Knowledge Engineering: learning and application guide


St. Petersburg State University
Graduate School of Management







T.A. Gavrilova, S.V. Zhukova

        KNOWLEDGE ENGINEERING: a learning and application guide










St. Petersburg
2012

Reviewers:

Professor A.G. Medvedev, Doctor of Economics, Graduate School of Management SPbSU Professor A.V. Smirnov Doctor of Science, deputy director of SPII RAS

Published in accordance with requirements of Curriculum Design and Development Committee, Graduate School of Management SPbSU



         Gavrilova T.A., Zhukova S.V.
        Knowledge Engineering: learning and application guide / T. A. Gavrilova, S. V. Zhukova; Graduate School of Management SPbSU. — SPb.: Publishing Centre “Graduate School of Management”, 2012. — p. 133.




           Knowledge Engineering is the discipline of mapping intellectual assets. Through this guide, students are introduced to the major practical issues of knowledge engineering techniques. Developing business information structuring skills are the key to successful knowledge representation and sharing in any organisation. Students are trained to use Mind Manager and CMap software in order to support understanding of highly multidisciplinary horizons of knowledge engineering. Applications of recent advances in information processing and cognitive science to management problems are introduced in a variety of interrelated exercises designed to form an e-portfolio. The design of an e-portfolio makes it possible to reveal the tradeoffs of visual knowledge modelling, invent and evaluate different alternative methods and solutions for better understanding, representation, sharing and transfer of knowledge.
           The guide is written to support “Knowledge Engineering” delivered to students of the “Master of International Management” graduate program.







      © Graduate School of Management SPbSU, 2012

    Contents


PREFACE....................................................................... 5
INTRODUCTION ................................................................. 7
CHAPTER 1.  CONCEPTUAL MODELLING ............................................. 8
   1.1. Intensional and extensional definitions............................... 8
   1.2. Mindmaps ............................................................. 9
   1.3. Concept maps......................................................... 11
   1.4. FRAMES............................................................... 13
CHAPTER 2.  DECISION MODELLING .............................................. 14
   2.1. Decision tables ..................................................... 14
   2.2. Decision tree ....................................................... 16
   2.3. Cause and Effect Diagram ............................................ 20
   2.4. Flowcharts........................................................... 21
   2.5. Causal chains........................................................ 25
   2.6. Fuzzy knowledge ......................................................27
   2.7. Knowledge elicitation and structuring.................................28
CHAPTER 3.  REPRESENTING KNOWLEDGE WITH ONTOLOGIES........................... 30
   3.1. Types of ontologies.................................................. 32
   3.2. Ontological Engineering.............................................. 34
CHAPTER 4.  SELF-TRAINING IN KNOWLEDGE ENGINEERING...........................35
   4.1. ROADMAP OF IN-CLASS ASSIGNMENTS.......................................35
   4.2. E-portfolio development...............................................36
   4.3 Preparing for a final exam.............................................40
APPENDIX A.  COMPUTER SCIENCE HISTORY FACTS.................................. 41
APPENDIX B.  ORCHESTRATING ONTOLOGIES........................................ 51
APPENDIX C.  BUSINESS ENTERPRISE ONTOLOGIES.................................. 75
APPENDIX D.  INFORMATION MAPPING SOFTWARE.................................... 91
APPENDIX E.  COURSE SYLLABUS “KNOWLEDGE ENGINEERING”......................... 95
APPENDIX F.  AN EXAMPLE OF AN E-PORTFOLIO................................... 105
CONCLUSION.................................................................. 128
REFERENCES .................................................................. 129

        Preface

       This guide is intended to support students in understanding the basics of knowledge engineering and structuring in order to apply intelligent technologies to various subject domains (business, social, economic, educational, humanities, etc.). The discipline of knowledge engineering gives students insight and experience in the key issues of data and knowledge processing in various companies. Via in-class discussion sessions and training, students reveal the tradeoffs of visual knowledge modelling, invent and evaluate different alternative methods and solutions for better representation and understanding, sharing and transfer of knowledge. This book is targeted at managers of different levels, involved in any kind of knowledge work. The course’s goals are focused on using the results of multidisciplinary research in knowledge engineering, data structuring and cognitive science in information processing and modern management. The hands-on character of this course fosters learning by doing, case studies, games and discussions. Practice is targeted at e-doodling with the Mind Manager and Cmap software tools.
       A good deal of the course focuses on auto-reflection and auto-formalisation of knowledge, training analytical and communicative abilities, discovery, creativity, systemic analysis of new perspectives, synthesis of evidence from disparate sources of information, and gaining new insights in this fascinating emerging field.
       Since knowledge engineering is the discipline of mapping intellectual assets, it introduces a lot of visualisation techniques to represent data and knowledge by means of business information structuring. Special software (mind mapping and concept mapping) makes it possible to amplify the positive effects of knowledge acquisition and save time for managers at the documentation stage of knowledge work.
       The assignments designed to form an e-portfolio examine a number of related topics fully described in the course syllabus, such as:
       •   system analysis and its applications;
       •   the relationship among, and roles of, data, information, and knowledge for different applications, including marketing and management, and various approaches needed to ensure their effective implementation and deployment;

Preface


       •   the characteristics of the theoretical and methodological topics of knowledge acquisition, including the principles, visual methods, issues, and programs;
       •   defining and identifying cognitive aspects for business knowledge modelling and visual representation (mind mapping and concept mapping techniques);
       •   developing different business diagrams, such as decision trees, decision tables, causal chains, etc.

       The examples in the appendices are partially comprised from real students” portfolio and may have some mistakes and errors.

        Introduction

       The need to exchange and reuse knowledge became a global problem for the scientific and research community with the exponential growth of the Internet. Knowledge engineering is not only a science that studies knowledge processing (elicitation, structuring and formalisation) for intelligent (or knowledge-based) systems development, but also contains techniques crucial for each and every modern company that considers knowledge a key intellectual asset.
       The domain of knowledge engineering has expanded greatly in recent years and now includes the elicitation (or acquisition), collection, analysis, modelling and validation of knowledge for knowledge management projects. One issue that presents particular interest is the symbolic representation of knowledge.
       Knowledge engineering principles. Since the mid-1980s, knowledge engineers have developed a number of principles, methods and tools that have considerably improved the process of knowledge acquisition. Some of the key principles may be summarised as follows:
       •  knowledge engineers acknowledge that there are different types of knowledge, and that the right approach and technique should be used for the knowledge required;
       •  knowledge engineers acknowledge that there are different types of experts and expertise, and that methods should be chosen appropriately;
       •  knowledge engineers acknowledge that there are different ways of representing knowledge, which can aid the acquisition, validation and re-use of knowledge;
       •  knowledge engineers acknowledge that there are different ways of using knowledge, and so the acquisition process can be guided by the goals of the project;
       •  knowledge engineers use structured methods to increase the efficiency of the acquisition process.
       Issues in knowledge acquisition. Some of the essential issues in knowledge acquisition are formulated as follows: experts are individuals and the owners of the

Chapter 1. Conceptual modelling

knowledge in their heads; experts have both tacit and explicit knowledge; experts are always busy and not interested in sharing knowledge; knowledge has a very specific life cycle.
      Requirements for knowledge acquisition techniques. Because of knowledge acquisition issues, special techniques are required: taking experts off the job for short time periods, allowing non-experts to understand the knowledge involved, focusing on the essential knowledge; capturing tacit knowledge, allowing knowledge from different experts to be collated, allowing knowledge to be validated and maintained.


        Chapter 1.    Conceptual modelling

      Knowledge is a high level concept of abstraction that encompasses a lot of interrelated facts from human experience. The formalisation of knowledge in clear hierarchies of concepts, terminology and explicit solutions has to overcome complicated issues of human intuition and cognition. One of the most successful ways to begin the extraction and articulation of knowledge is the visualisation of concepts and the creation of visual models of knowledge. Visualising techniques make it possible to focus on the so-called WHAT-knowledge aspects, to arrange and clarify relationships between concepts, thoughts and ideas, to observe the borders of concepts’ meanings within the domain under consideration, when a particular management problem is on the agenda.


    1.1. Intensional and extensional definitions

      A rather large and especially useful portion of our active vocabulary is taken up by general terms, words or phrases that stand for whole groups of individual things sharing a common attribute. But there are two distinct ways of thinking about the meaning of any such term.
      The extensional of a general term is just the collection of individual things to which it is correctly applied. Thus, the extension of the word "chair" includes every chair that is (or ever has been or ever will be) in the world. The intension of a general term, on the other hand, is the set of features which are shared by everything to which it applies. Thus, the intensional of the word "chair" is (something like) "a piece of furniture designed to be sat upon by one person at a time." You can find another example of intensional/extensional definitions in Fig. 1.

Chapter 1. Conceptual modelling

9

Intensional

Extensional

Fig. 1.1. Intensional and extensional definition of the term “Dog”


    1.2.  Mind maps

       The area of application for mind maps is very broad, since this type of diagramming serves to capture thoughts and ideas on paper. According to the evidence produced by the neurosciences, the human brain is a powerful biological computer with parallel nonlinear processing of electro-chemical signals. The parallel nature of thinking is reflected in the process of mind mapping that starts with putting the main concept or problem in the centre of the picture. All other items related to the key concept or problem find their place on the radially arranged branches starting from the centre of a map. In accordance with the peculiarities of visual perception, the more important one of the aspects of the key word is, the more distant it is from the centre. It is advisable to limit the number of branches to nine, as was shown by Miller in 1956; a human being has a limited capacity for processing information and cannot handle more than nine objects of attention simultaneously. The depth of knowledge on the subject of the key concept (word, problem) is explored by means of hierarchical representation of more and more detailed issues placed on the sub branches of the map. Mind maps are used to clarify one’s vision in a form that is easily transferred between managers working for any organisation. The orchestration of branches depends greatly on the semantic and logical connections between portions of information.
       Mind maps are used to structure and visualise ideas, which is one of the most important stages of any decision-making and problem-solving process. The origins of mind mapping date back to ancient papers by Porphyry of Tyros, Aristotle, and Llull.

Chapter 1. Conceptual modelling

The modern technique of mind mapping was reinvented recently by Tony Buzan. The main idea is to direct managers’ attention away from their habitual right-left and topdown processing of the pictures’ content and towards nonlinear perception of the whole map at one glance, including all the details. As the brain functions in a nonlinear way, it is proper to use curves instead of straight lines to mimic the nature of thinking, in order to increase the success of note-taking. This tool is highly advantageous when used in brainstorming sessions, when people are expected to present their raw thoughts within strict time limits.
       There are some useful tips to develop effective mind maps. These tips are based on advances in neurobiology and cognitive sciences and can be summarised as follows:
       •   Begin with the key concept and place it in the centre.
       •   Use different colours (no more than three) to emphasise the related items.
       •   Support the curves with self-explanatory pictograms and symbols.
       •   Represent the importance of the item by means of the hierarchy level of the branch in the way that the font size of the text placed on the curve and thickness of the curve decrease as the level increases.
       •   Place items of the same scale of abstraction on the same level.
       •   Limit the text above a branch to several very concise and appropriate words to articulate the item.
       •   Connect the branches with the central concept.
       •   Make the lines the same length as the word/image.
       •   Use a mind map to show the associations.



                            Getting Started.


The process of research.

                        Brainstorming
                        Taking Notes Planning an Essay                        .
                        Sifting and organising your ideas
                    \ Outlining
From Outline to Finished Work

                    Version 0.1, Jan 2 2008
Abo u: th is map z James Wilson* ^^^^^^^\Notes on this тар§

Fig. 1.2. Example of a Mind map ( by Free Mind)

Chapter 1. Conceptual modelling

11


    1.3.  Concept maps

       Concept maps are graphical tools for organising and representing knowledge. They include concepts, usually enclosed in circles or boxes of some sort, with the relationships between the concepts indicated with a connecting line linking two concepts. The words on the lines are referred to as linking words or linking phrases, and specify the relationship between the two concepts. The label for most concepts is a word, normally a noun.
       Propositions are statements about some object or event in the universe, either naturally occurring or artificially constructed. Propositions contain two or more concepts connected using linking words or phrases to form a meaningful statement. Sometimes these are called semantic units, or units of meaning. Some examples of concept maps are given in Fig. 1.3, Fig. 1.4.
       Another characteristic of concept maps is that the concepts are represented in a hierarchical fashion, with the most inclusive, most general concepts at the top of the map and the more specific, less general concepts arranged hierarchically below. The hierarchical structure for a particular domain of knowledge also depends on the context in which that knowledge is being applied or considered.


Fig. 1.3. Example of a concept map (Cmap tool)