http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
윤현님(Hyunnim Yoon),김형일(Hyungil Kim) 한국정보기술학회 2009 한국정보기술학회논문지 Vol.7 No.6
Current e-books exist as digital content that integrates voice and video to the text. Because e-books exist as digital content, they are easy to distribute. Furthermore, e-books are highly portable because they can be stored in a storage device. e-Books can be quickly and continuously updated. e-Books often exist in diverse forms. If the popularity of e-books will increase, users will spend a lot time trying to find the e-books they want. In order to solve this problem, it is necessary to find technique that can analyze e-books of various forms effectively. In this paper, we propose a method for extracting appropriate e-books to individual users. The proposed method mixes feature using category and this method analyzes the categories of e-books and a list of e-books borrowed by the users, and extracts e-books suitable for users. In simulation, the method that considered the characteristics of user showed the mean hit ratio of 36% and the proposed method showed 73%. Several simulation results that show the effectiveness of the proposed method are also presented.
HCI 기반 의료정보 시스템을 위한 뇌영상의 객체 분류를 이용한 이상 부위 추출
김형일(Hyungil Kim),윤현님(Hyunnim Yoon),김용욱(Yonguk Kim) 한국정보기술학회 2012 한국정보기술학회논문지 Vol.10 No.1
In case medical information systems provide only general image information and medical information, they restrict convenience and efficiency in physicians’ diagnostic and therapeutic activities. In this paper, we propose a HCI-based medical information system that extracts objects suspected to be brain tumor and provides them together with patient information and image information in order to support physicians’ diagnostic and therapeutic activities. This system performed SVM?based object classification using various object characteristics of brain images in order to extract abnormal objects suspected to be brain tumors. The experiment used abnormal objects suspected to be brain tumors, white matter, gray matter, and cerebrospinal fluid, and the experimental results showed the average accuracy of 74.4%.