• Journal of the Korea Institute of Military Science and Technology
• /
• v.1 no.1
• /
• pp.82-91
• /
• 1998

This paper presents a neural networks based approach for the problem of sensor failure detection and accommodation for ship without physical redundancy in the sensors. The designed model consists of two neural networks. The first neural network is responsible for the failure detection and the second neural network is responsible for the failure identification and accommodation. On the yaw rate sensor of ship, simulation results indicates that the proposed method can be useful as failure detector and sensor estimator.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.709-710
• /
• 2023

최근 농가 인구의 고령화 심화로 인한 농업 인력 감소로 농업 지속 가능성이 위협받고 있다. 국내 농가의 주요 형태인 시설 재배지에서는 병해에 의한 연쇄 피해가 발생할 수 있으므로 농업 생산성 증대를 위해 병해의 조기 진단이 필요하다. 본 논문에서는 병해의 조기 진단과 대처를 위해 YOLOv5를 이용한 딸기 병해 진단 모델을 제작, 데이터셋과 학습 세부사항에 변화를 주며 실험하였다. 실험 결과 데이터셋과 epochs 증량은 모델 성능에 영향을 주지만 임계점에 다다르면 성능 향상에 도움이 되지 않는 것을 알 수 있었다. 한편 학습한 모델 중 가장 좋은 성능을 가진 모델의 경우 F1 Score 0.98, mAP 0.99를 나타내 높은 정확도로 딸기의 병해 여부 진단이 가능하였다.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.49 no.2
• /
• pp.131-139
• /
• 2022

This study aimed to evaluate the effectiveness of deep convolutional neural networks (CNNs) for diagnosis of interproximal caries in pediatric intraoral radiographs. A total of 500 intraoral radiographic images of first and second primary molars were used for the study. A CNN model (Resnet 50) was applied for the detection of proximal caries. The diagnostic accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curve, and area under ROC curve (AUC) were calculated on the test dataset. The diagnostic accuracy was 0.84, sensitivity was 0.74, and specificity was 0.94. The trained CNN algorithm achieved AUC of 0.86. The diagnostic CNN model for pediatric intraoral radiographs showed good performance with high accuracy. Deep learning can assist dentists in diagnosis of proximal caries lesions in pediatric intraoral radiographs.

• Journal of The Korean Association For Science Education
• /
• v.27 no.4
• /
• pp.346-353
• /
• 2007

In order to be efficient in teaching, a teacher should understand the current learner's level through diagnostic evaluation. This study has examined the major issues arising from the noble diagnostic assessment tool based on the theory of knowledge space. The knowledge state analysis method is actualizing the theory of knowledge space for practical use. The knowledge state analysis method is very advantageous when a certain group or individual student's knowledge structure is analyzed especially for strong hierarchical subjects such as mathematics, physics, chemistry, etc. Students' knowledge state helps design an efficient teaching plan by referring their hierarchical knowledge structure. The knowledge state analysis method can be enhanced by computer due to fast data processing. In addition, each student's knowledge can be improved effectively through individualistic feedback depending on individualized knowledge structure. In this study, we have developed a diagnostic assessment test for measuring student's learning outcome which is unattainable from the conventional examination. The diagnostic assessment test was administered to middle school students and analyzed by the knowledge state analysis method. The analyzed results show that students' knowledge structure after learning found to be more structured and well-defined than the knowledge structure before the learning.

• Digital Contents
• /
• no.5 s.144
• /
• pp.68-69
• /
• 2005

습자가 원하는 학습자원을 컴퓨터가 스스로 찾아내서 학습자에게 전달해주고, 더 나아가 새로운 지식까지 추론해서 제공해 줄 수는 없을까? 의미의 웹이라 불리고 있는 시멘틱 웹(Semantic Web)은 의미적으로 연결돼 있는 학습 정보를 컴퓨터가 의미를 이해해서 학습자가 원하는, 학습자 수준에 맞는 정보를 제공해주고 더 나아가 지식까지도 추론해서 학습자에게 가장 적합한 형태로 전달해 줄 수 있는 강력한 메커니즘으로 부각되고 있다. 이에 필자는 살이 있는 e러닝이 되기 위해서는 시멘틱 웹과의 통합이 필요하다고 생각해 2회에 걸쳐 시멘틱 웹과, 시멘틱 웹을 e러닝에 어떻게 적용할 것인지에 대해 진단해 보고자 한다.

• The Journal of Korean Association of Computer Education
• /
• v.15 no.6
• /
• pp.33-41
• /
• 2012

This paper presents the online assessment questions management system and method. The proposed system consists of a database to store learner information and zone-specific items grouped by difficulty and item bank. This database includes: an item selection department and authoring assessment to select questions about a particular learner or specific learning item. In this paper, we propose: an item bank database which stores online output assessments; and an online test department to collect and sort learner evaluation data and answer selection order for online tests, click statistics, response time, and analysis unit response patterns department by analyzing the data collected by the online learners' test assessment, learners' level and ability, the diagnosis and assessment of report propensity. The proposed system will diagnose and effectively evaluate the learner's learning levels and learning ability by: answer selection order, number of clicks, and response time reflected in the results of the learners' evaluations.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.4
• /
• pp.551-559
• /
• 2017

In this paper, a real-time training method to improve the performance of bearing fault diagnosis. The traditional bearing fault diagnosis cannot classify a condition which is not trained by the classifier. The proposed 4-step method trains and recognizes new condition in real-time, thereby it can classify the condition accurately. In the first step, we calculate the maximum distance value for each class by calculating a Euclidean distance between a feature vector of each class and a centroid of the corresponding class in the training information. In the second step, we calculate a Euclidean distance between a feature vector of new acquired data and a centroid of each class, and then compare with the allowed maximum distance of each class. In the third step, if the distance between a feature vector of new acquired data and a centroid of each class is larger than the allowed maximum distance of each class, we define that it is data of new condition and increase count of new condition. In the last step, if the count of new condition is over 10, newly acquired 10 data are assigned as a new class and then conduct re-training the classifier. To verify the performance of the proposed method, bearing fault data from a rotating machine was utilized.

• The Journal of Information Technology
• /
• v.2 no.2
• /
• pp.13-22
• /
• 1999

In this paper, we purpose automatic diagnosis in online, as the fundamental study to diagnose the partial discharge mechanism and to predict the lifetime by introduction a neural network. In the proposed method, we use AE(acoustic emission) sensing system and calculate a quantitative statistic parameter by pulse number and amplitude. Using statically parameters such as the center of gravity(G) and the gradient if the discharge distribute(C), we analyzed the early stage and the middle stage. the quantitative statistic parameters are learned by a neural network. The diagnosis of insulation degradation and a lifetime prediction by the early stage time are achieved. On the basis of revealed excellent diagnosis ability through the neural network learning for the patterns during degradation, it was proved that the neural network is appropriate for degradation diagnosis and lifetime prediction in partial discharge.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.31-38
• /
• 2022

This paper applies an expert independent unsupervised neural network learning-based multivariate time series data analysis model, MSCRED(Multi-Scale Convolutional Recurrent Encoder-Decoder), and to overcome the limitation, because the MCRED is based on Auto-encoder model, that train data must not to be contaminated, by using learning data sampling technique, called Subset Sampling Validation. By using the vibration data of power plant equipment that has been labeled, the classification performance of MSCRED is evaluated with the Anomaly Score in many cases, 1) the abnormal data is mixed with the training data 2) when the abnormal data is removed from the training data in case 1. Through this, this paper presents an expert-independent anomaly diagnosis framework that is strong against error data, and presents a concise and accurate solution in various fields of multivariate time series data.

• Information and Communications Magazine
• /
• v.31 no.12
• /
• pp.73-80
• /
• 2014

본고에서는 교육 분야에서 다양한 데이터를 수집 및 분석하여 개인화된 학습 서비스를 제공하려는 학습 분석(Learning Analytics) 서비스의 개념과 앞으로 기대되는 유즈케이스를 소개한다. 국제적으로 주목 받고 있는 학습 분석 기술은 현재 개념화 수준에 머물러 있지만, 글로벌 기업들이 주축이 된 민간단체에서는 데이터 수집체계와 같은 구체적인 구현 방법에 대한 논의도 추진되고 있어서 관련 현황에 대한 진단도 해본다. 특히 국제 표준화 기구와 단체를 통해 추진되고 있는 내용을 중심으로 소개한다. 다양한 데이터 응용 기술을 융합해서 기대할 수 있는 학습 분석 서비스 모형을 제시하면서 관련 정책과 제품개발에 기여할 수 있을 것으로 기대한다.