• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.464-469
• /
• 2001

CRM(Customer Relationship Management : CRM) is an advanced marketing supporting system which analyze customers\` transaction data and classify or target customer groups to effectively increase market share and profit. Many engines were developed to implements the function and those for classification and clustering are considered core ones. In this study, an improved clustering method based on SOM(Self-Organizing Maps : SOM) is proposed. The proposed clustering method finds the optimal number of clusters so that the effectiveness of clustering is increased. It considers all the data types existing in CRM data warehouses. In particular, and adaptive algorithm where the concepts of degeneration and fusion are applied to find optimal number of clusters. The feasibility and efficiency of the proposed method are demonstrated through simulation with simplified data of customers.

• 한국컴퓨터정보학회논문지
• /
• 제13권6호
• /
• pp.51-58
• /
• 2008

버전 제어 시스템은 급속한 환경의 변화나 개발 환경이 복잡한 프로그래밍 환경에서 사용되고 있으며, 정의된 형상 규칙 정보를 정확하게 알지 못하는 경우나 미리 정의되지 않은 정보에 대해서는 형상 형성 정보 제공 문제가 발생한다. 본 논문에서는 사용자가 원하는 컴포넌트를 복합적으로 형상형성 제어 할 수 있는 라이브러리 시스템을 제안하고, 모델링하여 구현하였다. 제안한 형상형성 제어를 위해 사용되는 라이브러리는 확장 facet 분류를 응용하여 하부 표현 구조에 관계없이 라이브러리로부터 컴포넌트를 검색할 수 있도록 하였다. 이것은 본 논문에서 제안한 TreeSearch 클래스와 형상형성 제어 함수를 이용하여 관리된다. 본 논문의 라이브러리는 다른 언어와의 인터페이스를 통해 사용될 수 있으며, 사용자에 의해 facet이 확장되는 장점을 가진다.

• 한국학교보건학회지
• /
• 제2권1호
• /
• pp.131-146
• /
• 1989

This study was conducted for the development of recording system of students' health problems, and for the application of International Classification of Health Problem in Primary Care(ICHPPC) as a tool of morbidity classification in school health care. The data were collected from 12th of September to 24th of September in 1988. The objects were composed of health problems written by 10 school nurses who take service in the elementary school. The results were as follows: 1. The features of students' health problmes. The health problems of students were 68 problems from the total numbers of 361 codes of ICHPPC. The 93.4% of health problems was contained in 20 descriptive diagnoses and 97.0% was contained in 30 descriptive diagnoses. According to frequency of main health problems, There were abrasion, scratch and blister(26.7%); disorder of stomach function, other disease of stomach and duodenum (20.4%); headache(10.6%); bruise and contusion (5.3%); acute URI (5.0%); laceration and open wound(4.6%); Insect bite and sting(4.0%); epistaxis(3.4%): abdominal pain(2.6%): superficial tissue(1.7%). Out of all health problems, Category 17(accident, injury and poisoning was 44.7%. and Category 9(digestive system Disease) was 22.2%. 2. Applicability of ICHPPC by the school nurses. School nurses used 68 codes, among the total number of 361 codes from ICHPPC. According to ICHPPC method, school nurses can classified more diverse health problems systematically and objectively than that in other studies on school nurses activities. ICHPPC was found as a useful and applicable tool of morbidity classification in the practice of school nurses.

• 한국문헌정보학회지
• /
• 제7권
• /
• pp.3-67
• /
• 1980

Because of the changes made in the modes of cataloguing and classification in its long history, the present catalogue of the National Central Library has become complicated and provides an inadequate guide to its collection. There can be no doubt that this is a serious deficiency in a closed access library since materials housed in the library are virtually inaccessible to unskilled readers. The whole breakdown of the efficiency of the catalogue is emminent and will ultimately create the most serious problems for the library. The main purpose of this survey is: (a) to identify problem areas created by the frequent changes in the cataloguing and classifying principles in the library and (b) to grope a rational direction for the future development. Analysing the various classification schemes and cataloguing rules adopted in processing materials (mainly books) in the library, the following conclusions have been made. A. The library adopted five different clasification schemes in different periods, of which KDCP was used for the most part of its collection. KDCP is recommended to use for the future colletion. A classification development office is recommended to be established within the library, of which the main function is to revise the KDCP in collaboration with the appropriate committee of the Korean Library Association. B. The present practice in the library is to apply three different cataloguing rules and two different author notation tables to the Oriental, classical, and Western collections. Efforts should be made to find out an efficient system so that this variety is simplified. An alphabetical index should be added to the classified catalogue, and improvements are required in the Japanese collection. C. The technical services division is inadequately staffed. The staff should be sufficiently numerous and specially qualified. D. The present financial support for the technical services of the library is inadequate. Sufficient financial provision should be made to ensure the effective work. E. A feasibility study should be carried out to develop a computer processing system for providing machine-readable catalogue records on magnetic tape for use by the library community in Korea.

• 전기전자학회논문지
• /
• 제20권4호
• /
• pp.398-411
• /
• 2016

기계학습 방법의 하나인 SVM은 뛰어난 일반화 성능으로 영상처리 분야에서 많이 사용하고 있다. 하지만 SVM을 이용한 시스템에서 미리 학습된 데이터가 아닌 다른 데이터를 이용하려하면 새로 학습을 시켜야 하는 경우가 생긴다. 특히, 임베디드 환경에서는 이러한 상황에서 학습 시간이 오래 걸려 SVM을 적절히 이용하지 못하는 경우가 있다. 본 논문에서는 이러한 문제점을 해결하기 위하여 SVM의 학습 및 분류를 모두 수행할 수 있도록 하나의 FPGA로 구현하였다. SVM 연산의 복잡성으로 인해 생기는 반복연산을 병렬처리를 통하여 해결하고 커널 사용으로 생기는 지수 연산을 변형하여 고정 소수점 연산이 가능하도록 하였다. 제안하는 하드웨어는 Xilinx사의 ZC 706보드에 구현하였고, 구현한 FPGA의 검증을 위하여 TSR 알고리즘을 이용하였다. 구현한 하드웨어는 100 MHz의 주파수로 동작하며, 2천개의 데이터를 이용한 학습 시 약 5sec가 소요되고 $1360{\times}800$ 해상도에서 분류 시 약 16.54msec가 소요됨을 확인했다.

• 한국멀티미디어학회논문지
• /
• 제22권10호
• /
• pp.1168-1177
• /
• 2019

In order to diagnose and prevent Alzheimer's Disease (AD), it is becoming increasingly important to develop a CAD(Computer-aided Diagnosis) system for AD diagnosis, which provides effective treatment for patients by analyzing 3D MRI images. It is essential to apply powerful deep learning algorithms in order to automatically classify stages of Alzheimer's Disease and to develop a Alzheimer's Disease support diagnosis system that has the function of detecting hippocampus and CSF(Cerebrospinal fluid) which are important biomarkers in diagnosis of Alzheimer's Disease. In this paper, for AD diagnosis, we classify a given MRI data into three categories of AD, mild cognitive impairment, and normal control according by applying 3D brain MRI image to the Faster R-CNN model and detect hippocampus and CSF in MRI image. To do this, we use the 2D MRI slice images extracted from the 3D MRI data of the Faster R-CNN, and perform the widely used majority voting algorithm on the resulting bounding box labels for classification. To verify the proposed method, we used the public ADNI data set, which is the standard brain MRI database. Experimental results show that the proposed method achieves impressive classification performance compared with other state-of-the-art methods.

• 농촌계획
• /
• 제29권4호
• /
• pp.13-26
• /
• 2023

Recently, South Korea has prepared laws and systems to systematically manage rural spaces in response to the era of population decline and is making various efforts to promote related policies. However, various basic studies that can support this are still insufficient. In particular, in this study, the functions and roles of each settlement class were established along with the classification of the sedentary classes in rural areas, and the classification system for rural living services was established, and the hierarchy by functional facilities and the minimum standards for vulnerable areas (accessibility) were established. Specifically, in this study, the settlement class was divided into 4 classes of "central area - midpoint area - small point area - hinterland", and each function and role was presented, and the rural living service classification system was finally reestablished as 10 sectors and 31 functional facilities. In addition, the hierarchy and accessibility standards of rural living service functional facilities was set within 5 to 15 minutes for 'lower and basic services', within 10 to 20 minutes for 'medium and basic services', within 15 to 30 minutes for 'intermediate and complex services', within 20 to 60 minutes for 'high car/complex service' and within 10 minutes for 'urgent service'.

• 한국정보통신학회논문지
• /
• 제9권2호
• /
• pp.322-329
• /
• 2005

본 논문에서는 실시간 시선 식별을 위한 능동적 적외선 조명을 기반으로 한 컴퓨터 비전 시스템을 제안하고자 한다. 현존하는 대다수의 시선 식별 방법은 고정된 얼굴에서만 원활하게 동작하고 개개인에 대한 교정 절차를 필요로 한다. 그러므로 본 논문에서는 교정 작업 없이 얼굴 움직임이 있는 경우에도 견실하고 정화하게 동작하는 시선 식별 시스템을 제안하고 있다. 제안된 시스템은 추출된 변수로부터 시선을 스크린 좌표로 매핑하기 위해 GRNN을 이용하였다. GRNN을 사용함으로서, 시선 매핑에 분석적 기능이나 얼굴 움직임에 대한 계산이 필요 없을 뿐 아니라 다른 개개인에 대하여 학습과정에서 매핑 기능을 일반화 할 수 있었다. 시선 예측 정확도를 개선하고자 계층적 식별을 제안함으로써 오분류를 줄일 수 있었고, 공간 시선 해상도는 스크린에서 Im 정도 떨어진 거리에서 수평으로 10cm, 수직으로 약 13cm, 즉 수평으로 8도 수직으로 5도 이었다 실험 결과, 재식별 하였을 경우 1차 시선 식별시 84$\%$보다 약 9$\%$ 정도 정확성이 향상되어 93%를 나타냄으로써 제안된 시스템의 유효성을 증명하였다.

• The Journal of Korean Physical Therapy
• /
• 제31권5호
• /
• pp.279-285
• /
• 2019

Purpose: This study examined the effects of spinal stabilization exercises using visual feedback on the gross motor function and balance of the sitting posture in children with cerebral palsy. Methods: The subjects were 18 children with cerebral palsy aged 8-15 years in the I-III stages of the Gross Motor Function Classification System. The subjects were divided into an experimental group (n=9) and control group (n=9). The experimental group was treated with 30 minutes of neurodevelopmental treatment and 20 minutes of spinal stabilization exercises using visual feedback. The control group was treated with 30 minutes of neurodevelopmental treatment and 20 minutes of spinal stabilization exercises without visual feedback. Both groups participated in the experiment twice a week for eight weeks. The Gross Motor Function Measurement was performed to evaluate the changes between pre- and the post-intervention in gross motor function. The Seated Limit of Stability Surface Area was measured to evaluate the changes in trunk balance. Results: Both experimental and control groups showed a significant increase in the gross motor function and trunk balance (p<0.05). The experimental group showed a significant increase in gross motor function compared to the control group (p<0.05). The experimental group showed a significant increase in the dynamic trunk balance in all directions when measuring the Seated Limit of Stability Surface Area (p<0.05). Conclusion: Spinal stabilization exercises using visual feedback for the neurodevelopmental treatment of children with cerebral palsy can improve their gross motor function and trunk balance when in a sitting posture more effectively.

• 전기학회논문지
• /
• 제64권6호
• /
• pp.922-934
• /
• 2015

In this paper, The classification between precipitation echo(PRE) and non-precipitation echo(N-PRE) (including ground echo and clear echo) is carried out from weather radar data using neuro-fuzzy algorithm. In order to classify between PRE and N-PRE, Input variables are built up through characteristic analysis of radar data. First, the event classifier as the first classification step is designed to classify precipitation event and non-precipitation event using input variables of RBFNNs such as DZ, DZ of Frequency(DZ_FR), SDZ, SDZ of Frequency(SDZ_FR), VGZ, VGZ of Frequency(VGZ_FR). After the event classification, in the precipitation event including non-precipitation echo, the non-precipitation echo is completely removed by the echo classifier of the second classifier step that is built as Type-2 FCM based RBFNNs. Also, parameters of classification system are acquired for effective performance using PSO(Particle Swarm Optimization). The performance results of the proposed echo classifier are compared with CZ. In the sequel, the proposed model architectures which use event classifier as well as the echo classifier of Interval Type-2 FCM based RBFNN show the superiority of output performance when compared with the conventional echo classifier based on RBFNN.