The ability to model business knowledge is one of the core competences in the discipline of Information Systems. In addition, the modeling of business processes and structures has found wide acceptance in study programs for economic sciences. In addition to the knowledge of modeling paradigms, notations and tools, an effective acquisition of skills in this area also requires a deeper understanding of a professional domain, preferably from the point of view of a service provider or producer. Exercises that only rely on the analysis of existing business models or on the implementation of purely textual descriptions are not sufficient to establish a deeper understanding of the subject as well as an effective competence acquisition. The paper aims to show how a didactic concept, which takes up existing domain knowledge and explicitly focuses on a positive error culture, contributes significantly to the acquisition of competence in knowledge engineering by students.

1. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 1
Prof. Dr. Vera G. Meister
Katowice, April 2017
Competency Acquisition in Applied Knowledge Engineering
An Approach based on Learning from Errors

2. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 2
Agenda
06.04.2017Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
▪ Definition of the Subject Domain
▪ Relevant Competencies
▪ Scope of Competency Acquisition in KE
▪ Learners and their Prerequisites
▪ Didactic Setting
▪ Opportunities of Errors
▪ “Learning from Errors” in Theory and Practice
▪ Survey and Results
▪ Lessons Learned

3. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 3
Knowledge in Organizations
09.04.2017
Can be found in …
➢ determined structures – organization charts
➢ described procedures – process models
➢ Implemented processes – programs and rules
➢ the heads of employees.
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
It is typically …
➢ heavily fragmented and continiously changing,
➢ distributed along departments, groups, and
persons,
➢ requires human interpretation to be accessed.
Knowledge Engineering is the processing of
organizational knowledge in a way, that it becomes
- described uniquely and unambiguously,
- both contentally and formally.
Formal
Modeling
Comp.: Studer, R.; Benjamins V.R.; Fensel, D.: Knowledge Engineering - Prinicples and Methods. Data & knowledge engineering 25.1-2 (1998): 161-197.

4. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 4
Items to be Modeled
09.04.2017Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
1. Business processes
a) Predefined procedures
e. g. order processing in eCommerce
b) Knowledge-intensive case processing
e. g. development of individual quotations
2. Decisions
e. g. about the proper discount for a customer
3. Vocabularies, glossaries and thesauri
e. g. for a specific knowledge domain
4. Structures and relations
e. g. in enterprises or between business objects

5. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 5
09.04.2017Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Subject Domain: Applied Knowledge Engineering
DMN
Digitized
Processes
IT Support for
Knowledge Work
Automated
Decisions
Big Data
Analysis
AI &
Linked Data
SKOS
A set of
bridging
technologies

6. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 6
Relevant Competencies
06.04.2017
(1) Analyze and define real world problems and structures
(2) Communicate with stakeholders
(3) Choose adequate levels of abstraction
(4) Formalize real world concepts, individuals and their relations
(5) Choose the proper notation or specification
(6) Apply notations and specifications syntactically correct
(7) Choose proper tools and use them adequately
(8) Reflect the own work critically
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors

7. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 7
Scope of Competency Acquisition in Knowledge Engineering
06.04.2017Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Competencies
Knowledge
Skills
Notations
Domain
Methods
Tools
Applications
Analysis
Abstraction
Formalization
Reflection
Patterns

8. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 8
Learners and their Prerequisites
09.04.2017
❖ Students at medium and higher levels
▪ Business Administration Master
▪ Business Information System Bachelor
❖ During their study they have already acquired
▪ Basic knowledge about modeling methods and notations
▪ Basic skills on tools and applications
▪ Basic competencies of abstraction and reflection
But the most important thing is: deep expertise in a dedicated knowledge domain
e. g. from work, from internships or from leisure activities
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors

9. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 9
06.04.2017
Overall Didactic Setting: One Cycle for each Topic
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
• Domain definition
• Knowledge description
• Knowledge engineering
• Personal consultation
• Remote feedback
• Group reflection
• Reflection questions
• Tool-based training
• Transfer assignments
• Impuls lectures
• Video lectures
• Reading material
Input Transfer
ProjectCoaching

10. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 10
09.04.2017
Opportunities for Errors
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Knowledge Engineering in an individual domain generates a multitude of error oportunities:
▪ choosing an inadequate level of abstraction,
▪ incomplete analysis and description of structures, procedures and relations,
▪ neglecting the restrictions or support options of tools,
▪ incorrect usage of notation syntax,
▪ weak or erroneous implementation of notation logic and semantics,
▪ implementing suboptimal patterns.
Knowledge Engineering in an individual domain is a creative work in a complex environment.
 Error opportunities cannot be forseen in total.
Each error and its discussion may lead to
✓ a deeper understanding of the domain,
✓ more solid modeling skills based on deeper understanding of the notation.

11. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 11
The Theory of „Learing from Errors“
10.04.2017
(1) Karl Popper: Objective Knowledge – An Evolutionary Approach. Clarendon Press, Oxford, 1972, p. 344f.
• Horizon of expectations is a kind of a frame of reference
• Important kind of problems: experience of something contrary
to our expectations – may act on the frame like a bombshell
• Learners are forced to reconstruct their whole frame of reference
 Horizon of expectations is raised to a higher level, „damaging“
observations are integrated in a new frame of explanation
(2) Janosch M. Türling e. a.: Teachers‘ Knowledge about Domain Specific Student Error. In: Eveline Wuttke,
Jürgen Seifried (eds.): Learning from Errors at School and at Work. Budrich, Opladen, 2012, p. 96.
Three facets of Professional Error Competence of Teachers:
a) Knowledge of possible error types
b) Available strategies of action/teacher reactions
c) A constructive view on errors and their use in classroom interaction
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors

12. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 12
10.04.2017
The Process of „Learning from Errors“?
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Logical Error: There is a logical problem.
First of all, the supplier list should be split
into two lists depending on the volume of
the year. A parallelization must then be
carried out, since both lists are processed
further. One (under € 1 million) must be
supplemented by the delivery risks and
sorted according to this criterion in
descending order. Then the top 3 can be
selected. The other is sorted according to
volume and the top 15 is selected. After
combining these two paths, the lists can
then be aggregated. Basically one could
also do this one after the other, without
parallelization.
Your modeling implies that only one vendor
is checked for its volume. This is not true.
To work with loop activities would not be
useful either, you will get a list and want to
cut this list according to certain criteria.
Style Error: This is not
clearly specified. Do you
mean every year in October?
Then write: each October
Syntax Error: Within a
pool sending and
receiving messages
doesn’t make sense, since
the sequence flow carries
all data and documents.

13. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 13
Improved (final) Model
10.04.2017Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Correct Style
Correct Syntax
Correct Logic
✓ deeper understanding of the domain
✓ more solid modeling skills
✓ deeper understanding of the notation

14. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 14
Survey Outline
10.04.2017
❖ Participants: all active students (N=21) of the course in winter term 2016/17
❖ Paper questionnaire, anonymized, carried out in February 2017
❖ Analyzed items:
• Self estimation of knowledge and skills at the start and at the end of the course
• Usage rates and helpfulness of different forms/media of learning
• Possibly missed alternative forms/media of learning
• Personal feeling of being confronted with „making errors“
• Agreement with the following statements:
➢ The opportunity to make my own errors and learn from them
has contributed significantly to my learning success.
➢ The opportunity to model in my own domain of expertise
has contributed significantly to my learning success.
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors

15. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 15
06.04.2017
Self Estimation of Knowledge and Skills Acquisition
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
Business Vocabularies
Business Decisions
Business Processes
Start End
non-existent excellent

16. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 16
10.04.2017
Usage Rate and Helpfulness of Forms/Media of Learning
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
never
used
often
used
most
helpful
not
helpful
On-site lectures
Video lectures
eMail feedback
Individual consultation
Usage RateHelpfulness

17. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 17
10.04.2017
Personal Feeling about „Learning from Errors“
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
bad
19%
unfamiliar
38%
challenging
38%
other
5%
• About 75 % had a kind of positive
personal feeling assessing the
learning setting as unfamiliar
(courious) or challenging.
• About 25 % gave a negative
assessment to the learning setting
as unpleasant or too hard.

18. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 18
10.04.2017
Agreement with Statements
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
(1) The opportunity to make my
own errors and learn from them
has contributed significantly to
my learning success.
(2) The opportunity to model in
my own domain of expertise
has contributed significantly to
my learning success.
0
1
2
3
4
5
6
7
totally
disagree
disagree slightly
disagree
slightly
agree
agree totally agree
Count
0
1
2
3
4
5
6
7
8
9
10
totally
disagree
disagree slightly
disagree
slightly
agree
agree totally
agree
Count
➢ About 2/3 agree with the
statements, 1/3 disagree.

19. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 19
10.04.2017
Lessons Learned
Competency Acquisition in Applied Knowledge Engineering ∙ An Approach based on Learning from Errors
(1) Students in the Business Administration programs are not very familiar with the
presented learning setting. They evaluated the course challenging, maybe too hard.
(2) All active students achieved at least good results, in an oral feedback round they all
reported about growing motivation because of personal domain expertise and intensive
coaching.
(3) The competency acquisition during that short period can be estimated as above-
average, mostly due to substantial practice in individual modeling.
(4) The same applies to the effort of the lecturer to give that different feedback on a
multitude of models and development steps.
 Example result page: http://univera.de/FHB/KITP16/Eichler/

20. Technische Hochschule Brandenburg · Brandenburg University of Applied Sciences Page 20
Thank you for your attention!
Prof. Dr. Vera G. Meister • vera.meister@th-brandenburg.de • +49-175-5634180
Research Group Business Modeling and Knowledge Engineering (BMaKE)