Steps towards an Ontology based learning environment
Onto hr poster_final_print
1. Ildikó Szabó, Réka Vas, Corvinno Technology Transfer Center, Hungary
Gábor Kismihók, Stefan Mol, University of Amsterdam, The Netherlands,
Francesco Zoino, Giovanni Sorrentino, Valentina Castello, Dida Network, Italy
w w w . o n t o h r. e u Patrick Duijts Qompas, The Netherlands w w w . o n t o h r. e u
COMPETENCY MATCHING BETWEEN VOCATIONAL EDUCATION Visit OntoHR!
AND THE WORKPLACE WITH THE HELP OF ONTOLOGIES
THE ONTOHR PROJECT Try out OntoHR!
Background Methods
• Broad educational qualifications are too crude for purposes of personnel selection • Job role selection: The role of ICT SYSTEMS ANALYST was selected as the required job knowledge and competencies are
• Why does General Mental Ability (GMA) predict job performance so well? “The reason is that people who are more intel- well documented. Furthermore, the selection was based on the fact that the competencies required for this job role were
ligent learn more job knowledge and learn it faster, the major determinant of job performance is not GMA but job knowl- relatively invariant across the Italian and Dutch contexts in which this research is being conducted. This latter finding was
edge” (Schmidt and Hunter, 2001). the result of structured interviews that were carried out with Subject Matter Experts in both countries.
• The development of job specific knowledge tests is highly labor intensive and requires a thorough understanding of the • Competency definition: A competency is a temporally stable, narrowly defined, and trainable latent ability to complete an
job in question; more so than off the shelf personality inventories or GMA tests. Yet, job knowledge, as opposed to general organizationally valued prospective job task successfully (cf. Tett et al., 2002). For the purposes of the current project we
mental ability or personality, has the major advantage of being malleable, thus allowing for a person centered approach focus specifically on those competencies that are contingent upon an identifiable, specific, and distinct educational knowl-
to personnel selection. edge domain.
• With the help of ontologies – previous experience and qualifications of applicants are evaluated against entry require- • Competency profiling: Inductive derivation, through desk research, of the universe of knowledge based competencies re-
ments (skills and competencies) for the job role. quired for the ICT systems analyst job role across countries (Italy and the Netherlands) and organizations on the basis of:
o Competency profiles developed by academics (e.g., O*Net and WISCO databases)
Purposes and expected outcomes o Competency profiles in use (Philippines National ICT competency standards)
• To create a knowledge and mental ability -based qualification to job matching system with the overriding purpose of tack- o Competencies listed in job vacancy announcements
ling the conversion of vocational education qualifications into job related competencies • Linking Competencies, Job Knowledge, and Mental Ability: Based on a review of the personnel selection literature, it was
• To assist candidates in identifying their knowledge deficiencies vis-à-vis a particular job role, and to offer remedial learn- established that both job knowledge and mental ability are among the strongest predictors of job performance, across
ing content to redress any such deficiencies jobs, organizations and countries. Using competencies as our operationalization of job performance, we therefore set
• To assist decision makers in making content, construct, and criterion valid selection decisions using an ontology-based out to populate the ontology based selection system with the prerequisite knowledge and mental abilities for each of the
knowledge and mental ability test. competencies.
• Address the weaknesses of particular VET curricula, and thereby provide ad-hoc support • In addition to a job role ontology that incorporates the aforementioned job knowledge domains, mental abilities and job
role competencies, the system is further equipped with an educational ontology that includes the Vocational Educational
output competencies aimed at the job role of ICT systems analyst.
The relevant competencies are chosen here. Skills, knowledge, attitudes.
Organization Layer Applicant Layer Educational Institution Layer
With using competencies to describe a scope of activity (job role) the work design, se- The extended version of this training a selection system will be capable of comparing
lection-, training-, or payment system will be more flexibly fitted to the organizational the job requirements related to ISA job profile to certain training programmes’ learning
Competency outcomes. This task will be achieving by matching two ontologies with each other: Job
requirements (Lawler, 1994), moreover it has strategic importance (Schoonover and based job
Andersen, 2000). So the usage of a competency model (a set of competencies that descripiton Role Ontology and Learning Outcome Ontology. The elements of these ontologies are
Personal
include the key behaviours required for excellence performance in a particular role) is Organisation Applicant competencies
Educational the instances of „Competence” class in Education Ontology Model. These competencies
Institution
more profitable for an organisation instead of listing general tasks. will be chosen in „Select relevant ontology components” phase on one hand and will be
Education
Ontology Model Output extracted from details descriptions of the traning programmes. To find the appropriate
Competency matching algorithm we take into consideration algorithm used logical constraints like
Description
in SAKE project (Futó et al., 2008), algorithms based on linguistic similarity (Alasoud et
Training al., 2008), or general algorithm like finding morphism between two ontologies (Kalfo-
Workplace programmes
Group of Tasks glou & Schorlemmer, 2003). Aſter executing the adequate algorithm the missing and
specifies
Education the surplus competencies of the training programmes will be revealed in Competency
serves Ontology Model
is part of Matching Report.
Scope of specifies
Competency
is part of Task is part of
Select relevant
Activities serves Competency Matching
ontology
Matching Report
components
requires requires
prerequisite
prerequisite
is part of
Competence
Mental Ability requires Competence element of
Module
ensures VE Ontology Domain Ontology
Job Student/Applicant
Open a
requires ensures position application
Competency
Adaptive Testing
prerequisite Matching
element of
is part of
Knowledge Curriculum
belongs to
Area Module
requires
knowledge of
Mental Ability A
is part of
A’
is part of is part of Test
refers to
premise
refers to
Basic Concept Theorem Example Set up the
conclusion selection test
Learning Environment
refers to
Knowledge Test B B’
Person 2
Education & Training refers to
Register the
applicants
Authoring
Environment
Education Ontology Model is an ontology that provides support for capturing regu-
larities in a single framework general enough to model the curriculum content man-
agement requirements of multiple institutions or application. A detailed description
Taking the test
of the structure and major classes of the ontology model is provided in Vas et al.
(Vas, Kovács, & Kismihók, 2009).
Computing test
results
Applicant
Ranking
Personal
Students’ knowledge is evaluated by using multiple-choice question tests. All ques- mental ability
Applicant report
tions and possible answers reside in the Test Bank and connected to a specific con- Competency Test results Personal results
cept in the ontology. This way, learners’ knowledge about a particular concept can be Allignment analysis analysis
assessed. The Test Item Editor component is responsible for visualizing the ontology
structure and allowing users to allocate questions to each node. Numerous questions Personal
knowledge gap
can be assigned to any concepts in the ontology. report
Traning Needs
Analysis We want to compare the personal selection test results of the applicants qualified in a
certain training programme to the results related to this particular programme in the
Evaluation of Competency Matching Report. Here we can see to what degree applicants from a par-
Decision educational
programmes
ticular educational institute have actually mastered the competencies that are claimed
Programme as output competencies by their vocational educational institutes.”
Evaluation
Report
Results achieved so far
• Established an extensive network with relevant industry and educational stakeholders, both in the Netherlands and Italy Innovative aspects of the solution
• Developed a valid competency profile for the job role that adequately and accurately describes the demands of the job. Although a handful of ontology based systems have been successfully implemented both within the fields of HRM and edu-
• Developed the domain ontology model for both the education and job role ontologies. cation, the readily apparent desirability of bridging the vocational education – workplace divide, by means of interconnected
• Deployed both models for the job role in question VET and domain ontologies as outlined in the current research to the best of our knowledge, is unique.
• Filling this increasingly conspicuous research gap may in due time put an end to the arduous process of first testing stu-
What remains to be done dents to allow them to successfully exit vocational education, only to test them again upon organizational entry.
• Piloting: At least two pilots are being set up to test the feasibility of the system. The pilots will be aimed at: • In due time, we foresee that this technology might further facilitate the blurring of vocational education to workplace
o Gauging user experiences: Variables to be assessed include: justice and fairness, perceptions, face validity, user friend- boundaries, by allowing the adequate and accurate measurement of time to proficiency in a particular occupation, while
liness, time needed to complete the assessment, etc. at the same time continuing the delivery of training content that is tailored to the needs of the individual student.
o Validation: Suggested selection decisions regarding particular applicants will be compared and contrasted with
• The decisions suggested by the selection system currently in use (e.g, structured interviews, general mental ability tests, Valorization and future research
personality tests, etc.). • Expand the scope of the system to include various interrelated job roles in a particular industry, so that the system may
• Self-ratings of applicants’ perceived demands abilities fit. be used for making both selection and placement decisions.
• Subject Matter Expert judgments about the suitability of applicants for the job role in question • Seek to employ artificial intelligence solutions to ease the work involved in populating and keeping up to date the knowl-
• Performance appraisals edge and competency elements of the system
• Seek to incorporate “soſt” competencies, such as interpersonal skills, in addition to the “hard” competencies that are the
focus of the current project.
OntoHR - Ontology Based Competency Matching Between the Vocational Education and the Workplace
Project Nr: 504151 - LLP -1 - 2009-1-HU-LEONARDO-LMP
PROJECT PARTNERS
This project has been funded with support from the European Commission. This communication reflects the views only of the author, and the Commission cannot be held responsible for any use which
may be made of the information contained therein.
OntoHR_A0_Poster_jav2.indd 1 2/16/11 5:55 PM