Development of an automated method for classifying digital images using a combination of image quality modification and Bayesian classification is the subject of this thesis. The specific example is classification of images obtained by monitoring molten plastic in an extruder. These images were to be classified into two groups: the “with particle” (WP) group which showed contaminant particles and the “without particle” (WO) group which did not. Previous work effected the classification using only an adaptive Bayesian model. This work combines adaptive image quality modification with the adaptive Bayesian model. The first objective was to develop an off-line automated method for determining how to modify each individual raw image to obtain the quality required for improved classification results. This was done in a very novel way by defining image quality in terms of probability using a Bayesian classification model. The Nelder Mead Simplex method was then used to optimize the quality. The result was a “Reference Image Database” which was used as a basis for accomplishing the second objective. The second objective was to develop an in-line method for modifying the quality of new images to improve classification over that which could be obtained previously. Case Based Reasoning used the Reference Image Database to locate reference images similar to each new image. The database supplied instructions on how to modify the new image to obtain a better quality image. Experimental verification of the method used a variety of images from the extruder monitor including images purposefully produced to be of wide diversity. Image quality modification was made adaptive by adding new images to the Reference Image Database. When combined with adaptive classification previously employed, error rates decreased from about 10% to less than 1% for most images. For one unusually difficult set of images that exhibited very low local contrast of particles in the image against their background it was necessary to split the Reference Image Database into two parts on the basis of a critical value for local contrast. The end result of this work is a very powerful, flexible and general method for improving classification of digital images that utilizes both image quality modification and classification modeling.

Este trabalho apresenta o sistema de planejamento baseado em casos chamado FAR-OFF (Fast and Accurate Retrieval on Fast Forward). Este sistema usa o planejador FF (Fast-Forward) como um sistema generativo para adaptar os casos resgatados, bem como uma nova regra de similaridade chamada ADG, uma nova política de remoção de casos chamada Minimo-Prejuízo e um método de melhora da qualidade de um plano chamado SQUIRE. Todos esses novos métodos permitem um sistema de planejamento baseado em casos tão eficiente quanto os sistemas de planejamento baseados em busca heurística. / This work presents the FAR-OFF (Fast and Accurate Retrieval on Fast Forward) case-based planning system. This system uses the FF planner (Fast-Forward) as an effective generative system to adapt retrieved cases. It also uses a new similarity rule, called ADG, a new case-deletion policy named Minimal-Injury and a new method to improve the solution quality called SQUIRE. All these features are responsible for the results of the FAR-OFF system that are so efficient as the results of the heuristic search based planning systems.

Com o crescimento do número e da heterogeneidade dos equipamentos presentes nas atuais redes de computadores, o gerenciamento eficaz destes recursos toma-se crítico. Esta atividade exige dos gerentes de redes a disponibilidade de uma grande quantidade de informações sobre os seus equipamentos, as tecnologias envolvidas e os problemas associados a elas. Sistemas de registro de problemas (trouble ticket systems) tem lido utilizados para armazenar os incidentes ocorridos, servindo como uma memória histórica da rede e acumulando o conhecimento derivado do processo de diagnose e resolução de problemas. Todavia, o crescente número de registros armazenados torna a busca manual nestes sistemas por situações similares ocorridas anteriormente muito morosa e imprecisa. Assim, uma solução apropriada para consolidar a memória histórica das redes é o desenvolvimento de um sistema especialista que utilize o conhecimento armazenado nos sistemas de registro de problemas para propor soluções para um problema corrente. Uma abordagem da Inteligência Artificial que tem atraído enorme atenção nos últimos anos e que pode ser utilizada para tal fim é o raciocínio baseado em casos (casebased reasoning). Este paradigma de raciocínio visa propor soluções para novos problemas através da recuperação de um caso similar ocorrido no passado, cuja solução pode ser reutilizada na nova situação. Além disso, os benefícios deste paradigma incluem a capacidade de aprendizado com a experiência, permitindo que novos problemas sejam incorporados e se tomem disponíveis para use em situações futuras, aumentando com isso o conhecimento presente no sistema. Este trabalho apresenta um sistema que utiliza o paradigma de raciocínio baseado em casos aplicado a um sistema de registro de problemas para propor soluções para um novo problema. Esse sistema foi desenvolvido com o propósito de auxiliar no diagnostico e resolução dos problemas em redes. Os problemas típicos deste domínio, a abordagem adotada e os resultados obtidos com o protótipo construído são descritos. / With the increasing number of computer equipments and their increasing heterogeneity, the efficient management of those resources has become a hard job. This activity demands from the network manager a big amount of expertise on network equipments, technologies involved, and eventual problems that may arise. So far, trouble ticket systems (TTS) have been used to store network problems, working like a network historical memory and accumulating the knowledge derived from the diagnosis and troubleshooting of such problems. However, the increasing number of stored tickets makes the manual search of similar situations very slow and inaccurate in these kind of systems. So, an adequate approach to consolidate the network historic memory is the development of an expert system that uses the knowledge stored in the trouble ticket systems to propose a solution for a current problem. Case-based reasoning (CBR), an approach borrowed from Artificial Intelligence that recently had attracted many researchers attention, may be applied to help diagnosing and troubleshooting networking management problems. This reasoning paradigm proposes solution to new problems by retrieving a similar case occurred in the past, whose solution can be reused in the new situation. Furthermore, the benefits of this paradigm include the experience learning capability, allowing new problems being added and becoming available to use in future situations, expanding the knowledge of the system. This work presents a system that uses case-based reasoning applied to a trouble ticket system to propose solutions for a new problem in the network. This system was developed with the aim of helping the diagnostic and troubleshooting of network problems. It describes the typical problems of this domain, the adopted approach and the results obtained with the prototype built.

The rapid growth of the Internet has shaped up the global economy. The stock market information is thus more and more transparent. Although the investors can get more helpful information to judge future trend of the stock market, they may get wrong judgments because the stock market data are too huge to be completely analyzed. Therefore, the purpose of this study is to develop an artificial stock market analyst by employing the information technology with high speed and performance, as well as integrating the artificial intelligence techniques. We exploit case-based reasoning to simulate the analysts in using history stock market data, employ the artificial neural network to imitate the analysts in analyzing the macrofactors of stock market, and apply the fuzzy logic to humanize the artificial stock market analyst in making judgments close to the real stock market analysts. The artificial stock market analyst would use the modified case-based reasoning system combined with the artificial neural network, and incorporate the designed membership functions for macrofactors of stock market. We expect the system to improve the accuracy of Taiwan electric stock price prediction by applying macrofactors from the technical analysis indicators and financial crisis factors, and make better stock trading strategies.

Quality of service is commonly used to shape network traffic to meet specified criteria. The various scenarios include limiting and reserving bandwidth for a particular application, host or user, prioritizing latency sensitive traffic or equal distribution of unreserved bandwidth. The DynShaper software distributes and controls a traffic quota by more sophisticated means than fixed per user limits and simple disconnection after the user reaches the limit. It distributes the quota on a daily basis, where each day receives the same share. The users are sorted into predefined groups with different bandwidths depending on their recent consumption. This classification is periodically updated to ensure the sorting order is maintained. The bandwidths of these groups is dynamically adjusted depending on the actual consumption to provide an efficient utilization. This thesis presents another distribution model using a case-based reasoning approach, a method for machine learning which is classified as conventional artificial intelligence. Case-based reasoning tries to solve new problems based on the solutions of similar problems from the past. Controlling the network traffic to remain within a fixed quota can be modeled as such a problem if the traffic patterns are recurring. Possible solutions can be derived from statistical data and altered to suit the new problems. When an untested solution is applied, the software supervises the execution and revises the solution accordingly, if the actual results deviate from the precalculated schedule.

Professors need alternative programs to support their online teaching. This dissertation reports an initial study in a long-term research agenda for developing a faculty online teaching solution. The primary purpose of the study is to explore faculty perceptions of a case library to help decision makers and researchers determine whether they would pursue the use of such a tool to support faculty online teaching. The secondary purpose of the study is to generate design knowledge to inform future development of and research on this or similar case libraries. The methodology of this study includes three components: development research, rapid prototyping, and qualitative methods. Development research and rapid prototyping provided a three-stage framework for this study: conceptualization, development, and research. I synthesized the literature to create conceptual models of an Online Teaching Case Library (OTCL) at the conceptualization stage, built a prototype to implement the models at the development stage, and conducted research to evaluate the prototype at the research stage. Qualitative methods guided data gathering and analysis. I recruited seven faculty participants based on a purposeful sampling technique. To gather the data, I followed a three-step data collection process: initial interviews, contextual interviews, and final interviews. This process allowed me to observe and interview faculty participants while they were exploring the prototype. I analyzed the data by following an 11-step procedure synthesized from the works of Miles and Huberman (1994) as well as LeCompte and Schensul (1999a). This study found that on one hand, faculty members might use an OTCL, because they perceived that this tool could support their apprenticeship approach to learning to teach. On the other hand, however, their perceived decision to use an OTCL would also be influenced by the perceptions of the usefulness and usability of the tool. The study identified the initial evidence supporting an OTCL as an online teaching resource and the challenges involved in developing and implementing such a solution. It provides a base for decision makers to determine whether they would adopt this tool. It also offers some design guidance for those who do want to pursue this solution to faculty development.

Up to 1998, the Institutes of Nursing in the United Arab Emirates have been using the traditional lecture-based teaching/learning process in their graduate-nursing program. In 1998, however, these Institutes adopted a new approach; namely, the case-based learning (CBL) for the education of their nursing students. This approach emphasizes the use of self-directed and cooperative learning that is supposed to help students increase their critical thinking (CT) level. As the students were experiencing changes in the teaching practices, it was important to determine the effect of the teaching and learning approaches on students' CT abilities, and to describe suggestions needed for improvement. Empirically, very little is known regarding the influence of CBL on a student's CT. The question then remains, as to whether students who have undergone case-based learning, differ significantly in their CT abilities from those who studied in the traditional method. This study investigates the critical thinking skills in relation to two types of nursing educational programs: (a) the traditional teaching and (b) the case-based learning. The professed purpose of the study in hand is to measure and compare the level of critical thinking in participants from each of the two programs. The instrument of measurement guiding this study is the model developed by Facione and Facione (1998). The design has been a comparative descriptive survey. The critical thinking abilities were measured by the CCTST, which was administered to 38 participants from the traditional curriculum and 65 from the case-based learning curriculum who agreed to participate in the study. When the scores were analyzed by using the independent sample 1- test, this study found that, in general, participants from both programs performed badly on the CCTST. Nevertheless, the CBL program participants performed significantly better when compared to the traditional program participants in all aspects of the CCTST. / Thesis (M.Cur.)-University of Natal, Durban, 2001.

The purpose of this study was three-fold: to measure the ability of the Medical Explorers case-based curriculum to improve higher order thinking skills; to evaluate the impact of the Medical Explorers case-based curriculum to help students be self directed learners; and to investigate the impact of the Medical Explorers case-based curriculum to improve student attitudes of the life sciences. The target population for this study was secondary students enrolled in advanced life science programs. The resulting sample (n = 71) consisted of 36 students in the case-based experimental group and 35 students in the control group. Furthermore, this study employed an experimental, pretest-posttest control group research design. The treatment consisted of two instructional strategies: case-based learning and teacher-guided learning. Analysis of covariance indicated no treatment effect on critical thinking ability or Motivation and Self-regulation of Learning. However, the Medical Explorers case-based curriculum did show a treatment effect on student attitudes toward the life sciences. These results seem to indicate that case-based curriculum has a positive impact on students’ perspectives and attitudes about the study of life science as well as their interest in life science based careers. Such outcomes are also a good indicator that students enjoy and perceive the value to use of case studies in science, and because they see value in the work that they do they open up their minds to true learning and integration. Of additional interest was the observation that on average eleventh graders showed consistently stronger gains in critical thinking, motivation and self-regulation of learning strategies, and attitudes toward the life sciences as compared to twelfth grade students. In fact, twelfth grade students showed a pre to post loss on the Watson-Glaser and the MSLQ scores while eleventh grade students showed positive gains on each of these instruments. This decline in twelfth grade performance is an endemic indicator of underlying problems that exists in this transitional year of education and supports the need to strengthen the transitional connections between high schools and institutions of higher learning. / Department of Biology

Development of an automated method for classifying digital images using a combination of image quality modification and Bayesian classification is the subject of this thesis. The specific example is classification of images obtained by monitoring molten plastic in an extruder. These images were to be classified into two groups: the “with particle” (WP) group which showed contaminant particles and the “without particle” (WO) group which did not. Previous work effected the classification using only an adaptive Bayesian model. This work combines adaptive image quality modification with the adaptive Bayesian model. The first objective was to develop an off-line automated method for determining how to modify each individual raw image to obtain the quality required for improved classification results. This was done in a very novel way by defining image quality in terms of probability using a Bayesian classification model. The Nelder Mead Simplex method was then used to optimize the quality. The result was a “Reference Image Database” which was used as a basis for accomplishing the second objective. The second objective was to develop an in-line method for modifying the quality of new images to improve classification over that which could be obtained previously. Case Based Reasoning used the Reference Image Database to locate reference images similar to each new image. The database supplied instructions on how to modify the new image to obtain a better quality image. Experimental verification of the method used a variety of images from the extruder monitor including images purposefully produced to be of wide diversity. Image quality modification was made adaptive by adding new images to the Reference Image Database. When combined with adaptive classification previously employed, error rates decreased from about 10% to less than 1% for most images. For one unusually difficult set of images that exhibited very low local contrast of particles in the image against their background it was necessary to split the Reference Image Database into two parts on the basis of a critical value for local contrast. The end result of this work is a very powerful, flexible and general method for improving classification of digital images that utilizes both image quality modification and classification modeling.

A look at today's US military will see them operating much beyond the scope of protecting and defending the United States. These operations now consist of, but are not limited to humanitarian aid, disaster relief, and conflict resolution. This broad spectrum of operational environments has necessitated a transformation of the individual military services into a hybrid force that can leverage the inherent and emerging capabilities from the strengths of those under the umbrella of the Department of Defense (DOD), this concept has been coined Joint Operations. Supporting Joint Operations requires a new approach to determining a viable military logistics support system. The logistics architecture for these operations has to accommodate scale, time, varied mission objectives, and imperfect information. Compounding the problem is the human in the loop (HITL) decision maker (DM) who is a necessary component for quickly assessing and planning logistics support activities. Past outcomes are not necessarily good indicators of future results, but they can provide a reasonable starting point for planning and prediction of specific needs for future requirements. Adequately forecasting the necessary logistical support structure and commodities needed for any resource intensive environment has progressed well beyond stable demand assumptions to one in which dynamic and nonlinear environments can be captured with some degree of fidelity and accuracy. While these advances are important, a holistic approach that allows exploration of the operational environment or design space does not exist to guide the military logistician in a methodical way to support military forecasting activities. To bridge this capability gap, a method called A Technique for Logistics Architecture Selection (ATLAS) has been developed. This thesis describes and applies the ATLAS method to a notional military scenario that involves the Navy concept of Seabasing and the Marine Corps concept of Distributed Operations applied to a platoon sized element. This work uses modeling and simulation to incorporate expert opinion and knowledge of military operations, dynamic reasoning methods, and certainty analysis to create a decisions support system (DSS) that can be used to provide the DM an enhanced view of the logistics environment and variables that impact specific measures of effectiveness.