Difference between revisions of "ISWC2010 Evaluation"

From Knoesis wiki
Jump to: navigation, search
(→‎System Architecture)
 
(37 intermediate revisions by 3 users not shown)
Line 1: Line 1:
−
==Linked Sensor Data==
+
==Ontology of Perception: A Semantic Web Approach to Enhance Machine Perception==
−
LinkedSensorData is an RDF dataset containing expressive descriptions of ~20,000 weather stations in the United States. The data originated at MesoWest, a project within the Department of Meterology at the University of Utah that has been aggregating weather data since 2002.[http://mesowest.utah.edu/index.html] On average, there are about five sensors per weather station measuring phenomena such as temperature, visibility, precipitation, pressure, wind speed, humidity, etc. In addition to location attributes such as latitude, longitude, and elevation, there are also links to locations in Geonames that are near each weather station. This sensors description dataset is now part of the LOD.
+
Today, many sensor networks and their applications employ a brute force approach to collecting and analyzing sensor data, and ignore the semantics inherent in the environmental data. Such an approach often wastes valuable resources – including both 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. Such mechanisms include the use of background knowledge to determine what aspects of the environment to focus and a strong interdependent relationship between our ability to observe and perceive. In this paper, we develop an ontology of perception – derived from cognitive theory – that may be used to more efficiently convert observations into perceptions. We evaluate this approach by collecting and analyzing observations of weather conditions, and show up to 50% reduction in the number of observations necessary for analysis.
  
−
==Active Perception System [APS]==
 
−
Active Perception System is a system that is inspired by human perception. This system emulates the human perception cycle and also showcases how a system can take advantage of human perception model in order to achieve efficient use of sensors in order to get a situational awareness with least effort. Effort here refers to tasking sensors to observe some phenomenon and is quantified by their power consumption.
 
  
 
==System Architecture==
 
==System Architecture==
−
APS system architecture is as shown.
 
−
[[Image:perception_cycle.jpg|none|thumb|500px| Figure 1. Perception Cycle.]]
 
  
−
[[Image:perception_process.jpg|none|500px|alt = Figure 2. Perception Process.]]
+
<strong>Perception Cycle</strong>
 +
[[Image:perception_cycle_new.jpg|none|thumb|500px| Figure 1. Perception Cycle.]]
  
−
[[Image:observation_process.jpg|none|300px|alt = Figure 3. Observation Process.]]
+
<strong>Perception Process</strong>
 +
[[Image:perception_process_new.jpg|none|thumb|500px| Figure 2. Perception Process.]]
  
−
==Evaluation==
+
<strong>Observation Process</strong>
 +
[[Image:observation_process_new.jpg|none|thumb|300px| Figure 3. Observation Process.]]
  
−
==Data Set==
 
−
We use the [http://wiki.knoesis.org/index.php/SSW_Datasets observation dataset] for running the perception cycle.
 
  
−
==Ontologies==
+
==Ontologies and Knowledge Bases==
 +
 
 +
<strong>Weather Background Knowledge</strong>
 +
[[Image:bipartite_graph_black.jpg|none|thumb|600px| Figure 4. ]]
 +
* [http://sonicbanana.cs.wright.edu/activeperception/ont/weather.owl Weather Background Knowledge]
 +
* [http://wiki.knoesis.org/index.php/SSW_Datasets Observation Knowledge Base]
 +
* [http://sonicbanana.cs.wright.edu/activeperception/ont/perception.owl Perception Ontology]
 +
* Perception Datasets:
 +
**[http://sonicbanana.cs.wright.edu/activeperception/all_active_25miles.zip 25 miles (17 observers)]
 +
**[http://sonicbanana.cs.wright.edu/activeperception/all_active_50miles.zip 50 miles (70 observers)]
 +
**[http://sonicbanana.cs.wright.edu/activeperception/all_active_100miles.zip 100 miles (170 observers)]
 +
**[http://sonicbanana.cs.wright.edu/activeperception/all_active_200miles.zip 200 miles (373 observers)]
 +
**[http://sonicbanana.cs.wright.edu/activeperception/all_active_400miles.zip 400 miles (516 observers)]
  
−
The perception ontologies for various radii are given in this section.
 
  
−
*[http://sonicbanana.cs.wright.edu/activeperception/all_active_25miles.zip 25 Miles]
 
−
*[http://sonicbanana.cs.wright.edu/activeperception/all_active_50miles.zip 50 Miles]
 
−
*[http://sonicbanana.cs.wright.edu/activeperception/all_active_100miles.zip 100 Miles]
 
  
 
==Statistics==
 
==Statistics==
 +
 +
====Percepts Generated during Evaluation: # and %====
 +
(p = precipitation, t = temperature, w = wind speed)
 +
 +
 +
<strong>25 miles (17 observers)</strong>
 +
[[Image:25miles.jpg|none|thumb|600px| Figure 5. ]]
 +
<strong>50 miles (70 observers)</strong>
 +
[[Image:50miles.jpg|none|thumb|600px| Figure 6. ]]
 +
<strong>100 miles (170 observers)</strong>
 +
[[Image:100miles.jpg|none|thumb|600px| Figure 7. ]]
 +
<strong>200 miles (373 observers)</strong>
 +
[[Image:200miles.jpg|none|thumb|600px| Figure 8. ]]
 +
<strong>400 miles (516 observers></strong>
 +
[[Image:400miles.jpg|none|thumb|600px| Figure 9. ]]
 +
 +
 +
====Percepts Generated during Evaluation: Trends====
 +
 +
<strong>Grounded Percepts</strong>
 +
[[Image:trend_grounded.jpg|none|thumb|600px| Figure 10. ]]
 +
<strong>Extraneous Percepts</strong>
 +
[[Image:trend_extraneous.jpg|none|thumb|600px| Figure 11. ]]
 +
 +
 +
====Theories Generated during Evaluation====
 +
[[Image:trend_theory.jpg|none|thumb|600px| Figure 12. ]]

Latest revision as of 16:31, 28 June 2010

Ontology of Perception: A Semantic Web Approach to Enhance Machine Perception

Today, many sensor networks and their applications employ a brute force approach to collecting and analyzing sensor data, and ignore the semantics inherent in the environmental data. Such an approach often wastes valuable resources – including both 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. Such mechanisms include the use of background knowledge to determine what aspects of the environment to focus and a strong interdependent relationship between our ability to observe and perceive. In this paper, we develop an ontology of perception – derived from cognitive theory – that may be used to more efficiently convert observations into perceptions. We evaluate this approach by collecting and analyzing observations of weather conditions, and show up to 50% reduction in the number of observations necessary for analysis.


System Architecture

Perception Cycle

Figure 1. Perception Cycle.

Perception Process

Figure 2. Perception Process.

Observation Process

Figure 3. Observation Process.


Ontologies and Knowledge Bases

Weather Background Knowledge

Figure 4.


Statistics

Percepts Generated during Evaluation: # and %

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


25 miles (17 observers)

Figure 5.

50 miles (70 observers)

Figure 6.

100 miles (170 observers)

Figure 7.

200 miles (373 observers)

Figure 8.

400 miles (516 observers>

Figure 9.


Percepts Generated during Evaluation: Trends

Grounded Percepts

Figure 10.

Extraneous Percepts

Figure 11.


Theories Generated during Evaluation

Figure 12.
Retrieved from "https://wiki.aiisc.ai/index.php?title=ISWC2010_Evaluation&oldid=2171"