SEN-455
Knowledge Based
Management System
Chapter Objectives
•
What Does Knowledge Codification Involve?
•
Benefits of Knowledge Codification
•
Pre Knowledge Codification Questions
•
Tools and Procedures
Knowledge Codification in the
KM
System Life Cycle
What Does Knowledge Codification
Involve?
•
Converting “
tacit knowledge
” into “
explicit usable
form
”
•
Converting “
undocumented
” information into
“
documented
” information
•
Representing
and
organizing
knowledge before it
is accessed
•
Tacit knowledge (e.g., human expertise) is
identified
and
expressed
through a form that is
•
Explicit knowledge is
organized
,
categorized
,
indexed
and
accessed
.
•
It is making institutional knowledge
visible
,
accessible
, and
usable
for
decision making
•
The organizing often includes decision trees,
decision tables etc.
•
Codification must be done in a form/structure
which will eventually build the
knowledge base
.
What Does Knowledge Codification
Involve?
Benefits of Knowledge Codification
• Instruction/training—promoting training of juniorpersonnel based on captured knowledge of senior employees
• Prediction—inferring the likely outcome of a given situation and flashing a proper warning or suggestion
for corrective action
• Diagnosis—addressing identifiable symptoms of specific causal factors
• Planning/scheduling—mapping out an entire course of action before any steps are taken
Points before initiating knowledge
codification:
• Recorded knowledge is often difficult to access
• Diffusion of new knowledge is too slow.
• Knowledge is nor shared, but hoarded (this can involve political implications).
• Often knowledge is not found in the proper form. • Often knowledge is not available at the correct time
when it is needed.
• Often knowledge is not present in the proper location where it should be present.
Modes of Knowledge Conversion
• Conversion from tacit to tacit knowledge producessocialization where knowledge developer looks for experience in case of knowledge capture.
• Conversion from tacit to explicit knowledge involves
externalizing, explaining or clarifying tacit knowledge via analogies, models, or metaphors.
• Conversion from explicit to tacit knowledge involves
internalizing
• Conversion from explicit to explicit knowledge involves
combining, categorizing, reorganizing or sorting different bodies of explicit knowledge to lead to new knowledge.
Codifying Knowledge
•
What organizational
goals
will the codified knowledge
serve?
•
Why is the knowledge
useful
?
•
How would one codify
Some Codification Tools
•
Knowledge Map
•
Decision Table
•
Decision Tree
•
Frames
•
Production Rules
•
Case-based Reasoning
•
Knowledge-Based Agents
Knowledge Map
• It is a visual representation of knowledge, not a repository • Knowledge maps originated from the belief that people
act on things that they understand and accept.
• Identify strengths to exploit and missing knowledge gaps to fill
• Can be applied in Knowledge Capture
• A straightforward directory that points people to where they can find certain expertise
• Capture both explicit and tacit knowledge in documents and in experts’ heads
Knowledge Map
• Knowledge mapping is very useful when it is required to visualize and explore complex systems.
• Examples of complex systems are ecosystems, the internet, telecommunications systems, and
customer-supplier chains in the stock market. • Knowledge Mapping is a multi-step process.
• Key can be extracted from database or literature and placed in tabular form as lists of facts.
• These tabled relationships can then be connected in networks to form the required knowledge maps.
Knowledge Map (Relationships among
Departments)
The Building Cycle
• Once where knowledge resides is known, simply point to it and add instructions on how to get there
• An intranet is a common medium for publishing knowledge maps
• Main criteria: clarity of
purpose, ease of use, accuracy of content
Decision Tables
•
More like a
spreadsheet
—divided into a list of
conditions
and their respective values and a
list of
conclusions
•
It consists of some conditions, rules, and
actions.
Decision Tables (Example I)
•
A phonecard company sends out monthly invoices
to permanent customers and gives them discount
if payments are made within two weeks. Their
discounting policy is as follows:
•
``If the amount of the order of phonecards is
greater than $35, subtract 5% of the order; if the
amount is greater than or equal to $20 and less
than or equal to $35, subtract a 4% discount; if the
amount is less than $20, do not apply any
discount.''
• We shall develop a decision table for their discounting decisions, where the condition alternatives are `Yes' and `No'.
Discount Policy (Example II)
Condition Stub Condition Entry 1 2 3 4 5 6
Customer is bookstore Order size > 6 copies
Customer is librarian/individual
IF Order size 50 copies or more (condition) Order size 20-49 copies
Order size 6-19 copies
Y Y N N N N Y N N N N N Y Y Y Y Y N N N Y N N Y N Allow 25% discount Allow 15% discount Allow 10% discount
THEN Allow 5% discount (action) Allow no discount X X X X X X
Decision Trees
• A decision tree is usually a hierarchically arranged semantic network.
• Composed of nodes representing goals and links representing decisions or outcomes
• All nodes except the root node are instances of the primary goal.
• Often a step before actual codification
• Ability to verify logic graphically in problems
involving complex situations that result in a limited number of actions
Discount Policy Customer is library or individual Less than 6 6-19 copies 20-49 copies 50 or more copies Discount is 5% Discount is 10% Discount is 15% Customer is bookstore Less than 6 copies Discount is NIL 6 or more copies Discount is 25% Discount ? Discount ? Discount ? Discount ? Discount ? Discount ? Order size ? Order size ? Bookstore Not a bookstore
Frames
• Represent knowledge about a particular idea in a
data structure
• Handle a combination of declarative and
operational knowledge, which make it easier to understand the problem domain
• Have a slot (a specific object or an attribute of an
entity) and a facet (the value of an object or a slot)
• When all the slots are filled with values, the frame is considered instantiated
Generic AUTOMOBILE Frame Specialization: VEHICLE Generalization: (STATION-WAGON, COUPE, SEDAN) . . . Year: Range: (1940 – 1990) If-Changed: (ERROR:
Value cannot be modified)
Specialization: AUTOMOBILE Generalization: (SMITH’S AUTOMOBILE, HANSON’S AUTOMOBILE) Doors: 2 SMITH’S AUTOMOBILE Frame Specialization: COUPE . . . Year: 1990
An Automobile Example
Production Rules
• Tacit knowledge codification in the form of premise-action pairs
• The premise is a Boolean expression that should evaluate to be true for the rule to be applied. • Rules are conditional statement that specify an
action to be taken if a certain condition is true • The action part of the rule is separated from the
premise by the keyword THEN.
Production Rules
• In case of knowledge-based systems, rules are based on heuristics or experimental reasoning.
• Rules can incorporate certain levels of uncertainty. • The action clause consists of a statement or a series
of statements separated by AND's or comma's and is executed if the premise is true.
•
Example:
IF
income is “average” and pay_history is “good”Production Rules (Role of inferencing)
•
Inferencing implies the process of deriving a
conclusion based on statements that only
imply that conclusion.
•
An inference engine is a program that manages
the inferencing strategies.
•
Reasoning is the process of applying
knowledge to arrive at the conclusion.
– Reasoning depends on premise as well as on general knowledge.
Case-Based Reasoning (CBR)
• CBR is reasoning from relevant past cases in a
manner similar to humans’ use of past experiences
to arrive at conclusions
• Goal is to bring up the most similar historical cases
that match the current case
• More time savings than rule-based systems • Requires strong initial planning of all possible
Generic CBR Process
User Partial Description of a New Problem Specify Attributes of Problem Match Attributes to Those in Case Base User Submits Similar Cases Case BaseRole of Planning (Earlier Steps)
•
Breaking
the KM system into
modules
•
Looking at
partial
solutions
•
Linking partial solutions via rules and
procedures to
arrive
at the
final solution
Role of Planning (Latter Steps)
•
Deciding on the
programming language
•
Selecting the right
software package
•
Developing
user interface
and
consultation
facilities
•
Arranging for the
verification
and
validation
Knowledge-Based Agents
•
An
intelligent age
nt is a program code which is
capable of performing
autonomous
action in a
timely fashion.
•
They can
exhibit
goal directed
behavior
by taking
initiative.
•
They can be programmed to interact with other
agents or humans by using some
agent
communication language.
•
In terms of knowledge-based systems, an agent
can be programmed
to learn from the user
behavior
and
deduce future behavior
for assisting
the user.
Knowledge Developer's Skill Set
Knowledge Requirements:
•
Computing technology
and
operating
systems
.
•
Knowledge repositories
and
data mining
.
•
Domain specific knowledge
.
Knowledge Developer's Skill Set
Skills Requirements:
•
Interpersonal
Communication.
•
Ability to
articulate
the project's rationale.
•
Rapid Prototyping skills
.
•
Attributes related to
personality
.
– Physical (Physique, Gender, Age, health)
– Psychological (Origin, social status, social values, attitude, occupation, group membership)