Mathematics, computer science and statistics.
Vicechair: Walter Brehm, Keesler Air Force Base
1:00 HIGH-SCHOOL GPA AND ACT SCORES AS PREDICTORS FOR STUDENT SUCCESS FOR THE FALL OF 1999 FIRST-TIME FRESHMEN AFTER ONE YEAR OF STUDY IN THE FALL 2000 AT JACKSON SATE UNIVERSITY
Elgenaid I. [Hamadain.sup.*], Okoye Steven, Lauren Jones, Jonathan Rodarte, and Wutasha Taylor, Jackson State University, Jackson, MS 39217
Data of 437 incoming freshmen after one year of study in Fall 2000 at Jackson State University were obtained from the Office of Institutional Research. The data contained information on students' high school GPA, ACT score, and college GPA. Several SAS procedures were used to analyze the data. Results indicated that data on all three variables were normally distributed. The three variables were moderately and positively correlated with each other. Based from multiple regression analysis, high school GPA and ACT score were significant. Individual simple regression analysis confirmed these results. Although the variables were not strongly correlated with each other, high school GPA and ACT scores appeared to be good predictors of college success (GPA). Two sets of samples (61 students each) were taken randomly and systematically using SAS. Descriptive statistics, regression and correlation analysis were performed for each set. Multiple regression model indicated that high school GPA and ACT Scores are good pred ictors of College GPA. High school GPA seems to be more important in determining college success (p-value = 0.0002), than ACT score (p-value = 0.3063). Descriptive statistics generated by random sampling seem identical to the one obtained through systematic sample. Mean high school GPA was significantly higher than that of college GPA, however; college GPA was more variable than high School GPA, according to t-test.
1:20 VISUAL REPRESENTATION OF INFORMATION NEEDS IN CARTOGRAPHIC INFORMATION RETRIEVAL USING GIS TECHNOLOGY
Lixin Yu, Alcorn State University, Alcorn State, MS 39360
Library catalogs for map collections are not well developed in most libraries. The cartographic information source differs from other kinds of information in that maps are usually rectangular in shape and defined by the coordinates of the four map corners. This coordinate information is difficult for an average person to use, unless certain user interface is designed and knowledge discovery in database algorithms are implemented. System with such an interface and algorithms can perform powerful queries that an ordinary text-based information retrieval system cannot. This presentation demonstrates a prototype system--GeoMatch, which allows users to interactively define geographic areas of interest on a background map. It also allows users to define, qualitatively or quantitatively, the relationship between the user-defined area and the map coverage. Two measurements, coverage and exclusiveness, are introduced for users to precisely represent their information needs. Three librarians were interviewed to study t he feasibility of the new system. The MARC record format is also discussed to illustrate that retrospective conversion of cartographic material records from an existing library online catalog system to GeoMatch can be done automatically.
1:40 APPLICATION OF ARTIFICIAL NEURAL NETWORKS FOR THE CLASSIFICATION OF REMOTE SENSING SPECTRAL REFLECTANCE DATA OF STRESSED SOYBEAN LEAF
Abdullah [Faruque.sub.*], Raj Bahadur, and Gregory A. Carter, State University of West Georgia, Carrollton, GA 30118; Mississippi Valley State University, Itta Bena, MS 38941; and Earth System Science Office, NASA, Stennis Space Center, MS 39529
This research paper describes the application of artificial neural networks as a superior pattern recognition tool for the classification of remote sensing spectral reflectance data of stressed soybean leaves. The objective of this study funded by National Aeronautics Space Administration (NASA) at Stennis Space Center was to record and classify the spectral reflectance differences of leaf stress caused by drought, fungal disease, and lead contamination of the soil. Reflectance spectra of drought stressed, lead contaminated and fungal infected leaves were measured using GERl500 Spectroradiometer for 512 spectral bands with 1.52 nm intervals from 308 nm to 1089 nm. Multi-layer feed-forward neural network model was used to train and predict the different classes of stressed leaves from their spectral signature. Network parameters and architectures were optimized to obtain maximum network classification performance. The classification performance of neural networks was compared to K-nearest neighbor and other st atistical pattern recognition techniques. The superior classification capability of neural networks refined with an additional research can be used to monitor more precisely the signs of damaging stress due to different factors on economic crops.
2:00 A SELF-TUTORING INTERACTIVE MODULE Kanchan Manaktala, Alcorn State University, Lorman, MS 39096
This is the first of a sequence of six interactive modules in analytic geometry that I am developing under the activities of the NSF project WELCOME awarded to Mathematics Association of America. The objective of this interactive module is to create a self-tutoring resource on the concept of a point. Math Wright, a powerful mathematical authorware is used to develop this module. The module teaches and tests the plotting of points, finding distance between two points, checking co-linearity of three points, and determining the equation of a line passing through two points. The feedback is instant and the questions are generated randomly, giving the user numerous monitored examples to master the concepts.
2:40 EXPLORING MATHEMATICAL THINKING
Garfield Burke, Jr., Mississippi Valley State University, Itta Bena, MS 38941
Mathematical thinking is much more than performing arithmetical computations with speed and accuracy. It involves reasoning and problem solving. Reasoning involves activities such as deciding if an answer is correct and/or defending the result. When reasoning is part of all mathematical activities, students learn that mathematics is not a collection of arbitrary rules but a system that makes sense and can be figured out. Problem solving is much more than finding an answer to a given problem. It requires analyzing, selecting and applying an appropriate strategy or combinations of strategics to an unfamiliar problem at hand. It also requires justifying solutions and if possible, transferring or generalizing knowledge gained from a problem-solving experience to similar situations. The purpose of this paper is to introduce the reader to some of the basic concepts and techniques centered around mathematical thinking/problem-solving activities and to illustrate how these concepts and techniques may be used to solve problems.
3:10 A DYNAMIC TOOL FOR CREATING MATHEMATICAL MODELS
Ravinder Kumar, Alcom State University, Alcom State, MS 39096
The objective of this work is to develop a dynamic tool for creating mathematical models that use the technique of linearity. Math Wright, a powerful authorware which blends CAS (Computer Algebra system) with multimedia tools, is used for the purpose. In many situations data collected renders itself to modeling by linearity. Typically, this happens when parameters or some functions of the parameters are connected by direct variation. This can be done in two ways: either manually fit a line of best fit in the scatter diagram of the (modified) data or use the technique of least squares to create a line passing through the origin that fits best in the (modified) data. If the resulting line is of the form y = kx, then k is the desired constant of proportionality. This dynamic tool allows the user to create a line of best fit through the origin in both ways. There are two built in examples and three examples for further exploration. Among the examples considered are fishing derby, panic stop, weight of the heart o f birds. These examples are taken from Giordano's book on Mathematical Modeling (FR Giordano, MD Weir, and W. Fox, A First Course in Mathematical Modeling (Second Edition) Brooks/Cole, 1997.)
3:30 Divisional Business Meeting
9:00 Divisional Poster Session
MATHEMATICAL PRINCIPLES OF RSA CRYPTOGRAPHY
Tancia [Boone.sup.*] and Przemo Kranz, University of Mississippi, University, MS 38677
Cryptography is considered an art or science of secret messaging. It is a scientific method with usage of mathematical functions to encode and decode message. All messages have to be converted into a form that is only understood by the intended recipient. The two elements that make up a cryptosystem is encryption and decryption. Encryption is the process that plain text is converted to, to make the coded message (cipher text). Decryption is the process of the cipher text being converted back into original formation by the intended audience. In the world today, the usage of this coding process includes ATM machines for electrical transfers, protects tapping in cell phone calling, encodes and decodes data entry while using the Internet, etc. However, the only secure form of cryptography known today is RSA (named for the three inventers-Ronald Rivest, Adi Shamir and Leonard Adleman) Cryptography. The purpose of this system is to create a one-way function system; a system that computes a function easily but makes inverting almost impossible. Suppose p and q are large prime numbers (numbers that are divisible by 1 and itself) containing 500 digits each. The results n = pq is a 1000 digit composite number. On a typical computer the multiplication takes under a second to compute. The variables p and q are considered private keys. The person decoding the message should know the private keys, whereas the general public knows the public key, which is n. Instead of given the primes, p and q, one is given the product n. From the number assigned to n, the two prime factors are to be founded. Using trail division to factor n would take centuries to factor.
MISSISSIPPI CENTER FOR SUPERCOMPUTING RESEARCH (MCSR) USER ADVISORY GROUP
Faculty and graduate student researchers will describe their research projects that involve HPC and the use of MCSR resources. IHL faculty and graduate students, with an interest in HPC and/or MCSR facilities and services, are also invited to attend. A meeting of the MSUAG meeting will follow in the afternoon. IHL faculty and graduate students, with an interest in HPC and/or MCSR facilities and services, are also invited to attend.
PARALLEL COMPUTING PROJECTS AT UNIVERSITY OF SOUTHERN MISSISSIPPI
Luis A. Parra, Jun Xie, Huiqian Yang, and Dexuan Xie, University of Southern Mississippi, Hattiesburg, MS 39406
DEVELOPMENT OF THE PARALLEL TURBO CODE
Jenping Chen, Mississippi State University, Mississippi State, MS 39762
VALIDATION OF MODAS-NRLPOM FORECASTING SYSTEM
Germana Peggion, University of Southern Mississippi, Stennis Space Center, MS 39522
LOW-ORDER SCALING METHODS FOR DYNAMICAL ELECTRODE CORRELATION
Svein Saebo, Mississippi State University, Mississippi State, MS 39762
P4 CLUSTER PERFORMANCE MEASURING AND TUNING
Andy Yaquan Xu and Haibo Wang, University of Mississippi, University, MS 38677
PARALLEL COMPUTING WITH NWCHEM AND MPQC
Haibo Wang, University of Mississippi, University, MS 38677
USE OF PARALLEL RESOURCES AT MCSR FOR THREE-DIMENSIONAL RECONSTRUCTION OF PROTEIN STRUCTURES
Mona T. Norcum, University of Mississippi, University, MS 38677
EXPERIENCE WITH PARALLEL DATABASE SERVERS ON DISTRIBUTED COMPUTING PLATFORM
Yanhai Sun and Haibo Wang, University of Mississippi, University, MS 38677
Divisional Talks Resume
9:30 A DYNAMIC AND INTERACTIVE MODULE ON PERMUTATIONS AND COMBINATIONS
Shomari Mosi, Alcorn State University, Lorman, MS 39096
Mathwright, a mathematical authoring software, is used to create an interactive and dynamic module on the topic of permutations and combinations. It allows the user to experiment with permutations and combinations and conjecture formulas through the animations created. Permutations and combinations are useful in probabilistic considerations. Usually, students have difficulty in differentiating between permutations and combinations. This module shows through the use of animations the difference between the two. The module is interactive subject to some limitations since a computer screen cannot contain a large number of objects. The module also randomly generates examples in probability and solves them step by step.
9:50 ASYNCHRONOUS CONVERGENCE OF ITERATIVE SOLVERS
Rochelle [Jenkins.sup.*] and Joseph Kolibal, University of Southern Mississippi, Hattiesburg, MS 39406
We exam the rates of convergence associated with relaxation methods such as Jacobi and Gauss-Seidel in which the iteration is modified to allow for asynchronous convergence, i.e., in which the selection of nodes to be updated is done randomly. The purpose is to investigate the parallelization of these algorithms, and to consider techniques for improving the rates of convergence.
10:10 A METHOD FOR OVERCOMING DISCONTINUITY BETWEEN NEIGHBORING UTM ZONE GRIDS
Chad A. Steed, Naval Research Laboratory, Stennis Space Center, MS 39529
The Universal Transverse Mercator (UTM) projection is ideally suited for use with large-scale geographic data. The Naval Research Laboratory at Stennis Space Center (NRLSSC) has investigated the UTM projection in support of the Naval Air Systems Command. In the investigation of the UTM grid, the properties and construction of the UTM projection have been analyzed to identify advantages and disadvantages in the system. A major disadvantage of the system is the lack of continuity between neighboring UTM zone grids. This problem is most apparent when dealing with an area of interest that spans multiple UTM zones. To alleviate this problem, the concept of flexible zone boundaries for UTM zones has been developed and applied to two specific situations in UTM area of interest processing. The first situation involves the creation of a single UTM grid from the grids of neighboring UTM zones in a defined area of interest. The second is the extraction of native UTM zone grids from a single UTM grid that spans multiple UTM zones.
10:50 THE DESIGN AND DEVELOPMENT OF GRAPHICAL USER INTERFACES TO CONSOLIDATE MANAGEMENT TASKS ASSOCIATED WITH NAVY AIRCRAFT OPTICAL DISKS
Stephanie A. [Myrick.sup.*], Marlin L. Gendron, and Michael E. Trenchard, Naval Research Laboratory, Stennis Space Center, MS 39529
This paper presents the design and development of a series of Graphical User Interfaces (GUI) that are used to build and manage aircraft optical disk images by U.S. military aircraft mission planners, requirements officers, and aircrew. The Naval Research Laboratory at the Stennis Space Center (NRLSSC) SC) has created this product in support of the Naval Air Systems Command Hornet and Harier Programs (PMA-265 and PMA-257, respectively). The GUIs are a component of NRLSSC's Moving-Map Composer (MMC) software that performs a wide variety of mission planning and aircraft image functions including the design and build of Mission Planning System Compact Disk Images and Aircraft Optical Disk (AOD) images. These images are comprised of user-specified digitized charts, static, non-georefererenced dataframes containing emergency checklist procedures and reconnaissance photographs. MMC is an X Windows/Motif GUI and low-level C language application written by NRLSSC scientists, and implemented on a Compaq Alpha computer running Open VMS. The Navy and Marine Corps employ a large number of AODs in support of their various missions and requires a system to manage, build, and edit the images they contain. This paper will address the GUI design and development issues raised during this project and the methods that were developed for their resolution.
11:10 THE DEVELOPMENTOF SOFTWARE TOOLS FOR SCANNING PAPER CHARTS INTO GEOTIFF IMAGE FILES
Marlin L [Gendron.sup.*], Stephanie A. Myrick, and Michael Trenchard, Naval Research Laboratory, Stennis Space Center, MS 39529
This paper presents the design and development of software tools that scans paper charts, converts the charts into digital raster images, geo-references the images, and outputs Geo TIFF-compliant image files. The GeoTIFF file format is an extension of the popular TIFF raster file format and defined by a tightly controlled specification. Sets of tags (i.e., information) are added to the TIFF file format and describe all cartographic information associated with the TIFF scanned map imagery. These tags describe such information as map projection, datum and geographic location. Many standard commercial and government applications, including moving-map programs, utilize GeoTIFF files. The Naval Research Laboratory at the Stennis Space Center (NRLSSC) has created a set of software tools in support of the Naval Air Systems Command Hornet and Harrier Programs (PMA-265 and PMA257, respectively). NRLSSC scientists have written low-level C routines that geo-reference digital charts and convert them to Geo TIFFs. Graphic al User Interfaces (GUIs) written in the X-Window System/Motif language were also developed for displaying charts during the scanning process and choosing geographic control points on a scanned image. Other GUIs designed allow the user to clip or remove undesirable areas on the charts and display the final registered raster images.
11:30 MISSISSIPPI CENTER FOR SUPERCOMPUTING RESEARCH (MCSR) USER ADVISORY GROUP MEETING
David G. [Roach.sup.*] and Germana Peggion, University of Mississippi, University, MS 38677, and University of Southern Mississippi, Stennis Space Center, MS 39522
The Mississippi Center for Supercomputing Research was established in 1987 by the Mississippi Legislature and the Institutions of Higher Learning (IHL) in order to provide high performance supercomputing (HPC) support for research and instruction at all state universities. The Mississippi Supercomputer User Advisory Committee (MSUAG) was established by the IHL Research Consortium to provide user input and advice to MCSR management and technical staff on policies and procedures for the Center's operations. It includes member representatives from all IHL institutions. The Advisory Group will meet at this MAS conference. Mr. David G. Roach, Director of the MCSR, and Dr. Germana Peggion, MSUAG Chair and Professor at USM Stennis, will conduct the meeting. The agenda includes an update on MCSR HPC facilities and services, introduction of new MCSR staff members, and site reports and ongoing research updates by MSUAG representatives. A Poster Session will follow the MSUAG meeting in which faculty and graduate student researchers will have the opportunity to describe their research projects that involve HPC and the use of MCSR resources. IHL faculty and graduate students, with an interest in HPC and/or MCSR facilities and services, are also invited to attend.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||various articles|
|Publication:||Journal of the Mississippi Academy of Sciences|
|Date:||Jan 1, 2002|
|Previous Article:||Marine and atmospheric sciences.|
|Next Article:||Physics and engineering.|