Friday, October 22, 2004
Background discussions on a proposed
Anticipatory Technology Challenge Problem
White Paper on Incident Information Orb Architecture (IIOA) à
BCNGroup’s offer in response to the War on Terror
(see also (mappingSocialDiscourse))
The problems faced by the
American Administration are difficult ones.
The BCNGroup is offering, as of Thursday, October 21, 2004, a Challenge
Problem in response to a Broad Area
Announcement from ARDA.
Our technology is designed
to effectively and accurately inform us of who may be contemplating terrorism and
what their geographical location may be.
We believe that our
technology will increase the measure of fidelity between what is information,
that might be misleading or incorrect, and information that is correct. As this occurs, we expect the foreign threat
to diminish and the stress leading to fundamentalist expression, at home, to
also diminish.
In the New War, our task is
to detect and understand memetic expressions that are targeting our way of life,
without destroying our tradition for multi-culturalism and freedom of
expression. This task is difficult even
if one knows who the potential terrorists are and where the threatening
expression is coming from.
Americans cherishe multicultural
traditions. Our observation is that these traditions are memetically complex,
in fact multiculturalism might best illustrate the expression of memetic
complexes. In these expressions,
reconciliation of conflicts can occur by accepting the differences between
expressed viewpoints as being natural and acceptable. Memetic simplexes are more narrowly expressed as a truth
that cannot allow alternative viewpoints.
As a consequence any attempted reconciliation process quickly becomes a
conflict that can only be resolved by domination of one “side’ over the other
“side”.
By holding to principle,
the simplex often becomes part of a justification for aggression. When this occurs, the justification is
reinforced by a rational appeal to basic principles. The rationale is expressed in the selection of words by authors
and this selection process involves the formation of mental events. The consequences of our understanding how
mental events form leads us to the development of anticipatory technology based
on the measurement of the co-occurrence of semantic primitives defined in the
substructural ontology. The generative methodology for the
development of substructural ontology is discussed in a thread of the glass
bead game started October 21, 2004.
Many simplexes are able to
exist without leading to un-resolvable conflicts, in spite of the natural
reflex to express belief. It may be
that there is a more complex social belief that moderates the simplex. For example, the current holding back on
expression by the al Qaeda terrorist movement may be due to a radical
fundamentalism embedded in something more complex. The ability to control or moderate fundamentalism may be held
within the mainstream Islamic community.
These are only some of the
difficult issues that will be better understood by mapping of themes expressed
by social symbols in social discourse. Centers of
Excellence are being planned to engage interdisciplinary groups in
the American academic community.
We make the argument that
such mapping of the social discourse has economic value as well. Interests groups can stay in tune with
currents topics within a large distributed community. Product suppliers can harvest the social discourse and use this
information to stock inventories, understand customer feedback and
concerns. The harvest of the thematic
structure of social discourse is currently done with polls. Orb constructions visualize the topics and
themes in web based virtual communities.
In the preliminary results from Orb developed mapping of social
discourse, visual icons indicate opinions.
Our claim is that these Orb construction when visualize and subject matter
indicators, create a greater detail than polling methods achieve. The automated develop of taxonomy is
discussed in the InOrb Technologies
bead game thread.
President George W Bush has
consistently argued that the War on Terrorism is a long-term struggle between
civilized societies and a well-organized community of radical terrorists.
Principle
Observation: This source of
threat is a complex system of social activities. The development of the threat depends essentially on human
communication. The communication is required
as a means to inform the various members of existing groups. Terrors cells grow new membership from the
large communities in which the terror community exists.
The bits and pieces of
terror-community text is acquired in real time, by the many web harvesting
processes now in place (such as the system that has (hopefully) evolved form
the NSC J-39 system)
and can be used to know what the social discourse is within and between the
embedded community cells. This can be
done immediately using the Readware software.
Moreover, the fidelity of the resulting map of terror discourse
increases radically as the system is tuned to catch the nuances that
characterize terror discourse.
The ReadWare Orb
constructions are small co-occurrence maps and exist at three
levels. The instrumentation is made at
a letter 3-gram level, where the 3-grams are constructed stems. It is worth noting that Readware has not be
applied to Arabic discourse as yet, since the core team has not had the
financial resources or client supporting this application. The next level is composed of an aggregation
of these 3-grams into constructions that have been specified in a
semi-automated fashion. The
semi-automated process is to be demonstrated using 25 Aesop fables, and concept
identification tests made over the entire collection of 312 fables.
Figure 1: Readware stratification of informational bits
(Power Point at {++})
These aggregations of
letter 3-grams are placed into a concept frame. The concept frame can be realized as concept indicators with an
additional constraint reflecting sets of ontological affordances discovered
within the Readware stratified ontology and generative methodology. The concept frame acts as a concept prime,
and creates a higher-level abstraction than does a specific word level harvest
of text. This higher level of
abstraction is also being appreciated in the applied of the generalization of
the Readware-Orb generative methodology as applied to the co-occurrence
patterns found in cyber attack scenarios (Stephenson and Prueitt, current
work).