• Proceedings of the Korean Society for Information Management Conference
• /
• 2001.08a
• /
• pp.167-172
• /
• 2001

클러스터링 성능 평가를 위한 척도로 여러 공식이 개발되어 사용되어왔다. 이들 평가척도는 가급적 범위가 제한되며, 가환성이 있고, 비편향적이며 단일 척도일 필요가 있다. 기존 평가 척도에 대해서 검토한 후 비편향적인 단일 척도 WACS를 개발하였다. 클러스터 수를 달리하는 클러스터링 결과에 대해 여러 평가척도를 적용해서 성능을 평가하는 실험을 통해서 WACS 척도가 평가척도로서의 요건을 만족시킨다는 것을 확인하였다.

• Journal of the Korean Society for information Management
• /
• v.17 no.4
• /
• pp.67-81
• /
• 2000

This study discussed why Precision(& Recall) is not a good effectiveness measurement of IR system providing ranked results, reviewed other effectiveness measurements appropriate for ranked results, and proposed new measurements based on the average rank of relevant documents retrieved. The 18 case-sets of ranked results were used for evaluating 10 effectiveness measurements including proposed measurements. Simple measurements were significantly similar with the 11-Point Precision requiring complicated calculation.

• Journal of vocational education research
• /
• v.37 no.1
• /
• pp.55-75
• /
• 2018

The purpose of this study was to develop the evaluation scales that can access hierarchically the lifelong education field training program based on relevant stakeholders. To fulfill this purpose, both qualitative and quantitative research methods were applied. First of all, four groups of stakeholders were selected such as trainees, professors, field instructors, and program inspectors. The in-depth interviews with 13 experts in the field were conducted, and as a result, the scale with hierarchy of evaluation area, evaluation section, and evaluation item for each stakeholder was developed. In order to verify the content validity of the evaluation scales, delphi surveys for each stakeholder were conducted, using 35 expert panels. After delphi survey, the scale was applied to 90 trainees, 22 professors, and 31 field instructors. SPSS program was used to analyze the descriptive statistics, validity, and reliability. Finally, the evaluation scale of the trainer was identified as 8 evaluation areas, 14 evaluation sections and 51 evaluation items. And the professor's scale consisted of 6 areas, 10 sections and 40 items. The field instructor's scale was determined as 3 areas, 7 sections and 31 items, and the program inspector's evaluation scale was confirmed as 7 areas, 15 sections, and 32 items.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.05a
• /
• pp.99-102
• /
• 2003

본 논문에서는 퍼지 클러스터링 알고리즘에 의해 구해진 퍼지 클러스터들에 대한 평가척도를 제안한다. 제안된 척도는 퍼 지 클러스터들간의 중첩성(overlapping)과 분리성 (separation)을 이용한다. 중첩성은 클러스터간 인접도를 이용하여 계산하며, 분리성은 데이터에 대한 상관성 정도를 나타낸다. 따라서 중첩성이 낮고 분리성이 높을수록 좋은 클러스터 결과라고 할 수 있다. 표준 데이터 집합을 대상으로 기존의 척도들과 비교실험 함으로써 제안된 척도의 신뢰성을 알아보았다.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.7
• /
• pp.28-35
• /
• 1998

본 논문에서는 아날로그 이동 전화 환경에서의, 객관적인 음질 평가 척도를 사용하 여 주관적 음질을 추정하는 이동전화 자동 음질평가 시스템을 개발하였다. 이동전화의 통화 품질을 유지하기 위해서는 이동전화의 네트워크를 계속하여 체크하는 것이 매우 중요하다. 주관적 음질 평가는 사람의 체감을 직접 나타내는 것이므로 실제적인 음질을 평가하는데 중 요한 척도가 되지만, 인력과 시간이 많이 소모되므로 다양한 지역에서 지속적으로 음질을 평가하는데 부적절하다. 이러한 문제를 해결하기 위하여 객관적 음질평가 척도를 이용하여 주관적 음질 평가 척도를 예측하는 자동 음질 평가 시스템이 필수적이다. 반복된 실험을 통 하여 BSD(Bark Spectral Distance)가 주관적 음질 평가 척도와 높은 상관관계가 있음을 확 인하였으며 원래의 음성과 이동 전화 채널을 통과한 음성과의 BSD를 측정한 후 이를 바탕 으로 MOS(Mean Opinion Score)를 추정하는 자동 음질 평가 시스템(Automatic Speech Quality Evaluator)을 개발하였다.

• Journal of Digital Convergence
• /
• v.12 no.4
• /
• pp.379-397
• /
• 2014

The purpose of this study was to develop the Job Functional Assessment Scale. This study was involved took third steps. First, 93 preliminary items were developed based on theoretical literature and International Classification of Functioning, Disability and Health (ICF) Model. To verify the content validity, 3 rehabilitation experts rated those items. Second, the scale consisting of 73 items was administered to sample of 880 people with disabilities. Finally, items of the scale were reduced to 58 items. Factor analysis showed evidence of construct validity of the scale. Internal consistency (Cronbach's ${\alpha}$) of total score showed .94 and Cronbach's ${\alpha}$ of 7 subfactor demonstrated .74~.90. Thus, Job Functional Assessment Scale demonstrated satisfactory convergent validity and discriminant validity. In addition, the practical use of the scale was discussed.

• Journal of the Korean Society for information Management
• /
• v.36 no.4
• /
• pp.129-159
• /
• 2019

The Purpose of this study is to develop an evaluation model for usability of research data service from the angles of evaluating usefulness of research data service itself and research data use experience-based usability. First, the various cases of evaluating usability of data services are examined and 4 rating scales and 20 measuring indicators for research data service are derived as a result of comparative analysis. In order to verify validity and reliability of the rating scale and the measuring indicators, the study conducted a survey of 164 potential research data users. KMO Bartlett Analysis was performed for validity test, and Principle Component Analysis and Verimax Rotating Method were used for component analysis on measuring indicators. The result shows that the 4 intrinsic rating scales satisfy the validity criteria of KMO Barlett; A single component was determined from component analysis, which verifies the validity of measuring indicators of the current rating scale. However, the result of 12 user experience-based measuring indicators analysis identified 2 components that are each classified as rating scale of utilization level and that of participation level. Cronbach's alpha of all 6 rating scales was 0.6 or more for the overall scale.

• The Journal of the Korea Contents Association
• /
• v.14 no.10
• /
• pp.628-638
• /
• 2014

The aim of this study was to explore the reliability and validity of the Korean version of the Recovery Assessment Scale in psychiatric patients. The original RAS was translated into Korean and the content was verified through back-translation procedures. This study included 142 patients who had chronic mental illness and were aged 18 or older. The subjects were assessed by using RAS-K, the 12-item Dispositional Hope Scale(DHS), the 28-item Mental Health Recovery Measure(MHRM), the 32-item Behavior and Symptom Identification Scale (BASIS-32). In the reliability test, Crobach's a coefficient and test-retest reliabilities were 0.92 and 0.79, respectively, indicating that the RAS-K has good internal consistency. In the analysis of the concurrent validity of the RAS-K, there were significant correlations between the RAS-K and DHS(r=.675, p<0.01), and between the RAS-K and MHRM(r=.816, p<0.01), but lower correlations with symptoms and clinician-related measures of psychiatric functioning. Factor analysis resulted in a five factor solution accounting for 62.476% of the common variance. Our study indicates that the RAS-K is an appropriate instrument to assess recovery in Korean psychiatric patients.

• Journal of Korean Society of Transportation
• /
• v.27 no.1
• /
• pp.97-105
• /
• 2009

It is expected that logistics standardization have an great effect on industry and national economy due to cost saving and improving efficiency. Therefore all countries of the world make a strenuous effort to take a lead of logistics standardization Despite such efforts of every country, our country remained a lukewarm attitude about logistics standardization. Especially, our country remained a lukewarm attitude about logistics standardization effect quantify. We have to suggest the necessity of standardization and the effect for carrying out logistics standard. So, this study performed literature review and case study for development of effect valuation method. Also, this study developed effect valuation standard for quantifiable standardization effect and drew the effect valuation standard model. First, the measure of preliminary effectiveness was chosen for MOE selection. As a results, vehicle load ratio, delivery cost, keeping space efficiency warehouse automation ratio, etc. were selected. Then, vehicle/truck load ratio, warehouse keeping efficiency, up unloading hour, packaging cost, consistency transportation ratio were chosen for MOE of pallet standardization.

• Journal of Digital Contents Society
• /
• v.15 no.2
• /
• pp.309-317
• /
• 2014

This study developed contents quality grading measurement for helping contents developers, sales persons, consumers. After survey of 28 professionals and 89 contents related employees, this quality grading measurement was established. Also this quality grading measurement contains requirement of contents developers, sales persons, consumers by supporting detailed techical, contents, qualitative metadata.