• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.502-503
• /
• 2019

In the existing system that recommends through review data analysis, it does not reflect personal preference details such as user's characteristics or product purchase tastes, in this paper, we propose a system that provides customized recommendation information to various users by selecting the criterion that the user thinks most importantly when searching for the product and purchasing the product, and analyzing it. This is because the user's personal preference is reflected by arranging the product list based on the criterion that the user occupies the largest portion of the product purchase, so that it is more efficient than the recommendation through the recommendation system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.338-340
• /
• 2019

Recently, the speed and size of data accumulation are increasing due to the development of networks. There are many difficulties in classifying these data. One of the difficulties is the difficulty of labeling. Labeling is usually done by people, but it is very difficult for everyone to understand the data in the same way and it is very difficult to label them on the same basis. In order to solve this problem, we implemented GAN to generate new image based on input image and to learn input data indirectly by using it for learning. This suggests that the accuracy of classification can be increased by increasing the number of learning data.

• The Korean Journal of Mycology
• /
• v.49 no.2
• /
• pp.243-248
• /
• 2021

In this study, screening of potent non-pathogenic wild yeast with high anti-dementia butyrylcholinesterase (BChE) inhibitory activity and production condition of a BChE inhibitor were described. Among 36 non-pathogenic wild yeasts, Saccharomyces cerevisiae WJSL 0113 showed the highest BChE inhibitory activity of 85.2%. The specific BChE inhibitor was maximally produced when S. cerevisiae WJSL 0113 was cultured at 30℃ for 48 h in a yeast extract-peptone-dextrose medium.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.252-254
• /
• 2021

According to the Institute for Occupational Safety and Health, the number of industrial injuries in 2019 was 109,242, an increase of 6.8% from 2018. In this situation, the government and companies are discussing the development of core technologies for preventing safety accidents on site based on ICT in the field of construction and construction. In these fields, technologies using computer vision and artificial intelligence have recently been widely used. In this paper, we built training data for safety management of factory workers and trained a model based on YOLOv4. It is believed that this can be used as an initial study to predict the risk situation of workers in factories.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.320-322
• /
• 2021

With the development of wireless Internet technology and the expansion of the game market, various forms of games are being developed that are mounted on various platforms, including mobile platforms. In this environment, ensuring the immersion of the game user's perspective will secure the game's competitiveness, so it is necessary to increase the immersion by satisfying each area presented by the Human Computer Interaction (HCI) theory. To this end, this high school implemented downhill games and experimented with kiosks by applying an interface that recognizes the human body's movements as a way to secure freedom and immersion of game users.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.258-259
• /
• 2021

Forest lodge are divided into forest resources, auxiliary facilities, and users, and are constantly being used along with well-being culture. In addition, attention to the safety of users is also required, and this study aims to study how users evacuate within a short time (golden time) in situations of natural disasters that may occur in forests. In order to search for the current location of the user and find the best evacuation route, 3D scans of the entire forest lodge(forest resources, auxiliary facilities, etc.) are performed, and the optimal trajectory to the evacuation site is found through recognition of the current location. It is believed that it is possible to provide a quick evacuation guide through a mobile device with gps.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.338-340
• /
• 2021

Among the recent natural disasters, flood damage is concentrated in the summer, leading to casualties or property damage due to the typhoon's rainy season. Much research is being done to reduce this damage. In this paper, we design and implement a system that detects paths and provides them to users using shortest-distance algorithms in the event of such sudden flooding. The proposed system uses QGIS to detect paths using topographical data that generated tracks and pointers. In addition, other shortest paths are detected and provided to users when sudden flooding occurs during evacuation to routes detected through scenarios. Therefore, it is assumed that the proposed system will allow users to safely evacuate from the risk of future disaster safety accidents.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.222-224
• /
• 2022

Deep learning-based computer vision applications in the field of object segmentation take a transfer learning method using hyperparameters and models pretrained and distributed by STOA techniques to improve performance. Custom datasets used in this process require a lot of resources, such as time and labeling, in labeling tasks to generate Ground Truth information. In this paper, we present an automatic labeling construction method using object segmentation techniques so that resources such as time and labeling can be used less to build custom datasets used in deep learning neural networks.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.582-584
• /
• 2014

Conventional Web crawlers are reducing overhead burden on the server to ensure the integrity of data optimization measures have been continuously developed. The amount of data growing exponentially faster among those data, then the data needs to be collected should be used to the modern web crawler is the indispensable presence. In this paper, suggested that the existing Web crawler and Web crawler approach efficiency comparison and analysis. In addition, based on the results, compared to suggest an optimized technique, Web crawlers, data collection cycle dynamically reduces the overhead of the server system was designed for. This is a Web crawler approach will be utilized in the field of the search system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.684-686
• /
• 2014

It is difficult to effectively search and data management due to the increasing number of web is now. While Semantic Web technologies and the development of next-generation wepin this as a way to overcome them, and monopolize the domestic utilization is not overwhelming introduction to the Semantic Web technology is being used in existing search engines. This causes the development of the Semantic Web is becoming slower, and reluctant to use the Semantic Web users who use search engines as well. In this paper, compared to the currently used web and the next generation of the web, and why utilization is low compared to the search engine you are using an existing Web technology that uses the Semantic Web technology is a search engine, what research was that the inefficient because, as a RDF-based Semantic suggest how to improve the efficiency solved by designing the web.