Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

ONT Re: Sequential Interactions Generating Hypotheses




o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o

SIGH.  Note 2

o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o

| Since 1987, I have been arguing that no ontology or
| knowledge base can ever be adequate unless it comes
| to grips with what I have called the "knowledge soup" --
| the loosely organized, semi-structured mix of whatever
| people have in their heads.
|
| John Sowa
| http://suo.ieee.org/email/msg10558.html

1.  Sequential Interaction Data

I will try to explain why this dataset is interesting to me,
and why it comes to mind whenever we turn to discussing the
associative "prima materia" or semiotic "massa confusa" out
of which our more rational knowledge precipitates, which is
one of the things that I take John Sowa to be talking about
under the label of "knowledge soup".

The general type of data that we are dealing with here
is a timed sequence of categorical or qualitative codes.
I think of it a sequence of "words" or "sentences" from
a very simple formal language, and the game afoot is to
discover the grammar of the discourse, to find whatever
patterns may be found in the data, whether they rule it
in the manner of an absolute law or a likely constraint.

2.  Family Interaction Dataset

A "episode" is coded as a 7-tuple x = <x_0, x_1, x_2, x_3, x_4, x_5, x_6>.

The components x_0, ..., x_5 are categorical codes
that are described in the codebook.

The component x_6 is a non-negative integer code
that records the elapsed time in seconds at which
the episode ends.

Here is an initial segment of an individual data file,
a set of records from a single session of observation:

| Family Interaction Data
|
|  0   -00100     0
|  0    99999     0
|  2   -00302     2
|  2    31213     5
|  2    11233     8
|  2    34213    10
|  2    33213    18
|  2    10133    19
|  2    14233    22
|  2    30113    24
|  2    99999    29
|  2    31113    40
|  2    11233    43
|  2    31212    46
|  2    11232    51
|  2    31213    54
|  2    11232    56
|  2    99999    61
| ...    ...    ...

Here is the meaning of the categorical data codes:

| Family Interaction Codes
|
| x_0 is a code denoting the overall session type:
|
|    "1" means "work", a task-structured session
|    "2" means "play", a free-interaction session
|
| x_1 and x_4 are codes that record the two family members
| that are present during the session, according to the
| following scheme:
|
| f#_family_member
|  f1_child                       "1" means "child"
|  f2_father                      "2" means "father"
|  f3_mother                      "3" means "mother"
|  f4_older_brother               etc.
|  f5_older_sister
|  f6_younger_brother
|  f7_younger_sister
|  f8_household_pets
|  f9_multiple_recipients
|  f0_objects
|
| x_2 and x_3 are codes for the "content type" and "content quality",
| respectively.  Together they form a composite code that denotes
| a signed category of activity according to the following scheme,
| where x_2 by itself denotes the "content type" and x_3 denotes
| the "logical quality" of the episode:
|
| c##_content
|  c1#_conversation               "1" means "conversation"
|   c11_positive_verbal
|   c12_talk
|   c13_negative_verbal
|  c2#_affiliate/distance         "2" means "affiliate/distance"
|   c21_endearment
|   c22_tease
|   c23_verbal_attack
|  c3#_clear_directive            "3" means "clear directive"
|   c31_request
|   c32_command
|   c33_coerce
|  c4#_ambiguous_directive        "4" means "ambiguous directive"
|   c41_request_ambiguous
|   c42_command_ambiguous
|   c43_coerce_ambiguous
|  c5#_response_to_directive      "5" means "response to directive"
|   c51_agree
|   c52
|   c53_refuse
|  c6#_vocal_behavior             "6" means "vocal behavior"
|   c61
|   c62_vocal
|   c63
|  c7#_nonverbal_behavior         "7" means "nonverbal behavior"
|   c71_positive_nonverbal
|   c72_neutral_nonverbal
|   c73_negative_nonverbal
|  c8#_low_grade_physical_contact       "8" = "low grade
|   c81_touch                                  physical contact"
|   c82
|   c83_physical_aggression
|  c9#_pronounced_physical_interaction  "9" = "pronounced
|   c91_hold                                   physical interaction"
|   c92_physical_interact
|   c93_physical_attack
|  c0#_compliance_behavior              "0" = "compliance behavior"
|   c01_comply
|   c02
|   c03_noncomply
|
| x_5 is a code that records the "valence"
| or the "affective quality" of the episode
| according to the following scheme:
|
| v#_valence
|  v1_exuberant_affect            "1" means "exuberant affect"
|  v2_positive_affect             "2" means "positive affect"
|  v3_neutral_affect              "3" means "neutral affect"
|  v4_negative_affect             "4" means "negative affect"
|  v5_extreme_negative_affect     "5" means "extreme negative affect"
|
| x_6 is a non-negative integer giving the elapsed time in seconds
|     at the end of the phase of activity coded in x_1 through x_5.
|
| Codes outside the above mentioned ranges indicate things like:
| "beginning of session", "pause in activity", "end of session".

Jon Awbrey

o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o