• Asia pacific journal of information systems
• /
• 제16권4호
• /
• pp.121-149
• /
• 2006

This study tries to explain how the relationship-specificity of invested assets affects E-collaboration and consequently on performance of Small and Medium Enterprises (SMEs). We classify the relationship-specificity of invested assets into four types such as business process specificity, physical asset specificity, domain knowledge specificity, and site specificity. We define E-collaboration as composed of Electronic Information Sharing (EIS) and Electronic Cooperation (E-Co). In addition, we articulate firm's performance as operational and strategic one, and investigate the impacts of EIS and E-Co on its performance. The data were collected from 187 SMEs and used for analysis. Based on the survey results, we find the following: (1) EIS is directly influenced by business process specificity and physical asset specificity, (2) E-Co is affected by site specificity and domain knowledge specificity, (3) EIS has a positive and significant impact on E-Co, (4) EIS affects firm's operational performance, (5) E-Co influences on firm's strategic performance. In conclusions, the higher the level of EIS, SMEs seem to get greater operational performance, Respectively, the higher the level of E-Co, they tend to get greater strategic performance.

• 한국언어정보학회지:언어와정보
• /
• 제11권2호
• /
• pp.67-86
• /
• 2007

The semantic literature on definiteness and specificity shows that the former is relatively an established notion, whereas the latter is still a vague notion that needs to be more clarified and confirmed. Given this, Ionin (2006) argues for the reality of specificity based on the informal use of this. She proposes this in spoken English as a specificity marker which has a semantic feature indicating "peaker intent to refer" and "noteworthiness." She also provides as evidence the results of some crosslinguistics studies including an L2 acquisition study with both L1-Russian and L1-Korean L2-English learners. However, this paper will argue that the informal use of this does not seem to mark specificity according to Ionin's definition of specificity. It will also be argued that the L2 acquisition study cannot be used as evidence for the reality of specificity. Based on these arguments, this paper will try to redefine specificity, based on the notions of existence and uniqueness.

• Clinics in Shoulder and Elbow
• /
• 제4권2호
• /
• pp.186-190
• /
• 2001

Purpose: The aim of this study was to investigate the diagnostic accuracy of Neer, Hawkins provocative tests and supraspinatus manual muscle test for the assessment of impingement syndrome, partial tear and small complete tear of De rotator cuff. Materials and Methods: Seventy-one female and 115 male patients were included in the study. Patients were divided into four groups of no impingement, impingement without tear, partial tear and small complete tear of the rotator cuff, which were confirmed by sonogram, magnetic resonance imaging and surgery. Neer and Hawkins provocative tests and supraspinatus manual muscle test were performed respectively. SAS 6.12 version was used in statistical analysis. Results: We found that Neer test had 94% sensitivity, 54% specificity for impingement without tear and 89% sensitivity, 78% specificity for partial tear and 96% sensitivity, 23% specificity for small tear. Hawkins test revealed 95% sensitivity, 54% specificity for impingement without tear and 93% sensitivity, 78% specificity for partial tear and 100% sensitivity, 23% specificity for small tear. Supraspinatus manual muscle test revealed 27% sensitivity, 94% specificity for impingement without tear and 29% sensitivity, 82% specificity for partial tear and 48% sensitivity, 82% specificity for small tear. Conclusion: Neer and Hawkins tests have high sensitivity, low specificity for impingement syndrome, partial and small tear. Supraspinatus manual muscle test had low sensitivity and high specificity. However this test was not effective to differentiate the partial and small rotator cuff tear. We thought that more effective provocative test should be designed to detect the partial and small rotator cuff tear.

• 인터넷정보학회논문지
• /
• 제18권6호
• /
• pp.127-136
• /
• 2017

The strategy that using data collection from Netflix uses for its services is different from traditional human interaction and the communication, and it is represented by the systematic algorithm that rooted from intelligent information system based on the human interaction and communication. These characteristics allowed the study to reflect the influence of 'Asset specificity' which affects the continuous consumption of the media services of Netflix users through economic psychological analysis based on transactional cost economics. The result from the survey on actual Netflix users, three types of specificity (Space specificity, time specificity, relational specificity) reduced perceived searching cost whereas perceived instrumentality has increased, eventually reinforces the commitment to the service. This implies that the service characteristics of Netflix, trying to communicate with the individuals based on intelligent information system are distinct from the existing platform services and it gives the significance of work very effective for user's continuous consumption of the media services.

• 한국한의학연구원논문집
• /
• 제13권3호
• /
• pp.125-131
• /
• 2007

Objectives: We intended to know sensitivity, specificity and relation of diagnosis methods of Sasang Constitution. Methods: We compared QSCC II, PSSC, Body measurement dignosis results with gold standard ones which were diagnosed by Sasang Constitution specialist, and analyzed by crosstables to get the sensitivity and specificity of those. Results and conclusions 1. Sensitivity and specificity of one diagnosis method were low, so call, in QSCC II, Taeumin is 49% in sensitivity, 97% in specificity, Soeumin is 67%, 69%, Soyangin is 59%, 75%, in Body measurement, Taeumin is 66%, 89%, Soeumin is 70%, 82%, Soyangin is 49%, 78%. in PSSC, Taeumin is 57%, 90%, Soeumin is 88%, 62%, Soyangin is 18%, 88%. 2. When two diagnosis methods are consistent each other, sensitivity and specificity are more higher than one diagnosis method. 3. When three diagnosis methods are all consistent, sensitivity and specificity are most highest, but in Soyangin, those are decreased comparing with the case of consistency between QSCC II and Body measurement.

• 응용통계연구
• /
• 제33권5호
• /
• pp.541-553
• /
• 2020

Receiver operating characteristic (ROC) 곡선은 민감도와 특이도로 표현되며, ROC 곡선을 이용하는 최적분류점도 민감도와 특이도만을 반영하지만, 본 연구에서는 질병률과 효용을 추가하여 고려하는 기대효용함수를 연구한다. 특히 교차하는 ROC 곡선들의 area under the ROC curve (AUC) 값들이 유사한 경우에 특정한 부분의 부분 AUC를 비교해야 한다. 본 연구에서는 정의된 민감도 직선과 특이도 직선을 바탕으로 각각 높은 민감도와 특이도를 나타내는 부분 AUC를 제안한다. ROC 곡선들이 교차하고 동일한 AUC 값을 갖는 다양한 분포함수를 설정하여, 민감도 직선과 특이도 직선을 이용하여 구한 부분 AUC를 비교하면서 모형의 판별력을 향상시키는 방법을 제안한다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제16권15호
• /
• pp.6177-6186
• /
• 2015

Sensitivity and specificity are the two most important indicators in selection of medical imaging devices for cancer screening. Breast images taken by conventional or digital mammography, ultrasound, MRI and optical mammography were collected from 2,143,852 patients. They were then studied and compared for sensitivity and specificity results. Optical mammography had the highest sensitivity (p<0.001 and p<0.006) except with MRI. Digital mammography had the highest specificity for breast cancer imaging. A comparison of specificity between digital mammography and optical mammography was significant (p<0.021). If two or more breast diagnostic imaging tests are requested the overall sensitivity and specificity will increase. In this literature review study patients at high-risk of breast cancer were studied beside normal or sensitive women. The image modality performance of each breast test was compared for each.

• 한국언어정보학회지:언어와정보
• /
• 제11권2호
• /
• pp.23-47
• /
• 2007

In this paper I discuss two markers for specificity, etten and han, in Korean. First, I discuss their structural properties and claim that etten is a determiner, and han is a numeral preceded by an implicit existential quantifier. Then I discuss four uses of etten and two uses of han, and show when they are used as specificity markers. There are various properties of the two specificity markers, some of which they have in common. I take them as properties of specificity markers in general. The properties only one of them has are taken to come from the original meanings of the two morphemes. I claim that specific indefinites range over individual concepts, which lead to the properties they have in common.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2005년도 한국컴퓨터종합학술대회 논문집 Vol.32 No.1 (A)
• /
• pp.913-915
• /
• 2005

유전자 칩의 정확성은 각 유전자들의 식별자로 활용되는 probe들에 의해 결정된다. 칩에 삽입되는 probe들은 반응오류를 피하기 위해 이중구조, 녹는점, 그리고 CG구조와 같은 요소들을 고려한다. 또한 다른 유전자들과의 교차반응을 최소화하기 위해 specificity를 고려한다. probe의 specificity 검증은 전체 유전자들에 대해 탐색해야 하므로 대규모 염색체에 대해서는 많은 계산이 요구된다. 본 논문에서는 specificity 검증을 위한 효율적인 알고리즘을 제시한다. 제시한 알고리즘은 해시테이블을 활용하여 probe가 specificity를 만족하지 못하게 하는 유전자 시퀀스들만을 탐색하여 비교한다. 제시한 알고리즘이 기존 알고리즘보다 효율적임을 실험결과를 통해 보인다.

• 품질경영학회지
• /
• 제30권1호
• /
• pp.144-151
• /
• 2002

Consider the problem of estimating the confidence interval of defect rate of a specified population through a screening test. Relationship between defect rate of the population and sensitivity and specificity of the screening test is discussed. More general statistical problem is introduced with the assumption of unknown imperfect specificity. An improved version of Hogan & Gladden(1978)'s estimator is proposed and discussed. Simulation studies are performed to compare the efficiency of the proposed estimator with current estimators.