Week 16

Learning From Data

‘Process’ of Learning

“A process of filtering and transforming data into valid and useful knowledge.”

‘Goal’ of Learning:

“Final goal is to improve the qualities of communication and decision making”

Top-down approach:

“Start with a hypothesis derived from observation or prior knowledge”

Bottom-up approach:

°         No hypothesis to test

°         Unknown Patterns

°         Key relationships

Data Visualization:

“Converting and exploring data into some meaningful data visually is known as Data visualization.”

Artificial Neural Network as Learning Model:

It is modeled after human brain’s network and Simulate biological information processing via networks of interconnected neurons. Neural networks are analog and parallel

Supervised Learning:

In supervised learning, the model defines the effect one set of observations, called inputs, has on another set of observations, called outputs. In other words, the inputs are assumed to be at the beginning and outputs at the end of the causal chain. The models can include mediating variables between the inputs and outputs.

Un-Supervised Learning:

In unsupervised learning, all the observations are assumed to be caused by latent variables, that is, the observations are assumed to be at the end of the causal chain. In practice, models for supervised learning often leave the probability for inputs undefined. This model is not needed as long as the inputs are available, but if some of the input values are missing, it is not possible to infer anything about the outputs. If the inputs are also modeled, then missing inputs cause no problem since they can be considered latent variables as in unsupervised learning.

Maheen Asif

Week 15

Knowledge Transfer in E-World

Strategies of Knowledge Transfer :

°         Tacit knowledge transfer , unique in complex, non-algorithmic projects, where knowledge is mentally stored

°         Explicit Inter-team Transfer — one team shares experience with another working on a similar job at another site

E-World:

°         Internet technology to serve the internal needs of an organization

°         Link knowledge workers and users (like line managers) 24 hrs a day

Extranets

°         Extranets are the technical community to generate revenue and ensure competitive advantage

°         Extranets ensure long-lasting bonds between partners and corporate members

Groupware

°         Software that helps people work together from a distance

°         Facilitates knowledge transfer between knowledge seekers and knowledge providers

Groupware Applications:

°         Newsgroups and work-flow systems

°         Chat rooms

°         Video communication

°         Knowledge sharing groupware

E-Business

°         Brings universal access to Internet to core business process of exchanging information

°         Connects critical business systems directly to critical communities

The Value Chain in E-Business:

“A way of organizing primary and secondary activities of a business, where each provides “value added” to total operation”

Supply Chain Management:

“Supply chain management is the management of the flow of goods. It includes the movement and storage of raw materials, work-in-process inventory, and finished goods from point of origin to point of consumption”

Customer Relationship Management (CRM):

“Customer relationship management (CRM) is a model for managing a company’s interactions with current and future customers. It involves using technology to organize, automate, and synchronize sales, marketing, customer service, and technical support.”

Benefits of CRM

°         Increased customer satisfaction

°         Cross-selling products efficiently

°         Making call centers more efficient

Mehwish Shehzad

Week 14

Selecting KB Problem:

The KM system can be assured to be successful if:

°         The user(s) have prior experience with systems applications.

°         The user is actively involved in defining /identifying the specific systems functions.

°         The user is actively involved in user acceptance testing and the final system evaluation.

Ease of Understanding the KM System:

°         Reliable documentation (especially during user training) plays a key role during deployment.

°         Documentation including examples, illustrations, and graphics may reduce training time.

Knowledge Transfer:

Two Approaches used for transferring KM system technology in implementation:

°         The system is actually transferred from the knowledge developer directly to the working unit in the organization.

°         Installing the system on the resident hardware.

Integration Alternatives:

°         Technical Integration

°         Knowledge Sharing Integration

°         Decision Making Flow Integration

°         Workflow Reengineering

Maintenance:

°         Maintenance implies the way of making the required corrections which can continue to meet user's expectations.

Organizational factors:

°         Strong leadership;

°         User participation in the process.

°         Organizational politics.

°         Organizational climate.

°         User readiness (willing)

Security Issues:

°         The new system should provide password / protocol protection.

°         Security procedures should be consistently observed.

°         Access should be restricted regarding update of the KB.

“Practice without knowledge is rootless and knowledge without practice is useless”

Fundamentals of KT :

°         Should be a daily, integral part of a learning organization.

°         Transmitting (or conveying) the knowledge of one source to another source

°         Goal is to promote/facilitate knowledge sharing, increase collaboration and networking.

°         Sources: knowledge bases, experts, etc.

°         Media: LAN, secure/insecure lines, encrypted/plain text, etc.

°         Consumers: another application, a manager, a customer, etc.

Guidelines for Successful KT and Sharing:

°         Reasoning (why to do) BEFORE Processing (how to do)

°         Knowing how the Organization handles Mistakes

°         Doing is BETTER than Talking

°        Employee’s

Vocational Needs:

°         Ability Utilization

°         Advancement

°         Level of Achievement

°         Level of Creativity

°         Compensation

°         Independence

°         Authority (supervision)

Mehwish Shehzad

Week 13

SYSTEM TESTING AND DEPLOYMENT

Quality Assurance: The KM system should meet user expectations. Performance usually depends on the quality of explicit/tacit knowledge stored in the knowledge base. For the expert: quality relates to a reasoning process which produces reliable and accurate solutions within the KM system framework. For the user, quality relates to the system’s ability to work efficiently. For the knowledge developer, quality relates to how well the knowledge source is and how well the user's expectations are codified into the knowledge base.

Knowledge Testing: It is required to control performance, efficiency, and quality of the knowledge base.

Types of testing:

• Logical Testing:

To make sure that the system produces correct results.

• User Acceptance Testing:

It follows logical testing and checks the system's behavior in a realistic environment.

 Issues:

• Subjective nature of knowledge (tacit)
• Lack of reliable specifications
• Verifying correctness/consistency
• Negligence in case of testing
• Time limitations for knowledge developers to test the system
• Complexity in case of user interfaces

Logical Testing Approaches

Two approaches:

• Verify the knowledge base formation:
• The structure of the knowledge as it relates to circular or redundant errors is verified.
• Consistency, correctness and completeness of knowledge base rules are also verified.
• Verify the knowledge base functionality:

Deals with confidence and reliability of the knowledge base.

Attributes:

• Circular Errors
• Completeness
• Confidence
• Correctness
• Consistency
• Inconsistency
• Redundancy Errors
• Reliability
• Subsumption error

User Acceptance Testing Approaches:

Steps:

• Selecting a person/team for testing.
• Deciding on user acceptance test criteria.
• Developing a set of test cases.
• Maintaining a log on different versions of the tests and test results.
• Field-testing the system.

Test Team/Plan:

A testing plan indicates who is to do the testing. Commitment initiates with management support and a test team with a test plan. The team is expected to

• be independent of the design/codification of the system
• understand systems technology/knowledge base infrastructure
• be well versed in the organization's business

• Deciding on user acceptance test criteria:
• Accuracy
• Adaptability
• Adequacy
• Appeal
• Availability
• Ease of use
• Performance
• Face validity
• Robustness
• Reliability
• Operational/Technical Test
• User Acceptance Test Techniques:
• Face Validation
• Developing a set of test cases
• Subsystem Validation
• Maintaining a log on different versions of the tests/test results
• Field testing the system

Kulsum Raza

Week 12

Decision tree: is also a knowledge codification technique. A decision tree is usually a hierarchically arranged semantic network.

Frame: A frame is a codification scheme used for organizing knowledge through previous experience. It deals with a combination of declarative and operational knowledge. Key elements of frames:

      Slot: A specific object being described/an attribute of an entity.

     Facet: The value of an object/slot.

Production Rule: They are conditional statements specifying an action to be taken in case a certain condition is true. They codify knowledge in the form of premise-action pairs.

Syntax: IF (premise) THEN (action)

Rules can incorporate certain levels of uncertainty. A certainty factor is synonymous with a confidence level, which is a subjective quantification of an expert's judgment. The premise is a Boolean expression that should evaluate to be true for the rule to be applied. The action part of the rule is separated from the premise by the keyword THEN. 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.

In case of knowledge-based systems, planning involves:

• Breaking the entire system into manageable modules.
• Considering partial solutions and liking them through rules and procedures to arrive at a final solution.
• Deciding on the programming languages.
• Deciding on the software packages.
• Testing and validating the system.
• Developing the user interface.
• Promoting clarity, flexibility; making rules clear.
• Reducing unnecessary risk.

Role of inference:

• Inference implies the process of deriving a conclusion based on statements that only imply that conclusion.
• An inference engine is a program that manages the inference strategies.
• Reasoning is the process of applying knowledge to arrive at the conclusion.

Case-Based Reasoning: Case-based reasoning is a technique that records and documents cases and then searches the appropriate cases to determine their usefulness in solving new cases presented to the expert. The aim is to bring up the most similar historical case that matches the present case. Adding new cases and reclassifying the case library usually expands knowledge. A case library may require considerable database storage as well as an efficient retrieval system.

Knowledge-Based Agents: An intelligent agent is a program code which is able to perform independent action in a appropriate manner. 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.

Kulsum Raza

Week 11

Other Knowledge Capture techniques

On Site Observation: Process which involves knowledge developer observing the expert while solving the specific problem.
Brainstorming: Unstructured approach of creating ideas about solution of a problem. Questions can be raised for clarification.

Role of Knowledge Developer in Brainstorming:

General procedure for the session of brainstorming are as follows:

• Introduce and coordinate the brainstorming session
• Presenting the problem to the expert
• Prompt experts to generate ideas
• Watch for signs of possible convergence 

Electronic Brainstorming: Promotes instant exchange of ideas b/w experts that identifies objectives and structures the agenda. Through brainstorming effective communication can be made about the sensitive issues.

Protocol Analysis: Scenarios are collected by asking experts to solve the problem.

Consensus: Process of group decision making. Inputs gather from all experts and make final and better decision/solution.

Consensus Procedure:

• A proposal for resolution is put forward
• Amend and modify proposal through discussion
• Those participants who disagree with the proposal have the responsibility to put forward alternative proposals
• The one who put forward the proposal, with help of facilitator, can choose to withdraw proposal if seems to be dead end.
• When a proposal seems to be well understood and no new changes asked for, the facilitator confirm any objections
• If no objections, the facilitator can call for consensus
• If there are still no objections, then after a moment of silence, you have the decision
• If consensus appears to have reached, the facilitator repeats the decision so everyone is clear

Maheen Asif

Week 10

KM Architecture:

We classify fundamentals for KM systems in Layers that represents internal technologies of company which are as follows:

The User Interface: Interface b/w user and KM system like a web browser which should be consistent and easy to use.
Authorized Access Control: This layer ensures the unauthorized access to the info stored in company repositories. It maintains security.
Knowledge-Enabling Application: It provides knowledge Bases(i.e how performance can improve?), dbms information and aurtomation tools.

Transport Layer: Ensures that company will become a network of relationships such as LANs. WANs. intranets, extranets, and the internet.

Middleware: It makes possible to connect old and new data formats.

Physical Repositories: Most bottom layer of KM architecture where all repositories are installed.

KM System Development Life Cycle:

~Evaluate existing infrastructure~Form the KM team~Knowledge capture~Design KM blueprint (master plan)~Test the KM system~Implement the KM system~Manage change and reward structure~Post-system evaluation

Chapter 5

Capturing Tacit Knowledge

Knowledge Capturing: The process of gathering information about the expert's experience and thinking.

3 Important Steps:

• Use appropriate technique to gather info
• Interpret the info
• Build rules with the interpretations.

Level of Experts:

• Highly expert persons. 
• Moderately expert problem solvers. 
• New experts.

Individual Expert: Ideal when building simple KM system, easily conflicts can be resolved.

Multiple Experts: It can benefit complex problem domain and allow alternative way of representing knowledge.

Types of Interview:

Structured: Questions and responses are definitive.

Unstructured: Neither the questions nor their responses specified in advance.

Semi Structured: Predefined questions are asked.

Maheen Asif