• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.612-628
• /
• 2020

This paper presents a new approach for detecting abnormal behaviors in complex surveillance scenes where anomalies are subtle and difficult to distinguish due to the intricate correlations among multiple objects' behaviors. Specifically, a cascaded probabilistic topic model was put forward for learning the spatial context of local behavior and the temporal context of global behavior in two different stages. In the first stage of topic modeling, unlike the existing approaches using either optical flows or complete trajectories, spatio-temporal correlations between the trajectory fragments in video clips were modeled by the latent Dirichlet allocation (LDA) topic model based on Markov random fields to obtain the spatial context of local behavior in each video clip. The local behavior topic categories were then obtained by exploiting the spectral clustering algorithm. Based on the construction of a dictionary through the process of local behavior topic clustering, the second phase of the LDA topic model learns the correlations of global behaviors and temporal context. In particular, an abnormal behavior recognition method was developed based on the learned spatio-temporal context of behaviors. The specific identification method adopts a top-down strategy and consists of two stages: anomaly recognition of video clip and anomalous behavior recognition within each video clip. Evaluation was performed using the validity of spatio-temporal context learning for local behavior topics and abnormal behavior recognition. Furthermore, the performance of the proposed approach in abnormal behavior recognition improved effectively and significantly in complex surveillance scenes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.133-139
• /
• 2024

In response to the challenges posed by a super-aged society, local authorities are conducting educational programs on smartphone usage tailored for the elderly. However, obstacles such as the limitations of one-to-many education and suboptimal learning outcomes for the elderly have hindered the efficacy of smartphone education. This study suggests an educational service intended for direct application in offline settings, considering the identified problems. Through the utilization of generative AI, the proposed app identifies specific challenges encountered by users during actual smartphone use, offering personalized exercises to facilitate customized and repetitive learning experiences for individual users. When integrated with existing local government education initiatives, this app is anticipated to enhance the efficiency of smartphone education by providing personalized, one-on-one training that is efficient in terms of time and content.

• Hwankyungkyoyuk
• /
• v.9 no.1
• /
• pp.39-57
• /
• 1996

All kinds of environmental problems are related to each local and geographical environment. For this reason, it is necessary for schools in each region to provide environmental education which suits the geographical character of their particular region. In order to provide solutions to the environmental problems of each school's geographic region, the goal of this research is as follows: 1. We can make students realize the relationship between the human race and the environment by teaching according to the environmental conditions in each local area. 2. By teaching students about the problems in their own local environment, we can increase their concern about the state of their local surroundings. 3. When teaching about the environment, it is useful to use educational material which suits the character of each local region. 4. Students' interest in environmental preservation can be aroused through extracurricular environmental activities. The ares concerned are Chonnam and Kwangju City, which are divided into urban, industrial, rural, coastal, and mountainous areas. The conclusion about considering environmental education in environmental school social studies is as follows: 1. Kwangju and Chonnam should be divided into five sections, each with similar geographical environments. This will be an improvement over the old uniform approach to environmental studies in which all regions were treated as being the same each region will now receive special attention. 2. It is necessary to maximize the efficient use of the Environmental Education Building. When Media, environmental data and special materials for environmental education are used effectively, teachers can lead class effectively and students will be more interested in the class. 3. We can detect the cause of pollution, increase interest in the environment and easily solve environmental problems by collecting and displaying environmental educational materials. 4. An environmental education corner could boost students' interest in environmental problems and could act as a kind of bridge between theoretical and practical education. 5. Media and environmental data must be specialized according to the geographic character of each region. In this way, we can expect to improve the quality of environmental education over the simplistic environmental education of previous years. 6. Students will become interested in the problems of the region in which they live through social studies, and primarily through the environmental curriculum. 7. We can prevent learning deficiencies by making a consistent teaching plan. The teaching and learning methods will be improved and the teachers will be proud of what they teach. 8. The purpose of the Education Procedure Content Analysis is to make teaching and learning concise and easy by systematizing environmental and related subjects. This can be done by adding an environmental unit to the geographic section of social studies. 9. Citizens' interest in their own residential environment can be increased through action by sustaining environmental preservation movements to local region people.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.5
• /
• pp.21-30
• /
• 2023

Accurate segmentation of the kidney tumor is necessary to identify shape, location and safety margin of tumor in abdominal CT images for surgical planning before renal partial nephrectomy. However, kidney tumor segmentation is challenging task due to the various sizes and locations of the tumor for each patient and signal intensity similarity to surrounding organs such as intestine and spleen. In this paper, we propose a semi-supervised learning-based mean teacher network that utilizes both labeled and unlabeled data using a kidney local guided map including kidney local information to segment small-sized kidney tumors occurring at various locations in the kidney, and analyze the performance according to the kidney tumor size. As a result of the study, the proposed method showed an F1-score of 75.24% by considering local information of the kidney using a kidney local guide map to locate the tumor existing around the kidney. In particular, under-segmentation of small-sized tumors which are difficult to segment was improved, and showed a 13.9%p higher F1-score even though it used a smaller amount of labeled data than nnU-Net.

• Journal of the Korean earth science society
• /
• v.31 no.5
• /
• pp.500-520
• /
• 2010

The purposes of this case study were (1) to explore one experienced teacher's views on Earth Systems Education and (2) to describe and document the characteristics of the Earth Systems Education (ESE) curriculum provided by an exemplary middle school science teacher, Dr. J. All the essential pieces of evidence were collected from observations, interviews with the experienced teacher and his eighth grade students, informal conversations, document analysis, and field notes. The $NUD^*IST$ for MS Windows was used for an initial data reduction process and to narrow down the focus of an analysis. All transcriptions and written documents were reviewed carefully and repeatedly to find rich evidence through inductive and content analysis. The findings revealed that ESE provided a conceptual focus and theme for organizing his school curriculum. The curriculum offered opportunities for students to learn relevant local topics and to connect the classroom learning to the real world. The curriculum also played an important role in developing students' value and appreciation of Earth systems and concern for the local environment. His instructional strategies were very compatible with recommendations from a constructivist theory. His major teaching methodology and strategies were hands-on learning, authentic activities-based learning, cooperative learning, project-based learning (e.g., mini-projects), and science field trips. With respect to his views about benefits and difficulties associated with ESE, the most important benefit was that the curriculum provided authentic-based, hands-on activities and made connections between students and everyday life experiences. In addition, he believed that it was not difficult to teach using ESE. However, the lack of time devoted to field trips and a lack of suitable resource materials were obstacles to the implementation of the curriculum. Implications for science education and future research are suggested.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.375-390
• /
• 2024

Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.

• 한국사회복지학회:학술대회논문집
• /
• 2004.04a
• /
• pp.535-559
• /
• 2004

While the endeavor to introduce the service activity in the college education is enlarged, the object of this study is to introduce the concept of service learning in social work education. The 'content centered service-learning' is focusing on the practical education, the liaison and participation in community and the promotion of civil awareness which are very important in social work education that focuses on the interest and participation in the human and social problem as practical study. This study contemplates the educational effect of the course of social work combined with service learning. This study verifies the self-efficacy, the altruism, the change of social responsibility of students and the effect by carrying out the course of social welfare and service activities side by side. To evaluate the effect of service learning, pre and post test and qualitative analysis in journal of service activity, the discussion during the class and the mid and final term evaluation were done. Through the service learning, students showed the promotion in self-esteem, the altruism, the social responsibility, the self insights as a social worker and were able to have the chance to consider their future job. From the view point of local centers, it was evaluated that the service activity of students was helpful to the institutions and the clients, and that the liaison between the community agency and the college and the experience of integration of service learning and the course of social welfare was helpful too. Based on the above mentioned results, this study provide some recommendations in implementation of service-learning as the teaching methodology of social work.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.3
• /
• pp.189-201
• /
• 2011

In this paper, we design an ontology which provides interoperability by integrating typical metadata specifications and defines concepts and semantic relations between concepts that are used to describe metadata for learning objects in university courses. And we organize a hierarchical structured e-Learning system for efficient retrieval of learning objects on many local storages that use different specifications to describe metadata and propose a query processing method based on inferences. The proposed e-Learning system can provide more accurate and satisfactory retrieval service by using the designed ontology because both learning objects that be directly connected to user queries and deduced learning objects that be semantically connected to them are retrieved.

• Journal of the Korean association of regional geographers
• /
• v.13 no.2
• /
• pp.220-233
• /
• 2007

In the context of region reconceptualized as a learning framework, using the Seoul region as an example, this paper focused on the debate and the condition of regional social studies curriculum in elementary schools, It means that regionalization of curriculum is the decentralization of power to develop and manage curriculum In that case, region is understood to he a resource used to connect the subject matter to children's experience, Furthermore, region is interpreted as Heimat, community and world around, Heimat is intended to be understood, loved and belonged to, Community is the resource for learning and the space of communication, participation and action, World around is the typical example to be searched, compared and explained by the concept and theory. On the base of the regionalization debate and the classification of learning a region, I analysed the local textbook, Social Studies Inquiry: Life in Seoul and suggested the framework of learning a region. In this paper, I tried to pick out the spatial demension of Seoul. It is classified into the space of experience, participation and inquiry, In conclusion, the region-specific materials and approach laying stress on the issue and topic in the geographical context of the region will determine that a proper regionalization occurs through not only methods hut also content.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.8
• /
• pp.393-400
• /
• 2016

Recently, there have been various researches on efficient and automatic analysis on urban environment methods that utilize the computer vision and machine learning technology. Among many new analyses, urban safety analysis has received a major attention. In order to predict more accurately on safety score and reflect the human visual perception, it is necessary to consider the generic and local information that are most important to human perception. In this paper, we use Double-column Convolutional Neural network consisting of generic and local columns for the prediction of urban safety. The input of generic and local column used re-sized and random cropped images from original images, respectively. In addition, a new learning method is proposed to solve the problem of over-fitting in a particular column in the learning process. For the performance comparison of our Double-column Convolutional Neural Network, we compare two Support Vector Regression and three Convolutional Neural Network models using Root Mean Square Error and correlation analysis. Our experimental results demonstrate that our Double-column Convolutional Neural Network model show the best performance with Root Mean Square Error of 0.7432 and Pearson/Spearman correlation coefficient of 0.853/0.840.