• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.475-478
• /
• 2004

The DCSU performs satellite payload data acquisition and storage functions before sending the data to the ground station. While imaging, the DCSU makes a file per each input channel and store into memory stack. For the successful imaging mission, proper DCSU mission parameters should be uploaded before the mission such as file name, file size, output channel for the download transmission and so on. This paper will describe the DCSU recording mechanism and some notices that might be helpful for the ground operators.

• The Journal of Information Technology
• /
• v.12 no.2
• /
• pp.49-55
• /
• 2009

Currently the Web, multiple files in the Client to upload to the server of the multi-upload method, files are stored in specific folders on the server, the file name field in DB(Database Field) stored in a way that the process has been widely utilized. Therefore, a minimum number of file upload as many file names in the Database and the Table should be created in the field, but also a variable number of files can be uploaded has no limits. In this paper we proposed the new method that files to be uploaded to the server, you can upload a limited number of files, nor due to upload files to reduce load on the database to improve the processing speed of a single field using a multi-method file uploads using PHP directly implemented

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.221-222
• /
• 2014

빠른 웹의 성장으로 대용량 데이터를 효과적으로 처리할 수 있는 플랫폼 기술에 대한 관심이 높아지고 있다. 특히, HDFS는 이상적인 분산 파일 시스템으로 각광받고 있으며 대용량 파일의 처리를 목적으로 개발되었다. 하지만, 실제 파일들의 집합에서 소용량 파일이 차지하는 비중은 높은 편이다. 많은 수의 소용량 파일은 HDFS 성능 감소에 치명적인 원인이 된다. 많은 수의 소용량 파일들이 HDFS에 저장된다면 NameNode의 메모리 소비량이 증가하게 되며 많은 수의 소용량 파일은 많은 수의 DataNode와 NameNode를 요구하므로 상대적으로 처리시간이 많이 소모된다. 따라서 본 논문에서는 HDFS에서 소용량 파일의 저장과 액세스 효율성을 향상시키기 위하여 나이브 베이지안 분류기 알고리즘을 적용한 파일 그룹화 시스템을 설계하였다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.596-599
• /
• 2022

심층 학습은 지속적으로 발전하고 있으며, 최근에는 실제 사용자에게 제공되는 애플리케이션까지 확장되고 있다. 특히 자연어처리 분야에서는 대용량 언어 말뭉치를 기반으로 한 언어모델이 등장하면서 사람보다 높은 성능을 보이는 시스템이 개발되었다. 그러나 언어모델은 높은 컴퓨팅 파워를 요구하기 때문에 독립적인 소형 디바이스에서 제공할 수 있는 서비스에 적용하기 힘들다. 예를 들어 스캐너에서 제공할 수 있는 파일명 자동 부착 서비스는 하드웨어의 컴퓨팅 파워가 제한적이기 때문에 언어모델을 적용하기 힘들다. 또한, 활용할 수 있는 공개 데이터가 많지 않기 때문에, 데이터 구축에도 높은 비용이 요구된다. 따라서 본 논문에서는 컴퓨팅 파워에 비교적 독립적이고 학습 데이터가 필요하지 않은 비지도 학습을 활용하여 파일명 자동 부착 서비스를 위한 파일명 추출 방법을 제안한다. 실험은 681건의 문서 OCR 결과에 정답을 부착하여 수행했으며, ROUGE-L 기준 0.3352의 성능을 보였다.

• Journal of Korean Library and Information Science Society
• /
• v.20
• /
• pp.285-315
• /
• 1993

Some conclusions can be derived from the study: 1. Person's name should be script by the one's mother tongue because of its uniqueness. 2. Japanese person's name should be script and pronounce their mother tongue for exchange and sharing of an academic information. 3. We can anticipate that Japanese language materials will be increase in near future. 4. The National Central Library which publish Korean National Bibliography must have to responsibility to lead other library. 5. The script of [Korean National bibliography] must contribute to standardization and national and Universal Bibliographic Control. 6. The area of education, newspaper, publishing are scripting Japanese person's name with script conversion schemes for Koreanization, devised by Ministry of Education. 7. The script of [Korean National Bibliography]'s name index can be used as authority file at selection of heading in library cataloging. 8. Most of libraries script Japanese person's name with Chinese character in Korean language pronunciation. 9. Korean Cataloging Rules (KCR) and Korean Machine Readable Cataloguing (KORMARC) description rules should be defined about the mother tongue script of Japanese person's name. 10. It is desirable to increase of credit of Readings in Japanese material course in college curriculum. 11. Because Japanese person's name is complex and variable that it is desirable to add Chinese character with mother tongue script.

• International Journal of Internet, Broadcasting and Communication
• /
• v.8 no.3
• /
• pp.31-40
• /
• 2016

In this paper, we propose a new anti-forensic method for hiding data in the timestamp of a file in the Windows NTFS filesystem. The main idea of the proposed method is to utilize the 16 least significant bits of the 64 bits in the timestamps. The 64-bit timestamp format represents a number of 100-nanosecond intervals, which are small enough to appear in less than a second, and are not commonly displayed with full precision in the Windows Explorer window or the file browsers of forensic tools. This allows them to be manipulated for other purposes. Every file has $STANDARD_INFORMATION and $FILE_NAME attributes, and each attribute has four timestamps respectively, so we can use 16 bytes to hide data. Without any changes in an original timestamp of "year-month-day hour:min:sec" format, we intentionally put manipulated data into the 16 least significant bits, making the existence of the hidden data in the timestamps difficult to uncover or detect. We demonstrated the applicability and feasibility of the proposed method with a test case.

• Journal of Korea Society of Industrial Information Systems
• /
• v.16 no.2
• /
• pp.59-64
• /
• 2011

A lot of computer users use external memory device. But, same time file efflux incidents are also increasing. There are two ways people use for efflux file. One is moving it after checking file which is running on computer and the other is checking file name only. Checking from running file case, we can identify vestige with running information of applied program but, the case of moving as external device without running file there is no evidence running applied program. So there are a lot of difficulty with forensic investigation. In this paper we suggest the way to help forensic investigation which is method of getting external memory device information of volume and time through its awareness method and connection information and moving to external device without running file after compare the external memory device volume information through link file analysis and getting information of link file formation & access time from link file.

• International Journal of Internet, Broadcasting and Communication
• /
• v.10 no.4
• /
• pp.25-30
• /
• 2018

When producing a business card, there is always a risk for exposing personal information as the information from the business card application is saved in the computer of the printing agency. The information that goes into the business card application file, such as name, mobile phone number, and e-mail address is not encrypted, which makes it easy to access. This study was conducted in order to find a way to automatize the business card application process by encrypting the information entered into the business card application file, simplifying the business card application process, minimizing the workload and by directly linking to the print shop to remove the print file after completing the printing of the business card.

• KIISE Transactions on Computing Practices
• /
• v.20 no.12
• /
• pp.632-639
• /
• 2014

In this paper, we propose a PC's self-monitoring system for protecting illegal outflow of important personal files, which are managed in the Windows environment PC. This paper is based on the idea that it should be a read (access) file operation in order to outflow files through the network. To compare the information (name, some content) obtained from all operations of the reading of the 'Windows File System Driver' and captured transmission packets of mini port of 'Windows NDIS (Network Driver Interface Specification) Driver', and, if two pieces of information match, this system will determine the transmission to user. In this paper, tentatively, we developed a function to gather file information about file read operation and developed a function for transmission packet capture. This demonstration implementation showed that the proposed process has validity and the proposed process verified that it does not significantly affect the PC's performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.545-548
• /
• 2005

Recently, the computing environment of P2P come out to solve the excessive load of the server in the computing environment of the client/server. Currently, operated computing environment of P2P is mainly spreading out P2P system of read name or anonymity base about a publisher, sender and receiver of the file. But, to the current file protection there is three problem. The first problem is to a host become attack target. The second is to received file loose attack of Brute Force. The third is to define target of attack of DOS. To solve the this problem. it divide file into block unit. Each block is safely scattered peers. This paper propose the distribution P2P system of file division. Both proposing system solve the this problem and promote efficiency of file application.