• Korean Journal of Radiology
• /
• 제20권5호
• /
• pp.749-758
• /
• 2019

Objective: To investigate whether a computer-aided diagnosis (CAD) system based on a deep learning framework (deep learning-based CAD) improves the diagnostic performance of radiologists in differentiating between malignant and benign masses on breast ultrasound (US). Materials and Methods: B-mode US images were prospectively obtained for 253 breast masses (173 benign, 80 malignant) in 226 consecutive patients. Breast mass US findings were retrospectively analyzed by deep learning-based CAD and four radiologists. In predicting malignancy, the CAD results were dichotomized (possibly benign vs. possibly malignant). The radiologists independently assessed Breast Imaging Reporting and Data System final assessments for two datasets (US images alone or with CAD). For each dataset, the radiologists' final assessments were classified as positive (category 4a or higher) and negative (category 3 or lower). The diagnostic performances of the radiologists for the two datasets (US alone vs. US with CAD) were compared Results: When the CAD results were added to the US images, the radiologists showed significant improvement in specificity (range of all radiologists for US alone vs. US with CAD: 72.8-92.5% vs. 82.1-93.1%; p < 0.001), accuracy (77.9-88.9% vs. 86.2-90.9%; p = 0.038), and positive predictive value (PPV) (60.2-83.3% vs. 70.4-85.2%; p = 0.001). However, there were no significant changes in sensitivity (81.3-88.8% vs. 86.3-95.0%; p = 0.120) and negative predictive value (91.4-93.5% vs. 92.9-97.3%; p = 0.259). Conclusion: Deep learning-based CAD could improve radiologists' diagnostic performance by increasing their specificity, accuracy, and PPV in differentiating between malignant and benign masses on breast US.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권4호
• /
• pp.2038-2057
• /
• 2017

Gaussian learning is a new technology in the computational intelligence area. However, this technology weakens the learning ability of a particle swarm and achieves a lack of diversity. Thus, this paper proposes a vector Gaussian learning strategy and presents an effective approach, named particle swarm optimization based on vector Gaussian learning. The experiments show that the algorithm is more close to the optimal solution and the better search efficiency after we use vector Gaussian learning strategy. The strategy adopts vector Gaussian learning to generate the Gaussian solution of a swarm's optimal location, increases the learning ability of the swarm's optimal location, and maintains the diversity of the swarm. The method divides the states into normal and premature states by analyzing the state threshold of the swarm. If the swarm is in the premature category, the algorithm adopts an inertia weight strategy that decreases linearly in addition to vector Gaussian learning; otherwise, it uses a fixed inertia weight strategy. Experiments are conducted on eight well-known benchmark functions to verify the performance of the new approach. The results demonstrate promising performance of the new method in terms of convergence velocity and precision, with an improved ability to escape from a local optimum.

• 지능정보연구
• /
• 제20권3호
• /
• pp.77-92
• /
• 2014

텍스트에 대한 사용자의 접근성을 향상시키기 위해, 이들 문서는 정해진 기준에 따라 카테고리로 분류되어 제공되고 있다. 과거에는 카테고리 분류 작업이 수작업으로 수행되었지만, 문서 작성자에게 분류를 맡기는 경우 분류 정확성을 보장할 수 없고 관리자가 모든 분류를 담당하는 경우 많은 시간과 비용이 소요된다는 어려움이 있었다. 이러한 한계를 극복하기 위해 카테고리를 자동으로 식별할 수 있는 문서 분류 기법에 대한 연구가 활발하게 수행되었다. 하지만 대부분의 문서 분류 기법은 각 문서가 하나의 카테고리에만 속하는 경우를 가정하고 있기 때문에, 하나의 문서가 다양한 주제를 갖는 실제 상황과 부합하지 않는다는 한계를 갖는다. 이를 보완하기 위해 최근 문서의 다중 카테고리 식별을 위한 연구가 일부 수행되었으나, 이들 연구는 대부분 이미 다중 카테고리가 부여되어 있는 문서에 대한 학습을 통해 분류 규칙을 생성하므로 단일 카테고리만 부여되어 있는 기존 문서의 다중 카테고리 식별에는 적용할 수 없다는 제약을 갖는다. 따라서 본 연구에서는 이러한 제약을 극복하기 위해, 카테고리, 토픽, 문서간 관계 분석을 통해 단일 카테고리를 갖는 문서로부터 추가 주제를 발굴하여 이를 다중 카테고리로 자동 확장시킬 수 있는 방법론을 제안하였다. 실험 결과 원 카테고리가 식별된 총 24,000건의 문서 중 23,089건에 대해 카테고리를 확장시킬 수 있었다. 또한 정확도 분석에서 카테고리의 특성에 따라 카테고리 분류 정확도가 상이하게 나타나는 현상을 발견하였다. 본 연구는 단일 카테고리로 분류된 문서에 대해 다중 카테고리를 추가로 식별하여 부여함으로써, 규칙 학습 과정에서 다중 카테고리가 부여된 문서를 필요로 하는 기존 다중 카테고리 문서 분류 알고리즘의 활용성을 매우 향상시킬 수 있을 것으로 기대한다.

• 대한지구과학교육학회지
• /
• 제5권1호
• /
• pp.114-123
• /
• 2012

This study investigates the effect of environmental club activities of the creative experiential learning activities on the improvement of students' Environmental Literacy and Pro-environmental Behavior. For that purpose, This study divided 24 students at Jinhae U Elementary School into test groups of 12 students and Control groups of 12 students. Test groups participated in the 'GomGomi' environmental Club for sixteen times from March 2011 to December 2011. Conclusions of this study include; First, we found that the environmental club activities is effective to improve the students' environmental Literacy overall. Specifically, the Environmental Club Activities was effective in the sub-catagories of environmental Literacy such as 'rights of nature' and 'eco-crisis'. However, there was no significant change in the sub-category of 'human exemptionalism' Second, we found that the environmental club activities is effective to improve students' pro-environmental behavior overall. Specifically, the environmental club activities was effective in the sub-domains of Pro-environmental Behavior such as 'cognitive domain' and 'affective domain'. However, there was no significant change in the sub-domains of 'behavioral domain'. Summary, the environmental club activities was shown to be effective for improving their environmental literacy and pro-environmental behavior; This study implies that the environmental club activities of the creative experiential learning activities would be a effective tool to help students to improve their environmental literacy and pro-environmental behavior.

• 한국지능시스템학회논문지
• /
• 제19권1호
• /
• pp.54-61
• /
• 2009

본 논문에서는 비전 패턴인식 알고리즘인 시공간적 계층 메모리 학습 알고리즘을 이용한 새로운 근전도 패턴인식 방법을 제시한다. 효율적인 근전도 신호의 학습과 분류를 위하여 단순화된 2 레벨의 공간적 집합, 시간적 집합, 그리고 관리 맵퍼를 이용한 수정된 시공간적 계층 메모리 학습 알고리즘을 제안한다. 인식 성능을 향상시키기 위해서 관리 맵퍼 학습뿐만 아니라 시간적 집합 학습에도 카테고리 정보를 사용한다. 실험을 통하여 열 가지 손동작이 성공적으로 인식됨을 검증한다.

• International Journal of Computer Science & Network Security
• /
• 제22권4호
• /
• pp.254-260
• /
• 2022

The purpose of this study was to identify the obstacles facing gifted students with learning disabilities (GSLDs) from the point of view of their teachers in the Makkah region and to find suggested solutions to overcome these obstacles. The study covered Makkah, Jeddah and Taif and used semi-structured interviews which included open-ended questions. The study findings indicated that there were several educational obstacles including the absence of adapted courses or specialized teachers for GSLDs category and the insufficient time for the students to express their talents. According to the findings, there were also societal obstacles including the society's failure to expect the presence of talents along with disabilities, or its denial or rejection of their talents in addition to ridiculing them. The findings also confirmed the existence of administrative obstacles including the lack of community partnership. There were also family obstacles such as the family's lack of encouragement for the students, and ignorance of the nature of GSLDs. The study came up with a number of solutions and proposals related to awareness, educational institutions, education and competitions for talented people with learning disabilities.

• 문화기술의 융합
• /
• 제8권5호
• /
• pp.507-513
• /
• 2022

본 연구는 영화 산업에서 온라인 시장인 IPTV의 VOD 이용 건수 예측 모델을 개발하였다. 한국영화진흥위원회에서 수집한 2017년부터 2021년까지 VOD 이용건수 데이터를 활용하여 머신러닝 기반 예측모델을 구축했다. 문헌조사와 군집분석을 통하여 오프라인 시장과 온라인 시장의 차이를 밝히고, VOD 이용 건수의 새로운 범주를 제안한다. 머신러닝 기반의 VOD 이용 건수 예측 모델 개발을 통해 IPTV 기업들의 의사결정 지원 뿐 아니라 마케팅 전략 수립을 돕는 것을 목적으로 한다.

• The Plant Pathology Journal
• /
• 제39권4호
• /
• pp.319-334
• /
• 2023

Plant disease is an important factor affecting crop yield. With various types and complex conditions, plant diseases cause serious economic losses, as well as modern agriculture constraints. Hence, rapid, accurate, and early identification of crop diseases is of great significance. Recent developments in deep learning, especially convolutional neural network (CNN), have shown impressive performance in plant disease classification. However, most of the existing datasets for plant disease classification are a single background environment rather than a real field environment. In addition, the classification can only obtain the category of a single disease and fail to obtain the location of multiple different diseases, which limits the practical application. Therefore, the object detection method based on CNN can overcome these shortcomings and has broad application prospects. In this study, an annotated apple leaf disease dataset in a real field environment was first constructed to compensate for the lack of existing datasets. Moreover, the Faster R-CNN and YOLOv3 architectures were trained to detect apple leaf diseases in our dataset. Finally, comparative experiments were conducted and a variety of evaluation indicators were analyzed. The experimental results demonstrate that deep learning algorithms represented by YOLOv3 and Faster R-CNN are feasible for plant disease detection and have their own strong points and weaknesses.

• Educational Technology International
• /
• 제11권2호
• /
• pp.43-75
• /
• 2010

The purpose of this study is to identify patterns of learner behaviors, conflicting and facilitating factors during collaborative work in an online learning community(OLC). This study further seeks to investigate the difference of learner behaviors between high- and low-performing groups, and conflicting and facilitating factors. The online postings from four groups(19 students) in the spring semester(study 1) and six groups(24 students) in the fall semester(study 2) were analyzed. A coding scheme was generated based on constant comparison using the qualitative data analysis tool, NVivo. The analysis identified 7 categories of learner behaviors in both studies. Among the seven categories, information seeking and co-construction were most frequently observed in both studies. One evident difference between the high- and low-performing groups was that the high-performing groups revealed more incidents of learner behaviors in both studies. In addition, six categories of conflicting factors and five categories of facilitating factors were emerged in both studies. The inefficiency of work category was one of the most frequently observed categories in both studies. Interestingly, the high-performing groups showed more incidents of conflicting factors than the low-performing groups. This study revealed two different types of conflicting factors and there is a need for different moderating strategies depending on its type. Based on the results of the study, effective design strategies for an OLC to facilitate active learning were suggested.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.183-189
• /
• 2023

A stroke is a medical disease where a blood vessel in the brain ruptures, causes damage to the brain. If the flow of blood and different nutrients to the brain is intermittent, symptoms may occur. Stroke is other reason for loss of life and widespread disorder. The prevalence of stroke is high in growing countries, with ischemic stroke being the high usual category. Many of the forewarning signs of stroke can be recognized the seriousness of a stroke can be reduced. Most of the earlier stroke detections and prediction models uses image examination tools like CT (Computed Tomography) scan or MRI (Magnetic Resonance Imaging) which are costly and difficult to use for actual-time recognition. Machine learning (ML) is a part of artificial intelligence (AI) that makes software applications to gain the exact accuracy to predict the end results not having to be directly involved to get the work done. In recent times ML algorithms have gained lot of attention due to their accurate results in medical fields. Hence in this work, Stroke disease identification system by using Machine Learning algorithm is presented. The ML algorithm used in this work is Artificial Neural Network (ANN). The result analysis of presented ML algorithm is compared with different ML algorithms. The performance of the presented approach is compared to find the better algorithm for stroke identification.