Intellego

From Knoesis wiki
Revision as of 10:08, 20 June 2011 by Cory (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Ontology of Perception: IntellegO

Intellego (Greek: "to perceive")

Today, many sensor networks and their applications employ a brute force approach to collecting and analyzing sensor data. Such an approach often wastes valuable energy and computational resources by unnecessarily tasking sensors and generating observations of minimal use. People, on the other hand, have evolved sophisticated mechanisms to efficiently perceive their environment. One such mechanism includes the use of background knowledge to determine what aspects of the environment to focus our attention. In this project, we develop an ontology of perception, IntellegO, that may be used to more efficiently convert observations into perceptions. IntellegO is derived from cognitive theory, encoded in set-theory, and provides a formal semantics of machine perception.


Formal Specification

The formal specification of IntellegO is encoded in set-theory; which provides a notation that is unambiguous, well-established, and suitably expressive.

Intellego-in-set-theory.png


Ontologies and Knowledge Bases

The implementation of IntellegO used in our evaluations utilizes and integrates a suite of ontologies and knowledge bases.


Focus Evaluation

Between April 1st and April 6th of 2003, a major blizzard hit the state of Nevada. Environmental data within the surrounding area was collected by weather-stations, encoded as RDF, and made accessible on the Web. This data has been converted to RDF and is accessible as Linked Data (ref:linked sensor data). For every two hour interval from April 1st through April 6th of 2003, and for each observer within a 400 mile radius of the blizzard, we execute the perception-cycle and generate a perceptual-theory. For each execution of the perception-cycle, the observer is a weather-station and the resulting perceptual-theory contains member entities representing the weather event occurring at that time and location (of the weather station). After each execution, the resultant perceptual-theory is checked for correctness and the total number of percepts, in the set of percepts, is counted. Below, we show some statistics and trends, and provide the datasets generated by this evaluation.


Percepts Generated during Evaluation: # and %

(p = precipitation, t = temperature, w = wind speed)

25 miles (17 observers)

25-miles.PNG

50 miles (70 observers)

50-miles.PNG

100 miles (170 observers)

100-miles.PNG

200 miles (373 observers)

200-miles.PNG

400 miles (516 observers)

400-miles.PNG


Percepts Generated during Evaluation: Trends

Percepts (Observed Discriminating Qualities)

Trend-percepts.PNG

Extraneous Qualities (Not Observed)

Trend-extraneous.PNG


Perceptual-Theories Generated during Evaluation

Trend-theories.PNG


Evaluation Datasets

The data generated during the evaluation was annotated used a slightly older version of Intellego (which can be found here).