ONT Re: Inquiry Driven Systems
¤~~~~~~~~~¤~~~~~~~~~¤~DISSERTATION~¤~~~~~~~~~¤~~~~~~~~~¤
| Document History
|
| Subject: Inquiry Driven Systems
| Contact: Jon Awbrey <jawbrey@oakland.edu>
| Version: Draft 8.4
| Created: 23 Jun 1996
| Revised: 16 Aug 2001
| Advisor: M.A. Zohdy
| Faculty: Lipman, Mili, Windeknecht
| Setting: Oakland University, Rochester, Michigan
| Excerpt: 1.1.2.1 The Paradigmatic & Process-Analytic Phase
Inquiry Driven Systems: An Inquiry Into Inquiry
1. Research Proposal
1.1 Outline of the Project: Inquiry Driven Systems
1.1.1 Problem
1.1.2 Method
1.1.2.1 The Paradigmatic & Process-Analytic Phase.
In this phase I describe the performance and competence of intelligent
agents in terms of various formal systems. For aspects of an inquiry
process that affect its dynamic or temporal performance I will typically
use representations modeled on finite automata and differential systems.
For aspects of an inquiry faculty that reflect its formal or symbolic
competence I will commonly use representations like formal grammars,
logical calculi, constraint-based axiom systems, and rule-based
theories in association with different proof styles.
Paradigm. Generic example that reflects significant properties
of a target class of phenomena, often derived from
a tradition of study.
Analysis. Effective analysis of concepts, capacities, structures,
and functions in terms of fundamental operations and
computable functions.
Work in this phase typically proceeds according to the following recipe.
1. Focus on a problematic phenomenon. This is a generic property or
process that attracts one's interest, like intelligence or inquiry.
2. Gather under consideration significant examples of concrete systems
or agents that exhibit the property or process in question.
3. Reflect on their common properties in a search for less obvious traits
that might explain their more surprising features.
4. Check these accounts of the phenomenon in one of several ways.
For example, one might (a) search out other systems or situations
in nature that manifest the critical traits, or (b) implement the
putative traits in computer simulations. If these hypothesized traits
generate (give rise to, provide a basis for) the phenomenon of interest,
either in nature or on the computer, then one has reason to consider them
further as possible explanations.
The last option of the last step already overlaps with the synthetic phase of
work. Viewing this procedure within the frame of experimental research, it is
important to recognize that computer programs can fill the role of hypotheses,
testable (defeasible or falsifiable) construals of how a process is actually,
might be possibly, or ought to be optimally carried out.
¤~~~~~~~~~¤~~~~~~~~~¤~NOITATRESSID~¤~~~~~~~~~¤~~~~~~~~~¤