• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.205-211
• /
• 2024

The purpose of this study is to implement an agent that intelligently performs tracking and movement through reinforcement learning using the Unity and ML-Agents. In this study, we conducted an experiment to compare the learning performance between training one agent in a single learning simulation environment and parallel training of several agents simultaneously in a multi-learning simulation environment. From the experimental results, we could be confirmed that the parallel training method is about 4.9 times faster than the single training method in terms of learning speed, and more stable and effective learning occurs in terms of learning stability.

• Journal of Internet Computing and Services
• /
• v.3 no.3
• /
• pp.11-17
• /
• 2002

The web-based distance education has some problems in managing students despite the merit of getting over the problem of time and space compared with the classroom education or residential course. These problems make the web-based distance learners difficult in achieving the purpose and the standard of goal. And they drive the manager in charge of the web-based distance education into worries. The manager of the distance education should keep on monitoring students' participation and response, and then give the appropriate feedback to the students. But, the existing distance education system neglects to support efficient management function, for it puts emphasis on the activity of teaching and learning. Therefore, the purposes of this study are to extract the factors that are necessary to manage the distance education for training teachers, to realize the effective management system which can offer a proper feedback in the distance education for training teachers through the application of those factors to the training program, to make it possible to change the management system flexibly according to the curriculum and the computing setting.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1486-1501
• /
• 2021

The removal of accessories from the face is one of the essential pre-processing stages in the field of face recognition. However, despite its importance, a robust solution has not yet been provided. This paper proposes a network and dataset construction methodology to remove only the glasses from facial images effectively. To obtain an image with the glasses removed from an image with glasses by the supervised learning method, a network that converts them and a set of paired data for training is required. To this end, we created a large number of synthetic images of glasses being worn using facial attribute transformation networks. We adopted the conditional GAN (cGAN) frameworks for training. The trained network converts the in-the-wild face image with glasses into an image without glasses and operates stably even in situations wherein the faces are of diverse races and ages and having different styles of glasses.

• Journal of Internet Computing and Services
• /
• v.23 no.1
• /
• pp.69-76
• /
• 2022

Education systems have made efforts to prepare students by providing technical and nontechnical courses. With video games, however, there is the potential to develop dedicated metaverses that can help teach soft skills even during casual pastimes. The research conducted will propose a set of design practices for metaverse and game development to promote soft skills. While there are many soft skills people can acquire, this paper will focus on certain aspects based on specific games and studies. There will be some information collected from the information to support the design model and arguments. This paper will provide developers with a starting point for imaginative game creation and impart users with soft skills to assist in their professions and social life.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3211-3229
• /
• 2022

To solve the problems existing in the process of Weibo disaster rumor recognition, such as lack of corpus, poor text standardization, difficult to learn semantic information, and simple semantic features of disaster rumor text, this paper takes Sina Weibo as the data source, constructs a dataset for Weibo disaster rumor recognition, and proposes a deep learning model BERT_AT_Stacked LSTM for Weibo disaster rumor recognition. First, add adversarial disturbance to the embedding vector of each word to generate adversarial samples to enhance the features of rumor text, and carry out adversarial training to solve the problem that the text features of disaster rumors are relatively single. Second, the BERT part obtains the word-level semantic information of each Weibo text and generates a hidden vector containing sentence-level feature information. Finally, the hidden complex semantic information of poorly-regulated Weibo texts is learned using a Stacked Long Short-Term Memory (Stacked LSTM) structure. The experimental results show that, compared with other comparative models, the model in this paper has more advantages in recognizing disaster rumors on Weibo, with an F1_Socre of 97.48%, and has been tested on an open general domain dataset, with an F1_Score of 94.59%, indicating that the model has better generalization.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.7
• /
• pp.2752-2768
• /
• 2020

Phishing websites can have devastating effects on governmental, financial, and social services, as well as on individual privacy. Currently, many phishing detection solutions are evaluated using small datasets and, thus, are prone to sampling issues, such as representing legitimate websites by only high-ranking websites, which could make their evaluation less relevant in practice. Phishing detection solutions which depend only on the URL are attractive, as they can be used in limited systems, such as with firewalls. In this paper, we present a URL-only phishing detection solution based on a convolutional neural network (CNN) model. The proposed CNN takes the URL as the input, rather than using predetermined features such as URL length. For training and evaluation, we have collected over two million URLs in a massive URL phishing detection (MUPD) dataset. We split MUPD into training, validation and testing datasets. The proposed CNN achieves approximately 96% accuracy on the testing dataset; this accuracy is achieved with URL schemes (such as HTTP and HTTPS) removed from the URL. Our proposed solution achieved better accuracy compared to an existing state-of-the-art URL-only model on a published dataset. Finally, the results of our experiment suggest keeping the CNN up-to-date for better results in practice.

• Journal of Internet Computing and Services
• /
• v.23 no.1
• /
• pp.95-103
• /
• 2022

The goal of education and training in military is to foster strong combatants who can fight and defeat enemies. The Korean military is deeply aware of the importance of education & training, and has been introducing various advanced training systems so far. Despite these efforts, however, the military environment to maintain and strengthen the level of training is becoming increasingly difficult. In this study, it was conducted on the effectiveness analysis and utilization of the game system for military education & training through literature review. As a result of literature analysis, the introduction of the game system could be expected to have various effects throughout the cognitive and behavioral areas. Based on this effect analysis, the concept and shape of game system operation for each purpose were derived, and an improved plan using the game system was proposed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.5
• /
• pp.163-173
• /
• 2020

In this study, on-site corporate instructors participated as student-cum-workers in a degree-linked high skills Meister course to improve job competency and practical ability as proposed in the Work-Study Career Vision. Evaluation questions were then developed and their validity was verified by assessing satisfaction related to expected goals in enhancing advanced training guidance and competency as an evaluator. Satisfaction assessment was conducted based on training preparation, training implementation, training effectiveness and training administration. The Delphi Method was adopted and a total of 48 items were developed in 6 categories under 4 main areas. There were 7 evaluation items on the satisfaction of training course development under training preparation, 21 evaluation items related to the satisfaction of Off-JT and OJT courses under training implementation, 16 evaluation items related to the satisfaction of increased competency as an on-site corporate instructor and the satisfaction of enhanced practical skills and skills application at work under training effectiveness, as well as 6 evaluation items to assess satisfaction with administrative support under training administration. The final conformity assessment was conducted based on the stability, content validity ratio, consensus and convergence indicators of the developed items. Results of this study do not only apply to quality management of the high skills Meister course which is being promoted as a pilot project for work-study programs, but also serves as a rationale that may be considered as a basic research tool in the collection of various opinions to derive overall system improvement factors for the work-study high skills Meister course.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.37-46
• /
• 2021

Recently, social technologies have been created to solve problems from businesses for the establishment of generational solidarity ecosystem in terms of employment, residential space, network and social capital, age, cognitive and environmental aspects. This is senior-friendly healthcare business system aimed at meeting the senior needs for health life to enjoy active consumption culture life even after retirement, becoming a catalyst for minimizing generational conflicts, preventing the cognitive and physical deterioration of seniority in the areas of life healthcare, fitness and well-aging, and expanding into systems necessary for seniority self-reliance. We would like to draw up the development and requirements of the concept of the service platform for the study of collective characteristics for generation solidarity with senior class and the establishment of a customized senior health life system for generation solidarity. This system is characterized by a platform that can prevent the decline of seniors' cognitive and physical functions and enhance emotional stability. It is significant in providing feedback on the risk perception index, fall index, and prevention training index information to the child through the analysis and extraction of the senior health index for risk perception, fall probability, and fall prevention.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.1118-1133
• /
• 2017

Computer vision-based human activity recognition (HAR) has become very famous these days due to its applications in various fields such as smart home healthcare for elderly people. A video-based activity recognition system basically has many goals such as to react based on people's behavior that allows the systems to proactively assist them with their tasks. A novel approach is proposed in this work for depth video based human activity recognition using joint-based motion features of depth body shapes and Deep Belief Network (DBN). From depth video, different body parts of human activities are segmented first by means of a trained random forest. The motion features representing the magnitude and direction of each joint in next frame are extracted. Finally, the features are applied for training a DBN to be used for recognition later. The proposed HAR approach showed superior performance over conventional approaches on private and public datasets, indicating a prominent approach for practical applications in smartly controlled environments.