• This field is for validation purposes and should be left unchanged.
Guidance and results for the data-rich, yet information-poor.

KNOWLEDGE MODELING

Translate Data Mining Results Into Real-World Terms & Actions

by Chandrasekhar Karra, Ph.D.

Available Exclusively for On-Site Offerings

ABOUT THIS COURSE

Once you have extracted valuable information hidden within your data, how do you leverage it against the judgment, experience and intuition of your internal experts… and make it actionable?  Knowledge Modeling collects, stores and disseminates the business rules presently held by a valuable yet transient resource: Your domain expert!

This two-day course presents knowledge-modeling methods and tools that can be used to elicit and model a domain expert’s experience. This course illustrates how automated decision systems are built that use the modeled knowledge to simulate the expert’s decision-making process. In addition, those in attendance will learn practical techniques for capturing a domain expert’s knowledge, and identifying areas where the technology has significant business impact.

WHO SHOULD ATTEND

PROJECT MANAGERS & TECHNICAL VICE PRESIDENTS: whose organizations have irreplaceable experts with valuable problem solving expertise; expertise which should be captured, stored and distributed throughout the enterprise.

IT/IS PROFESSIONALS: whose expert problem-solving skills are in high demand, and whose time is both expensive and over-committed.

RECOGNIZED EXPERTS: who repeatedly solve similar types of problems and wish to delegate the process to an automated system that simulates decision-making.

RESEARCHERS: who frequently judge or estimate how a domain expert would arrive at a solution or a recommendation.

TECHNOLOGY PLANNERS: who survey emerging technologies in order to prioritize corporate investment.

CONSULTANTS: whose competitive environment is intensifying and whose success requires competency with knowledge modeling and related emerging information technologies.

BENEFITS OF ATTENDING

  • Identifying areas where the technology can be used to improve efficiency and substantially extend the capabilities of a data mining model
  • Observing how to efficiently model a domain expert’s decision making processes
  • Participating in the development of an elaborate knowledge-based system that instantly provides consistent solutions and recommendations
  • Creating an automated knowledge-based enterprise portal. The resulting knowledge repository will become a valuable asset to your organization

THE BUSINESS CHALLENGE

Every available corporate information resource should be leveraged in order to operate more efficiently and extend competitive edge. Almost every organization has at least two kinds of underutilized information resources:

  • Operational data collected and stored in large databases
  • Domain knowledge and expertise specific to their business

Knowledge-Based System (KBS) technology excels in applications such as diagnostics, process control, help desk applications and distributed training. Deploying knowledge-based technology often frees domain experts from repetitive tasks, and allows them to focus their creative energy on discovering new business rules. KBS technology is particularly useful when combined with data mining models. Data mining has recently become a popular tool for discovering valuable patterns and relationships in transactional data. But without expert interpretation of the output signals, data mining models are difficult to automate and translate throughout the enterprise.

Comprehensive decision support systems employ a hybrid system consisting of a data mining model and a KBS system. The result is a powerful, automated complex decision support system that leverages both data and human intelligence.

WHAT MAKES THIS COURSE UNIQUE

The primary focus of this course is to showcase technologies and techniques used in implementing knowledge-based systems. This vendor-neutral offering will use various tools and methods to develop a knowledge-based system.

A live workshop will demonstrate the techniques presented in the instructional sessions. A sample application will be selected and an attendee familiar with the problem area will perform in the role of the domain expert. The instructor along with the remaining attendees will work as Knowledge Engineers. The knowledge elicited during the KE session will be modeled into a working knowledge base.

The resulting knowledge base will be used to develop an active knowledge-based system. Source code of the modeled knowledge will be made available to attendees at the end of the course.

COURSE OUTLINE

A successful knowledge-based system implementation consists of two major components: the decision-making mechanism, and the knowledge base.

Day 1: Introduction of the fundamental principles of knowledge-base technology, concepts of knowledge modeling, practical techniques, and the process of developing a knowledge base. Discussions will follow on issues relating to the design and development of practical system.

Knowledge-Based Systems

  • History
    • Disciplines and Technologies
    • Definition
    • Background
  • Architecture
    • Inference Engine
    • Heuristics
    • Data
  • Knowledge representation
    • Semantic Networks
    • Object-Attribute-Value
    • Frames
    • Production Rules
  • Inferring strategies
    • Forward Chaining
    • Backward Chaining
    • Mixed mode
  • Knowledge Acquisition
    • Domain Experts
    • Simulation
    • Historic Data
    • Data Input
    • Pre-loaded
    • Interactive
    • Database Access
    • External Device Input
  • Typical Applications
    • Diagnostics (Medical, Equipment, etc.)
    • Process Control
    • Pattern Interpretation
    • Guidance Systems

Knowledge Modeling

  • Fundamentals
  • Definition
  • Team Composition
  • The Knowledge Engineer
    • Characteristics
    • Roles
    • Recommendations and Pitfalls
  • Tasks
    • Acquisition
    • Visualization
    • Modeling
    • Validation and Verification
  • Environment
  • Processes and Techniques
    • Unstructured Interview
    • Structured Interview
    • Open-Ended Interview
    • Procedural Simulation
    • Observation Protocol
    • Constrained Processing Task
  • Pitfalls
    • Fuzzy words
    • Complexity of explanation
    • Subjective Criteria
  • Representation
    • Graphical (Decision Trees)
    • Decision Tables
    • Text

Day 2: An application workshop will focus on implementing the methods discussed on Day 1 to solve a real-world problem. Attendees are encouraged to share applications that are suitable for a knowledge-based implementation.

  • Selection of a practical problem
  • Knowledge Modeling Workshop Session
    • Identify Domain Expert
    • Conduct Knowledge Acquisition
    • Model the knowledge
    • Program the knowledge
    • Test and validate the knowledge
  • Review the modeling process

THE PRESENTER

CHANDRASEKHAR KARRA, Ph.D. is a certified knowledge engineer with many years of experience working with knowledge-based systems, fuzzy logic, neural networks and data mining.

Dr. Karra has successfully implemented artificial intelligence technologies in areas such as manufacturing, diagnostics, image processing, knowledge and data modeling. He has designed and implemented a lube oil expert diagnostic system called MSCXpert for the Military Sealift Command, Department of the Navy. Dr. Karra also served as the knowledge engineer for the MSC project and developed the knowledge bases used by the system.

Dr. Karra also implemented a complete receivables management solution for Equifax. The system ranks accounts for collection using neural networks and interactively guides collectors through the call process using a dynamic knowledge-based system. He has developed several other solutions utilizing knowledge engineering technology as well.

This Course May Be Delivered At Your Site

Call (888) 742-2454 or send an email inquiry to receive a value-based spreadsheet quotation for training at your site.