Behavioral Traits Knowledge Model

Overview

The Behavioral Traits Knowledge Model (display name: Behavioral Traits EN v#) categorizes a document based on the personality traits or attitudes it describes or contains.

The taxonomy in divided into seven categories:

Sociality
Action
Openness
Consciousness
Ethics
Capability
Moderation

Each category has three levels of polarity:

Low
Fair
High

indicating to which degree the behavioral trait belongs to the category that represents it.

Categorization and taxonomy

1000 Sociality
    1100 Sociality low
        1101 Asociality
        1102 Impoliteness
        1103 Ungratefulness
        1104 Emotionality
        1105 Isolation
        1106 Disagreement
    1200 Sociality fair
        1201 Seriousness
        1202 Introversion
        1203 Unreservedness
        1204 Humour
        1205 Sexuality
    1300 Sociality high
        1301 Extroversion
        1302 Pleasantness
        1303 Trustfulness
        1304 Gratefulness
        1305 Empathy
2000 Action
    2100 Action low
        2101 Sedentariness
        2102 Passivity
    2200 Action fair
        2201 Calmness
    2300 Action high
        2301 Initiative
        2302 Dynamism
3000 Openness
    3100 Openness low
        3101 Rejection
        3102 Apathy
        3103 Apprehension
        3104 Traditionalism
        3105 Conformism
        3106 Negativity
        3107 Bias
    3200 Openness fair
        3201 Cautiousness
    3300 Openness high
        3301 Progressiveness
        3302 Acceptance
        3303 Courage
        3304 Positivity
        3305 Curiosity
4000 Consciousness
    4100 Consciousness low
        4101 Superficiality
        4102 Unawareness
        4103 Disorganization
        4104 Insecurity
        4105 Ignorance
        4106 Illusion
    4300 Consciousness high
        4301 Awareness
        4302 Spirituality
        4303 Concern
        4304 Knowledge
        4305 Self-confidence
        4306 Organization
5000 Ethics
    5100 Ethics low
        5101 Violence
        5102 Extremism
        5103 Discrimination
        5104 Dishonesty
        5105 Neglect
        5106 Unlawfulness
        5107 Irresponsibility
    5300 Ethics high
        5301 Inclusiveness
        5302 Honesty
        5303 Compassion
        5304 Commitment
        5305 Lawfulness
        5306 Solidarity
6000 Capability
    6100 Capability low
        6101 Lack of intelligence
        6102 Inexperience
        6103 Incompetence
    6200 Capability fair
        6201 Rationality
    6300 Capability high
        6301 Smartness
        6302 Creativity
        6303 Competence
7000 Moderation
    7100 Moderation low
        7101 Dissoluteness
        7102 Gluttony
        7103 Materialism
        7104 Addiction
    7200 Moderation fair
        7201 Healthy lifestyle
    7300 Moderation high
        7301 Self-restraint

Categories at the first and the second level of the hierarchy work as super-groups and sub-groups.
Only the categories at the third level of the hierarchy—the leaf categories—are predicted, but the full path starting from the super-group is returned as an additional property for each category.

Warning

Grouping is purely ontological. The names of the groups have no positive or negative connotations: in particular, the high, fair and low qualifications in sub-groups are not to be interpreted as a moral judgment, but as the intensity with which encompassed behavioral traits represent the super-group.

Output structure

The model output has the same structure as any other model and is affected by the functional properties of the workflow block.
The peculiar part of the output is the result of categorization, i.e. the categories array.

As stated above, the categories that can be returned in output—the recognized behavioral traits—are only those at the 3rd level of the hierarchy, the "leaves" of the tree.
The 1st and 2nd level categories are used to group the others. The information of the group and sub-group a behavioral trait belongs to is available in output in the hierarchy property, which represents the full path of the output category inside the tree.
For example:

...
"frequency": 75.25,
"hierarchy": [
    "Moderation",
    "Moderation low",
    "Gluttony"
],
"id": "7102",
"label": "Gluttony"
...

Example

Considering the following text:

Research commissioned by Newsround has found that 10-12 year-olds feel worried and pressured about looking good in the photos they share on social media. The majority take at least four selfies before choosing one to share, and three quarters say they edit photos before posting them. But what's shaping their attitudes?

the categorization is:

"categories": [
        {
            "frequency": 0,
            "hierarchy": [
                "Action",
                "Action high",
                "Initiative"
            ],
            "id": "2301",
            "label": "Initiative",
            "namespace": "behavioral_traits_en",
            "positions": [
                {
                    "end": 209,
                    "start": 201
                }
            ],
            "score": 10,
            "winner": true
        },
        {
            "frequency": 0,
            "hierarchy": [
                "Openness",
                "Openness low",
                "Apprehension"
            ],
            "id": "3103",
            "label": "Apprehension",
            "namespace": "behavioral_traits_en",
            "positions": [
                {
                    "end": 78,
                    "start": 71
                },
                {
                    "end": 98,
                    "start": 93
                }
            ],
            "score": 40,
            "winner": true
        }
]