Semester

Summer

Date of Graduation

2006

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Industrial and Managements Systems Engineering

Committee Chair

Rashpal Ahluwalia

Abstract

The issues that are to be considered while selecting data analysis methods are data type and expected solution type. Incorrect selection of either can lead to incomplete solution leading user taking an uninformed decision. The software developed for this thesis (SANE - Statistical And NEural network data analysis) is aimed at enabling a user to use advanced data analysis techniques to handle a given research issue. An effective navigation system is designed for user to select the analysis method by responding to a set of questions. SANE provides web based data analysis using statistical and neural networks methods. The statistical analysis module has methods for finding a relationship between variables, predicting group membership and finding group differences. The neural net module has back propagation and cascade correlation algorithms. Users can apply different methods on same dataset and compare the results. This software is implemented in ASP.NET2.0 with backend in SQL2005.

Share

COinS