 System searches for stored cases with problem characteristics similar to the new one, finds the closest fit, and
applies the solutions of the old case to the new case.
 Successful solutions are tagged to the new case and both are stored together.
 Unsuccessful solutions also are appended to the case database along with explanations as to why the solutions
did not work
 CBR is found in: diagnostic systems (medicine) or customer support

Expert System Case-Based Reasoning Fuzzy Logic Systems

Capture the tacit knowledge of
individual experts
IF-THEN-ELSE rules
Capture organization’s collective
knowledge and expertise
Presented in cases, continuously
expanded and refined by users
Capture tacit knowledge
where there is linguistic
ambiguity.
Represent description in a
small number of flexible rules
(range of values)

Fuzzy logic systems

 Rule-based technology represents imprecision by creating rules that use approximate or subjective values.
 It can describe a particular phenomenon or process linguistically and then represent that description in a small
number of flexible rules.
 Provides solutions to problems requiring expertise that is difficult to represent in the form of crisp IF-THEN rules:
o control room temperature automatically
o autofocus device in cameras
o selects companies for potential acquisition

Machine Learning

 The study of how computer programs can improve their performance without explicit programming.
 A machine that learns is a machine that:
o can recognize patterns in data
o change its behavior based on its recognition of patterns, experience, or prior learnings (a database)
 Examples:
o Google Search
o Recommendation systems on Amazon and Netflix

Neural Networks

 Find patterns and relationships in massive amounts of data that would be too complicated and difficult for a
human being to analyze.
 “Learn” patterns from large quantities of data by searching for relationships, building models, and correcting
over and over again.
 Humans “train” the network by feeding it data inputs for which outputs are known, to help the neural network
learn solution by example.
 Very sensitive and may not perform well if their training covers too little or too much data.
 Best used as aids to human decision makers instead of substitutes for them.
  Examples:
o The financial industry - discern patterns in large pools of data that might help predict the performance
of equities, corporate bond ratings, or corporate bankruptcies.
o Detect credit card fraud (Visa International) – by monitoring all Visa transactions for sudden changes in
the buying patterns of cardholders.

Expert System Neural Network

Emulate a human expert’s way of solving problems
Highly specific and cannot be retrained easily
Provide explanations for their solutions
Do not program solutions & do not aim to solve specific
problems
Generalized capability to learn
Do not always explain why they arrived at a particular
solution

Genetic Algorithms

 Useful for finding the optimal solution for a specific problem by examining a very large number of possible
solutions for that problem.
 Inspired by evolutionary biology: Search among solution variables by changing and reorganizing component
parts using processes such as inheritance, mutation, selection, and crossover (recombination).
 Able to evaluate many solution alternatives quickly.
 The problem must be one where the range of possible solutions can be represented genetically and criteria can
be established for evaluating.
 Used in optimization problems (minimization of costs, efficient scheduling, etc.) in which hundreds or thousands
of variable exists.

Intelligent Agents

 Software programs that work without direct human intervention to carry out specific, repetitive, and predictable
tasks for user, process, or application, such as deleting junk e-mail and finding the cheapest airfare.
 Help businesses locate and act on information that is considered important.
 Use limited built-in or learned knowledge base
o But some are capable of learning from experience and adjusting their behavior like Siri.
 Chatbots: software agents designed to simulate a conversation with one or more human users via textual or
auditory methods.
 Agent-based modeling applications:
o Systems of autonomous agents
o Model the behavior of consumers, stock markets, and supply chains and to predict the spread of
epidemics.

Hybrid AI Systems

 Genetic algorithms, fuzzy logic, neural networks, and expert systems can be integrated into a single application
to take advantage of the best features of each.
 Example: Matsushita “neurofuzzy” washing machine that combines fuzzy logic with neural networks.
 NOTES:
 Data Mining: capture undiscovered knowledge residing in large databases; providing managers with new insight
for improving business performance.
 AI Technology: Such systems would be able to learn languages, accomplish physical tasks (robotics), use a
perceptual apparatus that informs and language, and emulate human expertise and decision making. AI
systems extend the powers of humans but in no way substitute
 Rules: more interconnected and nested than in a traditional software program
 EXPERT VS. CBR: Expert systems primarily capture the tacit knowledge of individual experts, but organizations
also have collective knowledge and expertise that they have built up over the years