Дисертації з теми "Computer-aided assessment systems"

One effective way for organizations to capture expert knowledge and experience is to encapsulate it within an expert system (ES) and make that system available to others. While ES users have access to the system's knowledge, they shoulder the difficult task of appropriately incorporating the ES recommendations into the decision-making process.One proposed application of an ES is in the realm of deception detection. Humans are inherently poor at recognizing deception when it occurs and their confidence in their judgments is poorly calibrated to their performance. An ES has the potential to significantly improve deception detection; however, joining an ES and a human decision maker creates many important questions that must be addressed before such a system will be useful in a field environment. These questions concern changes in decision outcomes, decision processes, and the decision maker that result from ES use.To examine these questions, a prototype system was created that implements new and unobtrusive methods of deception detection. Kinesic analysis examines the body movement of a potential deceiver and linguistic analysis reviews the structure of utterances from a potential deceiver. This prototype, complete with explanations, was utilized in two experiments that examined the effects of access to the prototype, accuracy level of the prototype, user training in deception detection, and novice or professional lie-catcher status of the users.Use of the prototype system was found to significantly improve professional and novice accuracy rates and confidence alignment. Training was found to have no effect on novice accuracy rates. Accuracy level of the prototype significantly elevated accuracy rates and confidence alignment among novices; however, this improvement was imperceptible to the novices. Novices using the prototype performed on a level equivalent to professionals using the prototype. Neither professional nor novice users of the prototype exceeded the performance of the prototype system alone. Implications of these findings include emphasizing the development of computer-based tools to detect deception and defining a new role for human users of such tools.

Diagrams are increasingly used in many design methods, and are being taught in a variety of contexts in higher education such as database conceptual design or software design in computer science. They are an important part of many assessments. Currently computer aided assessments are widely used for multiple choice questions. They lack the ability to assess a student's knowledge in a more comprehensive way, which is required for diagram-type student work. The aim of this research is to develop a semi-automatic assessment framework, which enables the use of computer to support the assessment process of diagrammatic solutions, with the focus of ensuring the consistency of grades and feedback on solutions. A novel trace model, that captures design traces of student solutions, was developed as a part of the framework and was used to provide the matching criteria for grouping the solutions. A new marking style, partial marking, was developed to mark these solution groups manually. The Case-Based Reasoning method is utilised in the framework to mark some of the groups automatically. A guideline for scenario writing was proposed to increase the efficiency of automatic marking. A prototype diagram editor, a marking tool and scenario writing environment were implemented for the proposed framework in order to demonstrate proof of concept. The results of experiments show that the framework is feasible to use in the formative assessment and it provides consistent marking and personalised feedback to the students. The framework also has the potential to significantly reduce the time and effort required by the examiner to mark student diagrams. Although the constructed framework was specifically used for the assessment of database diagrams, the framework is generic enough to be used for other types of graph-based diagram.

Thesis (M.S. in Systems Engineering)--Naval Postgraduate School, September 2008.
Thesis Advisor(s): Rhoades, Mark M. "September 2008." Description based on title screen as viewed on November 4, 2008.. Includes bibliographical references (p. 75). Also available in print.

Academic assessment that is used to assist in producing knowledge having a lasting effect should be the concern of all stakeholders. Subjective assessment contributes to achieving this type of knowledge. In particular for multi-steps questions (MSQ) human assessors play important roles. This evaluation includes assessing solution strategies, working produced from employing those strategies, and the final answer. This assessment can reveal levels of conceptual understanding displayed by students. Subjective assessment gives options to students to express their understanding of the topic. However, when computer-aided assessment (CAA) systems are used, there is much less subjectivity. These systems subdivide MSQ into stub-steps, and students can provide one step’s answer without showing their workings. This technique is similar to assessing objective questions such as multiple-choice questions (MCQ) which do not, generally, examine conceptual understanding. The inability of CAA systems to assess solutions of MSQ is considered a complex problem. To solve a complex problem, accumulating knowledge about the problem, producing solutions and acquiring ability to apply those solutions are required. Design science research (DSR), which is the paradigm to investigate complex problems- in information systems research and design science - has been used in this research. Two hundred and fifty-eight student scripts containing working with solutions of MSQ were analysed using an inductive qualitative content analysis and the quantitative survey approach. The findings from the analyses of scripts were re-analysed iteratively to produce the knowledge that contributed to the understanding of the research problem and producing solutions. This knowledge includes types of solution strategies and student errors that can also be used in designing questions to be used in CAA systems for assessing objective questions. The methodologies that were used in producing this knowledge could also be used in similar disciplines to produce similar output. A significant contribution of this research is to analyse student workings for extracting solution strategies, and this helps understanding ways of solving MSQ and obtaining these strategies. This approach does not appear to have been used elsewhere. The questionnaire that was created for measuring the significance of types of strategies and student errors is an innovative instrument. It has not been used previously and can be adapted for similar studies. The information in the analyses of student workings also provides directions when solutions of MSQ are assessed by humans. Furthermore, all the above information as well as domain knowledge was used to develop DSR constructs and models artefacts. These models represent processes and data in the student workings, which was divided into sub-tasks with each task being represented by one model. The implementation of a model in a DSR prototype artefact accomplishes the assessment of that sub-task. One of these models was implemented, overcoming difficulty in designing user interfaces that can be used without disclosing the domain knowledge students are examined. These user interfaces show an innovative way of extracting conceptual understanding of topics. Furthermore, these models are highly generalisable to a very broad class of problems.

碩士
中國文化大學
生活應用科學研究所
87
The purpose of this study was to design a computer added weight control system by using software Visual Basic 6.0. The system was suitable for any age above 3 years old by manipulating personal computer. The system will achieve the goal of nutrition education to maintain the ideal body weight. The system consists of three parts: ideal body weight calculation、weight reduction system and physical fitness. In the first part, in accordance with user''s age、gender、height、weight and daily physical activity to figure his daily calorie need and range of ideal body weight. When user is younger than 19 (age 19 counted), system use weight-height index (WHI) to calculate his ideal body weight. If person is above 19, system applies body mass index (BMI) to calculate his ideal body weight. For calculation of daily energy requirement, the figures and the formulas of basal metabolism rate (BMR) factor and physical activity index are adapted to from the Department of Health. Next, the system will give the proper advice of the diet and exercise plan for the people whose body weight are ideal or lighter. In the diet plan, the carbohydrate、protein and fat contain was 58-68%、10-14% and 20-30% of total calorie. The system recommended the user whose weight was 10 % less than ideal body weight to intake extra 500 calories of the daily energy intake. The energy expenditure by exercise was counted for expending 200 calories estimated by resting metabolism rate (MET). If user''s weight was within the range of ideal body weight or lighter, we recommended him to take exercise more than 3 times a week. If one was overweight or obese, it will connect to the second part: weight-reduction system. The second part was weight-reduction system. At first, we provide user a diet plan with 300 calories lower than the daily energy intake and 200 calories exercise expenditure. If user wants to decide weight reduction goal by himself, he can choose one of time periods which system provides. The system will calculate how many calorie intakes should be reduced according to users body weight and time periods for weight reduction. The system won''t provide any weight reduction suggestion if daily food intake was less than 1200 calories or the reduction of energy intake was over 1000 calories. After that, user also can choose the pattern energy expenditure and intake; therefore, the system will look for apposite diet plan and physical activity chosen by the system was appropriate to individual case. If user was younger than 12, or his body weight was overweight two times of ideal body weight, system won''t suggestion for losing weight; however, the system advice him to see a doctor or nutritionist for consulting. Besides that, system also provide nutrition education about obesity, including the epidemiology of obesity、the complication of obesity、the cooking tips for high and low calorie food and correct concept for eating-out. The third part was physical fitness. User inputs the type of exercise and duration time; system will output the exercise energy expenditure for each individual event and daily total exercise energy expenditure. Besides, it provides the maximum heart rate calculation、definition and introduction of aerobic exercise、the method to determine healthy related physical fitness、simple stretch and demonstration of aerobics in order to give the user basic knowledge about healthy related exercise. Finally, the system reviews how much daily energy intakes by connecting to nutritional component analysis system and how much energy expended by daily activity and exercise to estimate the balance between energy intake and expenditure. In conclusion, this system not only provides the calculation of ideal body weight and daily calorie need、suggestion of diet plan、physical activity and energy balance analysis, but also give user a correct concept of weight reduction with physical fitness. The system could be a personal software available for everyone in hospital、school and home.

碩士
國立臺灣科技大學
營建工程系
88
This study focuses on developing an automated safety assessment system for fire damaged RC structures. Safety assessment for fire damaged RC structures is an important task for engineers in construction safety evaluation region. Assessment results can effect not only the property owner’s benefit, but also directly impact public safety. In this research, the system was developed to assist the expert in determining the safety condition for a structure component. The system includes a MYCIN expert reasoning engine integrated with a database management system (DBMS), improving the assessment process by providing a logical and systematic manner to analyze collected data in a real time basis. In the aspect of construction where experience accumulation is retarded, effective management of knowledge acquisition and problem solving is crucial. Therefore, this study presents a model that significantly acquires experts’ knowledge and experience in a systematic way, analyzes collected data and identifies safety conditions of structural components, thus preventing possible disasters in development.

碩士
國立臺灣科技大學
營建工程系
100
Construction safety is a major issue. Constructions are often conducted in unfavorable environemnts (e.g. open-air, hillside, waterfront) with powerful machineries. Furtherore, a construction project often involves many sub-contractors, and many substroctors often have limited resources and are unable to establish effective management. As a result, construction hazards occurs frequenetly, and improvements are urgently needed. To improve the practice of construction safety management, Council of Labor Affairs of Taiwan (CLA) announced a series of guidelines. These guidelines require risk assessments for labor health and safety must be included in the construction design phase. The guidelines have established the need for implementing safety managements, work flows, methods, and application forms into the construction design phase. The guide-lines hope to reduce the complexity of safety management and tendency of having high accident rates at construction sites. This work developes a web-based risk assessment system in construction engineering. The system assists and guides the risk assessment in planning, design, construction phases. Assessment results are recorded in the system, and needed tables and charts are generated accordingly. The system further collaborates with BIM (Building Information Model) applications to help engineers visualize temporary and spatial variation of risks identified by using the system. The model can further act as a communication tool to coordinate discussions between engineers for developing safety measures.

Preference assessments are an evidence-based procedures used to identify potential reinforcers for persons with developmental disabilities. There is a need to develop effective and efficient procedures to teach students and staff to conduct preference assessments, but only a small number of studies have been conducted and only two have used self-instructional materials. A recent study by Ramon et al. (2012) found that a self-instructional manual was more effective than a method description extracted from published articles for teaching university students to conduct multiple-stimulus without replacement preference assessments for persons with developmental disabilities. The present study extended this research by (a) adapting the self-instructional manual from Ramon et al. for online delivery, (b) adding video modeling as a teaching component, and (c) delivering the training package using a modified computer-aided personalized system of instruction (CAPSI, Pear and Kinsner, 1988). The training package was evaluated using a multiple-baseline design across three university students, replicated across three more students; and a multiple-baseline design across a pair of staff members, replicated a across a second pair. During the baseline phase, participants studied a two-page written description of the assessment procedure adapted from published studies. During the self-instructional manual phase, participants completed all of the following online: studied the self-instructional manual presented in eight units, viewed video demonstrations of the procedure, and completed review exercises scored by the computer program to demonstrate mastery of each study unit. Performance accuracy of each participant was scored using a standard behaviour checklist during a simulated preference assessment conducted following each phase. Clear and immediate improvement in performance accuracy was observed in all participants immediately following the self-instructional training package. Overall, students improved from a mean of 35% correct in baseline to a mean of 94% correct following CAPSI and staff improved from a mean of 23% correct in baseline to a mean of 87% correct following CAPSI. During retention and generalization assessments conducted from 7 to 17 days following self-instructional training, five of the six students and one of the four staff members performed at or above 85% correct (the mastery criterion). The findings showed that online delivery of the self-instructional manual plus video modeling has tremendous potential for providing an effective method for teaching a preference assessment procedure without face-to-face instruction.

碩士
國立陽明大學
物理治療暨輔助科技學系
101
Abstract The ultimate goal of prosthesis training for amputees is to regain independent walking ability. Static weight-bearing ability is a very good predictor of whether subject can walk independently. Good weight-bearing ability is a prerequisite for proper prosthetic control and balance ability, so it may be to walk independently. There is no objective measurement or device to evaluate and train weight-bearing ability and control ability for transtibial amputees in clinical. Purpose: Our study was to develop a computer-aided standing weight-bearing ability assessment and training system for transtibial amputees and to evaluate effects of system functions for improving weight-bearing and control ability. Methods: This system was composed of specific-designed force plates, PC and using LabVIEW to write specifically for weight-bearing assessment and training software. The system can instantly analyze two feet force changing and provide immediately visual and auditory feedback. System validity verification used the weights for the gold standard values. Twenty-six healthy subjects and fifteen experienced prosthetic users were recruited for reliability verification. In addition, we also executed clinical trial for weight-bearing training preliminary test. Three subjects in the intervention group and two subjects in the control group were recruited respectively. The intervention group received the system training for weight-bearing and stepping. The control group received traditional training. The intervention included three days in a week for four weeks and each session lasted about 30-40 minutes. Intraclass correlation coefficient(ICC) was used to estimate the validity and reliability. The comparable parameters between pre-training and post-training were assessed whether the progressive training effect. Results: ICC was 0.99 on validity verification, and p<0.01. Reliability verification, ICC values only on ND ST% and D ST% were 0.612 in healthy subjects, ICC values of other parameters were between 0.892 and 0.979, and p<0.011.After four weeks of system training intervention, weight-bearing ability--Intervention group, three subjects increased 11.27%, 9.30% and 19.85% in Ps%,--Control group, two subjects increased 5.39% and 4.95% in Ps%. Control ability--Intervention group, subjects increased 94.69%, 102.7% and 63.37% in Ps max%. Control group, subjects increased 33.32% and 79.67% in Ps max%. And two subjects did not touch parallel bar by hand in intervention group. Conclusion: ICC values are very high in reliability and validity, and we can use this system to evaluate standing weight-bearing ability for transtibial amputation patients. After training, the average of (S%-Ps%) is 3.01% in intervention group, better than in control group(7.24%). The average of (S ST%-Ps ST%) is 4.1% in intervention group, also superior to the control group (18.9%). Therefore, our system provides an objective and effective assessment and training method for transtibial amputation patients.

碩士
長庚大學
醫療機電工程研究所
94
Psoriasis is a common skin disease affecting affects 2–3% population worldwild. Accordingly, there are approximately 100 thousand individuals having psoriasis in Taiwan. The etiology of psoriasis remains unknown, however, it is believed to be a multifactorial and polygenetically inherited disease. The main features of psoriasis include abnormal epidermal keratinocyte differentiation and proliferation, superficial vascular remodelling and a T-cell-predominant inflammatory infiltrate. Chronic plaque psoriasis, the most common form of psoriasis, is a papulosquamous disease defined by erythematous plaques with a silvery scale involving trunk, extremities, scalp, face, and even whole body surface. Of the severity scoring systems currently available for psoriasis, the PASI scoring system has been most extensively tested. It consists of the involved surface areas of the body and the severity of individual psoriatic lesions, such as erythema, scaling, and thickness. Although it is convenient, the PASI scoring system is a subjective method based on a physician’s experience. Thus, the reliability of the visual grading scoring system is not satisfactory. The marked interindividual variation between physicians may have a significant impact on the medical strategy decision, and accuracy of new medical trial before and after evaluation.Oral medicine, topical medicine and Ultra-Violet phototherapy and have been used successfully for the treatment of psoriasis. Oral medicine, including methotrexate; cyclosporine, oral retinoids, are frequently associated with skin xerosis, itchy, and serious side effects including myelosuppression, impairment of liver/renal function. Phototherapy and topical therapies remain the gold standard of treating psoriasis.Ultraviolet A (UVA, 360nm) and narrow band ultraviolet B (UVB, 310nm) phototherapy are usually administered in specialty centers due to the bulky size of phototherapy devices. The adjustment of therapeutic dosimetry is based on the physicians experience and patients skin response before each light therapy. Optimal therapeutic dosimetry is undetermined because studies on psoriatic skin-light interaction is lacking. Thus, studies focus on the issue of psoriatic skin-light interaction is mandatory to disclose the optimal light therapy dosimetry. In first part of research, a Computer Aided Psoriasis Area and Severity Assessment system had been developed. It involves area assessment module, scaling assessment module, redness assessment module, thickness assessment module, network-based anamnesis database, and clinical verifying psoriasis area and severity assessment. The results show that there is no significant difference in comparison with the data of doctor in area assessment, scaling assessment, and redness assessment, while P>0.05; There is significant difference just in thickness assessment, while P<0.05; There is no significant difference in overall PASI assessment, while P>0.05. In second part, a Computer Aided Skin Assessment optical simulation analysis Integration system had been developed. By optical simulation, respectively, simulating normal and psoriasis skin in 1-6MED (minimal erythema dose, MED), the result shows that the intensity of normal skin is larger than that of psoriasis skin in a unit of area. This is, psoriasis skin has poor absorbing rate. Then, in clinical verification, psoriasis skin still can get effective treatment. In third part, a Computer Aided Skin phototherapy Integration system had been developed. It involves clock circuit module, lamp module, and elevator module. In this research, a Computer Aided Psoriasis Area and Severity Assessment and Skin phototherapy Integration system had been developed, and can immediately be used clinically. It can improve the treatment effect and provide the quantitative degree of psoriasis area and severity Assessment. Therefore, it provides a objective and quantitative assessment for long term assessment. According to the configuration of skin optical parameter, it practically simulates light therapy in 1-6MED (minimal erythema dose, MED), and provides strategies for improving treatment model efficiently and shorten treatment time.

碩士
國立嘉義大學
特殊教育學系研究所
95
The Computer Aided Assessment System (CAAS) for the blind student in the elementary school is developed by the method of Action Research (AR) in this thesis. The assessment of the blind student can be adequately accomplished with these 3 practices – “Electronic Testing”(ET),”Braille Computer”(BC) and “Assessment Accommodation ”(AA) - in the CAAS. The composition of the CAAS includes: 1. The ET is made following the principles of “Make-up” and ”Item-type” modifications. 2. 37 skills for BC operation is summarized for the assessment process. 3. AA is provided depending on the content of individual subject. The administration of CAAS can be divided into the following 3 phases – Testing Preparation, Assessment Process, Result Review. If the utilization of ET and BC could not afford to compensate the handicaps of the blinded in the assessment, AA such as extended time and audio assistance should be provided. The effect of the CAAS can be evaluated by the “multiplication” relationship between ET and BC. The result in the case study of Tsung-Tsung indicates that the CAAS is more suitable than the traditional Braille Assessment System. Future research for CAAS include: to develop individual CAAS for different sessions of school and to adopt CAAS in the official entrance examinations.

Learning the knowledge and procedures of the Assessment of Basic Learning Abilities (ABLA) is a major practical priority for direct-care service providers who work with individuals with intellectual disabilities (IDs) and autism spectrum disorders (ASDs). The present study compared the effects of two training conditions involving a self-instructional manual (SIM) alone and its combination with a Web-based computer-aided personalized system of instruction (WebCAPSI) program (SIM plus WebCAPSI) on teaching knowledge and implementation of the ABLA to university students. A multiple-baseline design across training conditions was used. The results indicate that the SIM combined with passing unit tests delivered via WebCAPSI was beneficial for knowledge acquisition while viewing demonstration videos was beneficial for conducting the ABLA. The findings suggest that a combination of a SIM combined with unit tests delivered via WebCAPSI and videos provides an effective approach to teach both knowledge and application of behavioural procedures to potential knowledge users.

博士
南台科技大學
電機工程系
102
According to the latest annual statistics report of the Ministry of Interior, as of 2013, there is an estimated 382,000 people with physical disability. This figure belongs to the largest percentage (34%) of the disabled population. Among them, about 40,000 people are patients with spinal cord injury (SCI), and annually, a continuous increase of around a thousand people is expected. In general, regardless of the field of experimental medical research, most studies are done through experiments on animals. The analysis of the motor behavior among animals is also often used in comparing experimental studies. Studies related to spinal injuries mostly use male rats of the Sprague-Dawley strains (SD) as the object of animal experiments. The Basso-Beattie-Bresnahan (BBB) open field locomotor scale test is used as the criterion to evaluate hind limb motor function. In investigating the relationship between cellular and molecular motor functions, this study firstly attempted to assess and compare the protective effect on hind limb functional recovery after human umbilical cord blood cells (HUCBC) or DMEM medium (without cell culture, as a vehicle) which is given through purified hematopoietic stem cells (CD34+), secreting cells or substances called conditioned medium (CM) cultured from human umbilical cord blood-derived CD34+ cells, combined CM and 17β-estradiol medium were given intravenously immediately after spinal cord injury. In this study, the following conditions for recovery after spinal cord injury were designed: (1) the increase in number and activity of apoptotic protease (caspase-3) of DNA terminal deoxynucleotidyl transferase-mediated dUTP DNA labeled cells (TUNEL - positive cell is evidence) is used as evidence in determining apoptosis; (2) the increase in performance of interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α as support) and the activity of myeloperoxidase (MPO) are evidences of activated inflammatory response; (3) the increase in nitric oxide (NO), 2,3-dihydroxybenzoic acid, and malondialdehyde (MDA as evidence) and the reduction of superoxide dismutase (SOD), glutathione peroxidase (GPx), and glutathione reductase (GR) are evidences of increased oxidative stress; (4) the amount of glial fibrillary acidic protein-positive cells (GFAP-positive cells) shows evidence of increased astrogliosis. The entire study found out that rats given conditioned medium, 17β-estradiol, and combined CM and 17β-estradiol medium immediately after SCI can effectively improve their situation on the 7th day of injury. Also, injured spinal cord is neuroprotected, and this effectively improves hind limb motor function. Subsequently, in treating SCI, combined CM and 17β-estradiol medium is better than a single CM or 17β-estradiol. These results illustrate that 17β-estradiol can strengthen the cultivation of CD34+ stem cell-conditioned medium derived from human umbilical cord blood cells, increasing the therapeutic effect on injured spinal cord of experimental rats. In coordination with the above-mentioned work, this study also attempted to establish a set of computer-assisted quantitative assessment system for animal behavior analysis on the basis of imaging technology in order to test the hind limb behavior of observed rats after spinal cord injury. The system mainly conducted image segmentation and coding of animal behavior through images taken using a high-speed camera. Parameters such as variation of joint angles, variances of gait cycle, stride length, and shift in center of gravity when walking, etc. were quantized through a pre-designed image analysis software program. Afterwards, the correlation between quantization parameters and traditional BBB scores was studied. This study also found that the variation of joint angle and the variance of the shift in center of gravity when walking have a good correlation with traditional assessment methods of BBB scores. Lastly, it is anticipated that the quantitative assessment system developed could provide an objective and reliable secondary assessment basis for future BBB score-related tests.