Not part of three month project

1
Simple and Advanced Orb Architecture
OntologyStream architecture

• The first part of this presentation (Slides 2
  12) can be demonstrated within six weeks and
  refined within the time frame of 3 and ½ months
• 50 time in May 2004 and 100 time June
  August 2004
• Dr. Prueitt and programming support from
  Nathan Einwechter
• The Simple Architecture develops Orbs using
  human in the loop activities to produce and use
  simple Orb resources
• The Advanced Architecture (slides 14 24)
  develops Orbs using human information production
  techniques plus Orb based data mining processes.
  Advanced architecture is an additional 6 months
  to one year effort.

Not part of three month project
Advanced Architectures
2
Simple architecture
Human Abduction about event Chemistry supports
Anticipation
Event Chemistry Repository
Encoding Engine Orb Visualization
Data
Real time analysis
Real time or legacy
Human knowledge of sub-structure supports
mutual induction
categoricalAbstraction sub-structure Repository
Diagram from Prueitt (2004)
Simple Architecture depends strongly on human
knowledge
3
Instrumentation/measurement

• Perl code simple and independent from Orb code
• Rules can be redeveloped by subject matter expert
  easily
• Output always in the form of lt a, r, b gt

Manual selection of measurement
Event parser
SLIP Analysis and Visualization
Knowledge Operating System
mutual induction
Generalized n-gram looks for conditional
co-occurrence of sub-structure in the data
abduction
Compact encoding of ordered triples in hash
tables
Orbs
Events of interest with selected sub-structure
content
Work product re-use
Diagram from Prueitt (2004)
Simple Architecture instrumentation and
measurement
4
Existing research tool capability
Can be demonstrated in prototype in May 2004
Orb technologies provide concept and metaconcept
metadata
Confirmed/refined Subject Matter Indictors
Selection from Reified Indicators
Editor, subject matter expert
Diagram from Prueitt (2004)
Simple Architecture existing research tool
capability
5
Confirmed/refined Subject Matter Indictors
Selection from Reified Indicators
Full-cycle event profile retrieval
retrieval engine
Knowledge work / feed-back
Compact encoding of ordered triples in hash
tables
Orbs
Case histories with selected subject matter
content
Diagram from Prueitt (2004)
Simple Architecture Human reification of Orb
structures
6
Use of ASCII encoding of event structure allows
event structure mining functions base on simple
general framework instrumentation and encoding
Text analysis technologies provide ontology
services
Diagram from Prueitt (2004)
Simple Architecture Encoding innovations
generalized from text analysis
7
Patent pending technologies provide knowledge
representational services, based on ASCII word
descriptors and co-occurrence patterns
Diagram from Prueitt (2004)
Simple Architecture encoding innovations
8
of relevant event-profiles retrieved
Recall
total of relevant event-profiles in collection
of relevant event-profiles retrieved
Precision
Precision
total of event-profiles retrieved
Recall
Diagram from Prueitt (2004)
Simple Architecture precision recall
9
Search Characteristics

• If you enter a descriptor,
• and that descriptor is in an Orb or database,
• you get all data linked to that descriptor,
• regardless of whether or not that descriptor is
  relevant

• High recall, low precision
• Time wasted with irrelevant data
• Relevant items may be retrieved but overlooked

Precision
Recall
Diagram from Prueitt (2004)
Simple Architecture precision recall
10
Search Characteristics (cont.)

• If highly relevant data is in the Orb or database
• but none of the descriptors you enter are really
  relevant,
• then the data needed will not be retrieved

• High precision, low recall
• Greater chance for error
• Inconsistent results
• Time wasted through redundant effort

Precision
Recall
Diagram from Prueitt (2004)
Simple Architecture precision recall
11
Question

• How do we bend the curve so we get
• more of what we really need
• and less of what we dont need?

Diagram from Prueitt (2004)
Simple Architecture precision recall
12
Answer

• The Orbs initially have local structure but no
  global structure and yet are easy to organize
  into a specific global structure
• Categorical Abstraction creates similarity
  relationships which can be modified in real time
  using mutual induction (human reification through
  cognitive priming and visualization)
• Event Chemistry stores past work product into
  things-to-try, where the resulting organization
  expresses real science regarding event structure
  and sub-structure to function relationships
• Now the retrieval task has by-passed schema
  structure imposed in classical database models
  and allows real time intuition to play an
  un-encumbered role

Diagram from Prueitt (2004)
Simple Architecture basic intuition behind Orbs
13
Advanced Architecture developed as part of
OntologyStream work on Total Information
Awareness architecture

• The foundation of this work is established in
  the Simple Architecture, which gives a stepping
  stone to the Advanced work
• The Simple Architecture develops Orbs using
  human in the loop activities
• The Advanced Architecture develops Orbs using
  human information production techniques plus Orb
  based data mining processes

Not part of three month project
Advanced Architectures
14
Subject-matter Experts
Topic Maps
Ontology Lens
Schema Logics
Presentation Layer
DOF
Transaction Layer
Event parsing
Parsing, tagging, routing, categorizing,
clustering
Orbs
RAM memory
Legend Orb Ontology referential base DOF
Differential Ontology Framework, including
stochastic, latent semantic indexing and ontology
services
Data encoding
Hash tables
System files
Not part of three month project
Advanced Architectures Complete architecture
15
Inter-Role Collaboration using Ontology
Role and Event Specific View
Knowledge Worker
Views
Knowledge Repository
Periodic Update
Synchronous Collaboration
Knowledge Worker
Views
Not part of three month project
Advanced Architectures Distributed
Collaboration
16
Production of Event Measurement Metrics
Implicit Ontology
T ground (B 1)
T ground (B 2)
T ground (B 3)
Explicit Ontology
d e C response
T ground (B 4)
T ground (B 5)
For each response, d, the implicit ontology
produces a set of metrics, mk , and these
metrics are used as the atoms of a logic. These
atoms are used to produce an explicit ontology.
The logic is then equipped with a set of
inference rules. Evaluations rules are then
added to produce an inference about the response
set.
Diagram from Prueitt (2002)
Advanced Architecture event metrics
17
Differential Ontology Framework

• By the expression Differential Ontology we mean
  the interchange of structural information between
  Implicit (machine-based) Ontology and Explicit
  (machine-based) Ontology
• by Implicit Ontology the variations of latent
  semantic indexing. These are continuum
  mathematics with only partial representation on
  the computer.
• by Explicit Ontology we mean an bag of ordered
  triples lt a , r, b gt , where a and b are
  locations and r is a relational type, organized
  into a graph structure, accompanied by first
  order predicate logic. This is a discrete
  formalism.

Implicit Ontology
Explicit Ontology
Diagram from Prueitt (2002)
Advanced Architecture differential ontology
framework
18
Tri-level Architecture bases for abduction of
event function and mutual induction from
substructure memory with human in the loop
The Tri-level architecture is based on the study
of natural systems that exist as transient
stabilities far from equilibrium. The most basic
element of this study is the Process Compartment
Hypothesis (PCH) that makes the observation that
systems come into being, have a stable period
(of autopoiesis) and then collapse. Human
cognition is modeled in exactly the same way.
Human mental events are modeled as the
aggregation of elements of memory shaped by
anticipation.
The tri-level architecture for machine
intelligence is developed to reflect the PCH. A
set of basic event atoms are developed through
observation and human analysis. Event structures
are then expressed using these atoms, and only
these atoms, and over time a theory of event
chemistry is developed and reified.
Diagrams from Prueitt (1996)
Advanced Architecture Tri-level Architecture
19
cA/eC categoricalAbstraction and
eventChemistry
Neuroscience informs us that the physical process
that brings the human experience of the past to
the present moment involves three stages. 1)
First, measured states of the world are parceled
into substructural categories. 2) An
accommodation process organizes substructural
categories as a by-product of learning. 3)
Finally, the substructural elements are evoked by
the properties of real time stimulus to produce
an emergent composition in which the memory is
mixed with anticipation. Each of these three
processes involves the emergence of attractor
points in physically distinct organizational
strata. The study of Stratified Complexity
appeals first to foundational work in quantum
mechanics and then to disciplines such as
cultural anthropology and social-biology.
categoricalAbstraction (cA) is the measurement of
the invariance of data patterns using finite set
of logical atoms derived from the measurement.
eventChemistry (eC) is a theory of type that
depends on having anticipatory processes modeled
in the form of aggregation rules, where the
aggregation is of the cA logical atoms.
Diagram from Prueitt (1995)
Advanced Architecture eventChemistry from Dr.
Paul Prueitt
20
gF two examples
generalFramework (gF) theory constructs cA/eC
knowledge based on conversation with humans.
The general form of a framework is constructed
based on specific knowledge of an application
domain
We have generalized from a physical theory of
about formative processes, to a computational
architecture based on frameworks. Various forms
are conjectured to exist as part of emergent
classes, and to have a periodic table like, in
many ways, the atomic period table. The
Sowa-Ballard Framework has 18 semantic atoms.
The Zachman has 30 atoms.
Ballard/Sowa Framework
According to Alvin Toffler, knowledge will
become the central resource of the advanced
economy, and because it reduces the need for
other resources, its value will soar. (Alvin
Toffler, Power Shift, 1990). Using architectural
constructs such as the Zackman Framework, can
prepare organizations to tap their inner banks of
knowledge to improve their competitive positions
in the twenty-first-century.
Zackman Framework
Diagram from Prueitt (2002)
Advanced Architecture gF from Dr. Paul Prueitt
21
mI mutual Induction between machine memory and
human Introspection
Mutual induction occurs when cognitive priming
triggers mental events in humans. If an
incomplete pattern is presented using SLIP
visualization of topological neighborhoods, in
Orb structure, then Human-centric Information is
Produced (HIP).
SAR structure activity relationships
Diagram from Prueitt (1995)
Advanced Architecture mutual Induction from
Dr. Paul Prueitt
22
Situational Logic Construction
A latent technology transform, T ground , is used
to produce simple metrics on membership of
sub-structure in event structure from the
response collection C response in the
categories defined by the contents of the bins C
exemplar B 1 È B n2 È . . . È B q
These bins are represented in the situational
logic as the logical atoms A , from which a
specific logic is constructed. These atoms are
then endowed with a set of q real numbers that
are passed to an Inference Processor.
The set of q real numbers are computed from a
formal evaluations of the structural
relationship between logic atoms using the
Ontology Lens. Atom a ? r1 , r2 , . . . ,
rq The process of developing a situational
logic treats new data structure as axioms, and
then a process of reduction of axioms to logical
atoms occurs (Russian Semiotics). The reduction
also requires the Ontology Lens, (invented 2002
by Prueitt).
Diagram from Prueitt (2003)
Advanced Architecture situational logics from
Dr. Paul Prueitt
23

• The Ontology Len (discovered by Prueitt, 2002) is
  a structural focus instrument that is designed
  to allow non-computer scientists to specify high
  quality exemplar sets. This is done with an
  Implicit Ontology to Explicit Ontology (IO-EO)
  loop.
• When the user puts a new unit into a bin or
  removes a unit from a bin, then the IO-EO loop
  will produce a different result.
• It is the human responsibility to govern the
  IO-EO loops so that the results have the
  properties that the human wants, mostly
  independence of categories, but perhaps some
  specific (and maybe interesting) category
  entanglements.

A graphic representation of what we call a
Latent Semantics Index structural similarity
matrix. The similarity is called structural
because the exact notion of semantic similarity
is not known from this algorithmic computation by
itself. The paragraphs of a small exemplar set
(see appendix A) are ordered as labels for the
columns and rows. One would expect that a
paragraph would be structurally similar with
itself, and this is in fact what one sees as a
set of dots (representing a value of 1) down
the diagonal.
Diagram from Prueitt (2002)
Advanced Architecture ontology lens from Dr.
Paul Prueitt
24
Minimal Work Flow Production of the Explicit
Ontology
Implicit Ontology
C ground C exemplar C response
T ground (B 4)
T ground (B 5)
T ground (B 3)
T ground (B 2)
T ground (B 1)
Ontology Lens
Schema-independent data
Schema-independent data is developed from the
Ontology Lens, in the form lt a, r , b gt
Where a and b are categories defined by the
exemplar set, , and r is a measure of
relationship.
Diagram from Prueitt (2002)
Advanced Architecture work flow