Although using statistical modeling techniques for detecting anomaly intrusion and profiling user behavior with network audit data has been studied for more than a decade, the minimum sample size required from each network site to fit a model is not clear. A large sample size requires more resources to collect and analyze data, while a small sample size increases both false positives and false negatives. Determining the minimum sample size and developing a better corresponding classification algorithm are two essential tasks in the area of intrusion detection. This research aimed to address these two tasks with the Markov Chain Monte Carlo, bootstrap simulation, and hierarchical random effects logistic regression modeling methods. The study cohorts were drawn from the 1998 Defense Advanced Research Projects Agency Intrusion Detection Evaluation offline, and the Third International Knowledge Discovery and Data Mining Tools Competition 1999 (KDD-cup) data. Sensitivity, specificity, area under the receiver operating characteristic curve, and misclassification rate were used for evaluating the performance of the proposed technique. The research demonstrated that a minimum sample size of 500 provides a sensitivity value of 0.85, a specificity value of 0.92, and a kappa-statistic of 0.77 in classification, and the results from the minimum sample-based model were comparable with the full sample-based model. The research also developed a multilevel classification algorithm that provides remarkably better classification performance. Finally, the research developed a standardized risk score to assess and evaluate the classification performance of the minimum sample size and multilevel algorithm. Compared with the KDD-cup 1999 top winning results, the risk score had similar performance, but in a far simpler format. In summary, the research provides statistically sound evidence for determining an appropriate sample size of audit data in intrusion detection. Since audit data are available and collectable from network traffic log files, the risk score and the classification algorithm can be implemented in common computer languages. The results of the research would significantly improve and facilitate real-time anomaly intrusion detection, typically in the high-speed mobile wireless network environment, and provide a new statistical approach to this area.

At Gallaudet University (GU), the undergraduate computer information systems (CIS) program serves deaf and hard of hearing (DHH) students who primarily depend upon visual sensory inputs to process information. Unique environmental variables have adversely affected student learning delivered through traditional one-pace-fits-all classroom lectures. These factors include GU's liberal undergraduate admission policy, lack of a universal signing standard, extensive use of fingerspelling in CIS signed lectures, and the practice-based skill acquisition nature of CIS content which requires learning labs replicating a real-world environment. Recent literature confirms many advantages inherent in using e-Learning in an online learning environment. Major benefits experienced among the hearing population from academic institutions and the corporate training arena serve as guideposts for addressing GU's teaching and learning challenges. To realize these benefits, additional aspects of the deaf education and learning environment must be accommodated. These include close social interaction and peer support and degrees of English language difficulty among many deaf learners. To incorporate recent advances of e-Learning into GU's inclusive classroom, the study aimed to design, develop, implement, and evaluate all instructional delivery system that would blend e-Learning with classroom instruction to enhance DHH college student comprehension o/CIS content. Due to limited literature guidance for blended learning design, the special study population, and the under-researched topic, extensive pre-trial preparation work was planned and performed in the fall 2005 semester. The preparation phase involved the design and development of the blended instructional delivery system prototype. Data collection instruments and protocols were developed and validated by an expert panel as part of the groundwork. The prototype was implemented and evaluated among three selected CIS courses during the spring 2006 semester. A descriptive and explorative multiple-case study methodology was chosen to collect, analyze, and report on data. Throughout the trial, iterations of concurrent data collection and analysis were conducted in parallel with modifications and new development driven by user feedback and preliminary findings. Learning outcome and learner satisfaction were evidently improved. Results were validated through triangulation, member checking, and external audit. The chain of evidence, thus established, has indicated that the study has successfully accomplished its goal.

The study questions: How can heterogeneous, technology-rich, classrooms support successful advanced or enriched e-learning experiences for teachers and gifted learners? Advanced learners in multi-ability classrooms are not afforded the opportunities to experience enriched or accelerated learning based on their individual needs due to the nature of traditional classroom instruction and management. Consequently, because instruction is targeted at the average learner, advanced students do not reach their full potential and do not show progress consistent with their ability. In light of No Child Left Behind, school reform initiatives, and standards-based learning, all children should have the opportunity to excel to their fullest potential. Differentiating for all students is an expectation of 21 51 century schools and heterogeneous classrooms. Technology is a largely untapped resource for reforming and restructuring instructional activities within the classroom. Strategies used with the gifted population as well as differentiation processes that help all students learn were used in an e-learning environment. Teachers were surveyed to establish current use and then provided with professional development for using e-learning technologies with their advanced learners. Advanced learners used e-learning to support individualized instruction. Teachers and students took part in three individual interviews as the e-learning modules were experienced in the classroom. They discussed the entire experience in a facilitated focus group at the end of the research process. Teachers learned how to manage e-learning activities to provide alternative instruction with advanced learners in their classrooms. Advanced learners experienced an independent method for learning that provided opportunities for enriched and advanced learning within heterogeneous classroom settings.

There has been extensive research on the design and development of software agents. However, little has been reported about educational applications involving agents or on how users interact with them in a multimedia learning environment. Furthermore, music departments in many school districts have a real need for instructional tools that would help students learn basic principles of ethnic wind musical instruments and music elements. Currently there are few, if any, music learning programs that offer personalized instruction to students. The investigator developed two programs: An interactive multimedia panpipe program (IMPP) and the same program with a virtual Internet agent (VIA) called a virtual Internet agent program (VIAP). She tested the hypothesis that a VIA improves the quality and effectiveness of online, interactive learning programs. The IMPP and the VIAP were designed and programmed based on accepted instructional and learning theories. The pentatonic panpipe program (PPP) has been demonstrated to be an excellent vehicle for teaching basic music elements. The IMPP and VIAP contain methodology for learning how to build and play the pentatonic panpipe. In order to assess the pedagogical value of VIAP, the researcher designed and developed two interactive Internet programs, one with agents (VIA), and the other without, '(or teaching basic principles on how to build and play the Andean pentatonic panpipe. The programs introduce basic units of learning on acoustics, Andean culture, and wind instruments. These programs helped to determine the effectiveness of VIA as a teaching enhancement. Results were validated using non-standardized pretests and posttests to both the experimental and control groups. There was also a survey questionnaire covering the IMPP and the VIAP and two additional instruments, a group appraisal form and a rating sheet, for collecting data not obtained from paper-and-pencil tests. Subjects were selected randomly from a population of 42 fifth grade children, attending an elementary school. These subjects were tested and observed in a computer lab environment. There were two sets of students, 20 in the experimental group, with VIA, and 22 in the control group, without VIA. The findings showed that agent -based technologies vastly improved the quality and effectiveness of online instructional systems. Posttest scores revealed a greater gain made by the experimental group than the control group. The difference in the mean scores was significant. Since the results varied significantly, Q < .05, among the experimental and controls, the null hypothesis was rejected (A = B) and it was concluded that the mean score of the experimental group was significantly greater than the control group. This validated the original hypothesis that the use of VIA in online interactive learning environments improves knowledge acquisition, retention, and satisfaction in an uninitiated sample group. VIAP is an interactive teaching program with virtual Internet agents that was proven to be a highly effective pedagogical tool.

The United States District Courts employ a decentralized automation strategy. This strategy allows each federal district to utilize automation resources in such areas as case management, court financial services, and decision support systems. With changes in technology and increased demand for automation support courts have adopted additional hardware platforms and communications technology at the direction of the Administrative Office of the Courts in Washington, D.C. with an increased demand for automation technology increased technical expertise is also required. This expertise needs to be accompanied by sound system administration practices. The literature indicates that no administrative model exists for the courts that thoroughly deals with the administration of hardware, software, security, operations, and networks. The literature supports that the lack of procedural formalization in automation operations can put information resources at risk. Consequently, this study is designed to assess t \ e practice of system administration in the courts. The assessment is based on the development of a system administration model that allows the courts to conform to sound UNIX administration practices as well as federal regulations and guidelines. The assessment is based on a survey instrument that addresses general and specific areas of system administration. The results of the survey accept the basic hypothesis that United States District Court system administration practices do not presently conform to accepted data center operation practices. This indicates that court information resources are at risk and expected administration tasks are not entirely supported by court information systems. However, it could not be ascertained that the man-machine interface of court systems is unreliable. This study provided a means to assist the courts to strengthen their administration practices. From the assessment it is recommended that the courts adopt a control systems methodology that can support the various court environments. Additional study is also required to better assess the reliability of the court system man-machine interface.

Although the title "Software Engineer" is widely used throughout industry, each software practitioner who bears the title is not a software engineer by many standards. In many instances, people believe that "Software Engineer" is a euphemism for a "programmer" who writes well-structured code while others believe it is about the software used by engineers. These suppositions expose the ignorance about the historical and legal meaning of "Engineer". Furthermore, the general use of the designation "Engineer" by other non-engineering professions has resulted in weakening the title. Licensing and certification are two possible methods to restrict the use of and restore respectability to the title "Software Engineer ". Unfortunately, efforts to license and certify software engineers have led to debate, controversy and opposition throughout the software community. This dissertation presumes that the debate over licensing and certification of software engineers is not truly a debate about licensing and certification but a debate about classification. This dissertation presents a system called SPECS (Software Practitioner Expert Classification System). The researcher suggests that a preliminary step prior to initiating the process of certification or licensing should be one of classification. The goal of classification is to associate the proper job title with the professional qualifications and services of the practitioner. The SPECS prototype has an underlying foundation in software engineering. Eight key components are used to build the classification model. These components include initial professional education, accreditation, skills development, certification, licensing, professional societies, continuing education and code of ethics. The significance of this dissertation is through the research and development of the classification model and the decision support system SPECS, an organization is provided a standard method to properly classify software practitioners and thus identify software engineers that qualify for certification or licensing. In addition, the research contributes to the field of expert system validation, knowledge discovery and knowledge acquisition.

The problem addressed was that many school districts do not have the operational funding to provide sufficient technology support. Whereas large companies strive to have at least one professional technology support person for every 50 computers, few school districts can afford to fund this ratio. Further, technology support is frequently ignored in the school district if it comes at the expense of hiring teachers to provide educational opportunities or reducing class sizes for students. The People Capability Maturity Model (P-CMM) is a tool for managing and developing an organization's workforce practices. A 6-month application of the P-CMM was applied to a Support Services Department in Volusia County School District in Florida. The goal was to explore the effectiveness of the application of the P-CMM in improving the use of best practices in a limited technology support personnel department. The results describe a single-case study of the application of the P-CMM model to a public school district technology support department. The IDEAL (Initiating, Diagnosing, Establishing, Acting, and Learning) process was used in applying the P-CMM. During the initiating phase, participation was gained and an overview of the P-CMM was provided. At the diagnosing phase, a questionnaire and a gap analysis found that the organization was assessed at Level 2 of the P-CMM. The area of Training and Development was identified as particularly weak in the organization, and during the establishing phase it was identified as the focus for the application. Competency Analysis was administered to determine what training was required. At the acting phase, an action plan was developed. During the learning phase, the questionnaire and the gap analysis were repeated to assess changes in workforce competencies. Although the organization did not reach Level 3 of the P-CMM, competencies were identified and improvements were made in Training and Development. This case study may be used as a model for replication in other school districts or organizations that need to improve the use of limited technology support in their organization.

This investigative study is being explored to design an architecture that will support users of computer systems in customizing their Graphical User Interface to their individual preferences. The architecture will give users the flexibility to personalize their user interface within a software application environment. Intelligent interfaces are autonomous and adaptive computer programs that can operate within a software environment such as an operating system, a database program or a computer network environment. These intelligent interfaces are known to assist users in routine computer related tasks, while still accommodating individual habits and preferences. Since the invention of Intelligent Interfaces, this is the best representation for users to have the opportunity to individualize their own GUIs. The main objective is to develop a model for understanding intelligent interface design and development issues associated with an effective and a successful graphical user interface. Examining effective GUI design characteristics will provide significant insight into future developments. Designing an Intelligent Interface based on successes will clarify the process and benefits of Graphical User Interface Technology. The design method that is being used in this paper makes pragmatic use of objects, classes and derived classes to represent the framework. Designing a user interface begins with the creation of a variety of models along with the function of the system. The format we used in our methodology was a pattern design developed by the rum Gamma group. Using their design pattern can assist a designer in improving the maintenance of the systems intent, specific classes and object interactions. Our architecture consists of two systems that are basic Graphical User Interfaces (GUIs) (including The Patricia GUI) and basic widgets to build the individual GUIs. The aggregate of both systems will have an interaction with delegated object composition to drive the coordinate building of the graphical user interface. The gamma group design patterns has a consistent structure that can be reused over and over again.

This study investigated the effects of hypertext computer-assisted instruction in mathematics on the achievement levels of fifth-grade students. The purpose of this study was to determine whether students who received the hypertext computer-assisted instruction performed significantly different in mathematics achievement than students who received the classroom instruction by the respective teacher. The classroom instruction for the students in this study involved manipulative teaching methods that included math relay games, team tournament, and Activities that Integrate Mathematics and Science (AIMS) work sheets. The research method used was a pre-test / post-test research design method with control and experimental groups of equal size. Random selection was used to select students for the groups from the population of students who had parental permission to participate in the study. Forty fifth-grade students were randomly selected for the control and experimental groups. All students were administered a pre-test on mathematics achievement at the beginning of the study to establish a baseline for any differences in the achievement levels between the control and experimental groups. An analysis of variance was performed on these pre-test scores that showed no significant difference between the two groups. The study consisted of nine lessons, each approximately thirty minutes long. The lessons were administered once a week over a period of three months. While the experimental group received hypertext computer-assisted instruction with the direction of the researcher, the control group received classroom instruction in mathematics under the direction of the respective classroom teacher. The lessons followed the curriculum guide for the state of Virginia and the county of Loudoun. At the conclusion of the study, a post- test was administered. Statistical analyses were run against the results to determine the significance of the hypertext computer-assisted instruction on the achievement levels of the experimental group when compared to the control group. The research hypothesis tested in this study was that students who received hypertext computer-assisted instruction demonstrated greater gains in mathematics achievement than students who received the classroom instruction. Results of the analysis of variance test showed that there were no significant differences between the two groups. Consequently, the research hypothesis was not supported by the data analysis of this study. It was recommended that further research be conducted to determine the effectiveness of computer-assisted instruction in schools. Continued research is needed to learn more information about the impact of computer-assisted instruction on student achievement. If schools are to determine how to best integrate computer technology into the curriculum and which methods are best, more empirical research is necessary. As computer technology matures, it will continue to increase in its effectiveness. As effectiveness is realized, computer technology will become more widespread and will potentially evolve into a significant tool in the educational process.

The purpose of this study was to investigate job satisfaction within the military which resulted from the information system used during the job description phase for new military personnel entering the service. The study was conducted by using 210 new military personnel who entered the United States Army. The participants in the study were divided into three major job classification groups. The three groups were Combat Arms, Combat Service Support, and Service Support. Each participant used either the traditional/current job description process (print and picture s) or an automated information system. The results of the study showed that it was a significant relationship bet ween the traditional/current job description process (print and pictures) and the automated information system. The participants who used the automated information system showed more job satisfaction. A statistical analysis was performed on the internal, external, and overall job satisfaction scores. All scores for the participants who used the automated information system were below the level of significance. The level of significance (alpha) for this study was. 05. The Minnesota Satisfaction Questionnaire was used to obtain the scores used in this study.