Personality Assessment Ontology (PAO)

1
PersonalityAssessmentOntology(PAO)

• Brian Donohue and J. Neil Otte
• University at Buffalo

2
personality

• a peculiar combination of affective, emotional,
  cognitive, and behavioral response patterns held
  by an individual

3
Outline

1. Linking personality tests to the Five Factor
   Model
2. Building PAO
3. Related Ontologies
4. Further Research

4
Existing Personality Testse.g. Myers-Briggs
5
Factorial Analysis

• Factorial Analysis A statistical method used to
  describe variability among observed, correlated
  variables in terms of a potentially lower number
  of unobserved variables.
• Using Factorial Analysis, one can (for example)
  determine that six observed variables reflect
  the variations in two unobserved variables
• Exhibition, Ascendance, Dominance ?
  Extraversion
• Borderline, Low Objectivity, Psychasthenia
  ? Neuroticism

6
Five Factor Model

• Tupes and Christal (1961) found five recurrent
  factors in analyses of personality ratings in
  eight different samples
• In many ways it seems remarkable that such
  stability should be found in an area which to
  date has granted anything but consistent results.
  Undoubtedly the consistency has always been
  there, but it has been hidden by inconsistency of
  factorial techniques and philosophies, the lack
  of replication using identical variables, and
  disagreement among analysts as to factor titles
  (p. 12)

7
Five Factor Model
8
Five Factor Model
9
From Personality Assaysto Five Factors
10
Building the Ontology

• Scope What do we need to account for?
• Factors, e.g. Agreeableness
• Subfactors, e.g. Feeling
• Assays, e.g. Myers-Briggs
• Persons with Roles
• Values representing the specified output of the
  assay
• Concern Discrepancies among the questionnaires!
• Quantity Few vs. Many
• Format True/False, Multiple Choice, Ordinal vs.
  Interval
• Thoroughness for some, only two factors
  evaluated
• Negatively correlated values

11
PAO 1.0

• Initial thought If subfactors are reducible to
  the five factors, e.g. Likability (Hogan) to
  Agreeableness, then a measurement of a subfactor
  just is a measurement of one of the five factors.
• So well just say theyre equivalent classes

12
(No Transcript)
13
(No Transcript)
14
MULTIPLE INHERITANCE
15
PAO 1.0

• How do processes and the data produced by the
  assays fit into the ontology?

16
Related Ontologies

• Mental Functioning Ontology
• Emotion Ontology
• Neuropsychological Testing Ontology

17
Emotion Ontology
18
Neuropsychological Testing Ontology
19
PAO 2.0Improvements

• Assays are processes (occurrents)
• Each personality test has parts, which are each
  assays of appropriate personality dimensions
  (Five Factors)
• Each assay has a specified output, ranging from 1
  to 100
• Personality dimensions are parts of personality
• Personality is a disposition of a person
• Persons have roles evaluator, evaluant

20
PAO 2.0Some Examples

• Hogan Personality Inventory has_part some
  Likability_Assay.
• Likability_Assay has_specified_output some
  Likability_Score.
• Likability_Score is_a Agreeableness_Score.
• Agreeableness_Score is_a Scalar_Personality_Measur
  ement Datum that contains information of
  measurement of definite value (1-100) of
  Agreeableness of some Personality of some Person.
• Agreeableness is_a Personality_Dimension
  characterized by

21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Negatively Correlated Scoresx 100 - n
27
Further ResearchWed like to see

• The incorporation of standards to automate
  diagnosis.
• A significance score for each test to allow for
  aggregation of tests and test scores.

28
Thank you! Q A