DSpace Community: Department of Food and Resource Economics

Modeling Nitrogen Loading Rate to Delaware Lakes Using Regression and Neural Networks

Sun, 29 Dec 2002 22:58:59 GMT

Title: Modeling Nitrogen Loading Rate to Delaware Lakes Using Regression and Neural Networks

Authors: Sudhakar, Prachi; Krishnan, Palaniappa; Bernard, John C.; Ritter, William F.

Abstract: The objective of this research was to predict the nitrogen-loading rate to Delaware lakes and streams using regression analysis and neural networks. Both models relate nitrogen-loading rate to cropland, soil type and presence of broiler production. Dummy variables were used to represent soil type and the presence of broiler production at a watershed. Data collected by Ritter & Harris (1984) was used in this research.

Modeling Nitrate Concentration in Ground Water Using Regression and Neural Networks

Sun, 29 Dec 2002 22:58:59 GMT

Title: Modeling Nitrate Concentration in Ground Water Using Regression and Neural Networks

Authors: Ramasamy, Nacha; Krishnan, Palaniappa; Bernard, John C.; Ritter, William F.

Abstract: Nitrate concentration in ground water is a major problem in specific agricultural areas. Using regression and neural networks, this study models nitrate concentration in ground water as a function of iron concentration in ground water, season and distance of the well from a poultry house. Results from both techniques are comparable and show that the distance of the well from a poultry house has a significant effect on nitrate concentration in groundwater.

Predictive Time Model of an Anglia Autoflow Mechanical Chicken Catching System

Sun, 28 Sep 2003 22:58:59 GMT

Title: Predictive Time Model of an Anglia Autoflow Mechanical Chicken Catching System

Authors: Ramasamy, Saravanan; Benson, Eric R.; Bernard, John C.; Van Wicklen, Garrett L.

Abstract: In this project, a predictive time model was developed for an Anglia Autoflow mechanical chicken catching system. At the completion of poultry growout, hand labor is currently used to collect the birds from the house, although some integrators are beginning to incorporate mechanical catching equipment. Several regression models were investigated with the objective of predicting the time taken to catch the chicken. A regression model relating distance to total time (sum of packing time, catching time, movement to catching and movement to packing) provided the best performance. The model was based on data collected from poultry farms on the Delmarva Peninsula during a six-month period. Statistical Analysis System (SAS) and NeuroShell Easy Predictor were used to build the regression and neural network models respectively. Model adequacy was established by both visual inspection and statistical techniques. The models were validated with experimental results not incorporated into the initial model.

Analysis of Christina School District's 2003 DSTP Performance

Mon, 29 Dec 2003 22:58:59 GMT

Title: Analysis of Christina School District's 2003 DSTP Performance

Authors: Mackenzie, John

Abstract: This report summarizes the performance of the Christina School District (CSD) on the 2003 Delaware State Testing Program (DSTP). The 2003 test data represent the final set of results attributable to CSD's former leadership team, which was replaced in July of 2003. The DSTP tests all public school 3rd, 5th, 8th and 10th graders in three areas: reading, math and writing. Despite its adequate resources, CSD has generally lagged behind most other school districts in Delaware in student DSTP performance. There is a persistent drop-off in student performance between 3rd and 5th grades, due in part to a significant exodus of high-performing 5th grade students to non-CSD schools, and there is little or no recovery in student performance levels between the 5th and 10th grade tests. The 2003 results identify schools and curriculum areas in particular need of improvement.