Research Experience
The University of Kanas
Depeartment of Computer Science and Electrical Engineering
2005 Research Experience for Undergraduates (REU)
Using Ensemble Learning for Detecting Data Abnormaties in Databases
Mentors: Drs. Gogineni, Tsatsoulis, and Miss. Lee
 
 

Abstract: Software engineers at the University of Kansas have developed SmartXAutofill, an intelligent data entry assistant for predicting and automating inputs for eXtensible Markup Language (XML) and other text forms based on the contents of historical documents in the same domain. SmartXAutofill utilizes an ensemble classifier, which is a collection of a number of classification algorithms where each individual internal classifier predicts the optimum value for a particular data field. As the system operates, the ensemble classifier learns which individual internal classifier works better for a particular domain and adapts to the domain without the need to develop special classifiers. The ensemble classifier has proven that it performs at least as well as the best individual internal classifier. The ensemble classifier contains a voting and weighting system for inputting values into a particular data field.

Because the existing technology can predict, suggest and automate data fields, the investigator tested whether the same technology can be used to identify incorrect data. Given existing data transmitted by sensors and other instruments, the investigator studied whether the ensemble classifier technology can identify data abnormalities and correctness in future sensor data transmission. The solution would be applied in a project funded by the National Science Foundation, Polar Radar for Ice Sheet Measurements (PRISM), using innovative sensors to measure the thickness and characteristics of the ice sheets in Greenland and Antarctica, with the goal of understanding how the ice sheets are being affected by global climate change.

PRISM sensors continuously send information that is collected and catalogued. The ensemble classifier will check the data for correctness by predicting which values should be there, and if the actual values are different, it will flag the data as possibly corrupted, and allow an operator to later study it and determine if it is correct or not. This technology will allow the PRISM intelligent systems to automatically determine the correctness of sensor and other data, and contributes to the PRISM project by adding a level of intelligence and prediction to the sensor suite.

 

The University of Kanas
Department of Computer Science and Electrical Engineering
2004 Research Experience for Undergraduates (REU)
UML Class Diagrams of PRISM Multi-Agent Subsystem Using UML Class Diagrams
Mentors: Drs. Gogineni and Tsatsoulis
 

Abstract: Hurricanes, tornados, thunderstorms, and other natural disasters can have many devastating outcomes. Global warming, the prime investigated natural disaster of the PRISM (Polar Radar for Ice Sheet Measurements) project, has a tremendous effect on the sea level rise. Scientists and researchers have theorized that the excess water is being allocated from the polar ice sheets of Greenland and Antarctica due to the long-term results of global warming; however, there are few resources to confirm the gain or loss ice.

Scientists and researchers of the PRISM project have applied their expertise on teams based on the areas of robotics, communications, intelligent systems, and radar. These areas were essential in measuring the ice thickness an determining the bedrock below the ice sheets in Greenland and Antarctica. In this research, the investigator worked with the Intelligent Systems team by learning the messaging patterns between the data producing agents and the requesting agents. The investigator also created Unified Modeling Language (UML) class diagrams of a messaging subsystem to represent the collaboration and communication between these two types of agents using eXtensible Markup Language (XML) with the Foundations for Intelligent Physical Agents (FIPA) standard codes. The class diagrams assisted scientists and researchers in planning new features for the multi-agent system.