Sherri Harms, Chair & Professor
Computer Science & Information Systems
Office: Otto Olsen 116
Course web pages available through the University of Nebraska-Kearney Blackboard Intranet.
This study will focus on developing a data mining and knowledge discovery tool to enable the prediction of drought conditions and assess the consequential landscape and vegetation response at regional scales based on ocean-atmosphere- land interactions and their relationships with drought. The objectives of the project are to: 1) investigate the relationships between drought and oceanic parameters over the central U.S. in order to identify triggering mechanisms for the onset, continuance, and end of drought at regional, state, and local levels; 2) build predictive models to assess and predict drought using an integrated database that includes satellite, climate, oceanic, and biophysical data such as available soil water-holding capacity and ecological type; and 3) evaluate the prediction model using crop yields and vegetation indices to identify and predict the impacts of drought on agriculture in the central U.S., and perform an evaluation using listening sessions to get feedback on the model outputs and improve the dissemination of user-tailored products. The benefits of the project include improving mitigation plans to reduce drought impacts over the central U.S.; the development of a science-based framework to model droughts; the creation of a knowledge-based decision making tool; and a greater understanding of the relationships among remote-sensing, climatic, oceanic, and biophysical data to enable improved monitoring and prediction of drought at greater spatial and temporal resolution over the central United States.
The negative consequences of the brain-drain of technologically skilled employees and businesses from the Midwest areas of the United States has resulted in technologically underserved rural communities and an increasingly aging, unskilled population. The proposed research will assess area businesses in four rural towns in Nebraska to determine the information technology related services, materials, and products that businesses are outsourcing beyond the local geographic area and beyond the state of Nebraska. Using survey research administered online through Opinio software, this study will identify rural area information technology resources and needs, as well as examine local social and economic relationships within the community and identify which local businesses have the potential to provide services to global businesses. Overall, the results of this study will be used to develop and enhance economic strategies for rural Nebraska communities.
We developed web tools that take an interdisciplinary approach to plant growth stages. The research cluster drew upon expertise in computer modeling, data mining, geographic information science, and bioinformatics coupled with experts in crop science, genetics, range ecology, climate change, and genotype x environment interactions. Website: http://phenology.unk.edu
We developed a suite of spatio-temporal tools and analysis models, backed by rigorous research, which provides a holistic look at the problem of agricultural risk assessment and exposure analysis. This project incorporates traditional geostatistical methods, geographic information systems (GIS), and data mining techniques in a framework that builds associations between oceanic parameters and major crop production regions (agroecozones) along with agricultural practices, climate and soils characteristics. The tools are linked together in approach by means of web-based and client-server delivery system designed to make them accessible and effective in solving critical problems of production risk in the agricultural sector. Most of our tools are based on research and prototype tools funded, in part, by the NSF Digital Government grant and a prior Nebraska Research Initiative grant. I developed the knowledge discovery design and was responsible for integrating the spatial-temporal knowledge discovery techniques into the information layer of the GDSS. We expanded the application of previous data mining efforts to include more data sets with longer historical patterns, and interpolating the results between weather stations.
National Science Foundation (NSF) Research Experiences for Undergraduates (REU): Knowledge Discovery Based on Geographical Regions, Supplement to the Digital Government grant listed below. $14,050 to UNK, funded Summer 2004, S. Goddard and Sherri Harms, PIs.
National Science Foundation (NSF) Research Experiences for Undergraduates (REU): Knowledge Discovery: From Prototype to Decision Support, Supplement to the Digital Government grant listed below.$14,050 to UNK, funded Summer 2003, S. Goddard and S. Harms, PIs.
National Science Foundation (NSF) Digital Government Initiative (DGI): A Geospatial Decision Support System for Drought Risk Management, $1,000,000 3 year grant awarded Spring 2001 to University of Nebraska-Lincoln, S. E. Reichenbach, lead PI ($40,000 to UNK)
1999-present Association for Computing Machinery (ACM)
1999-present Special Interest Group on Knowledge Discovery and Data (SIGKDD) of the ACM
2004-present Phi Kappa Phi National Honor Society, UNK Chapter
2004-present UNK Graduate Faculty
2003-present ACM Symposium on Applied Computing (SAC) data mining track program committee