• 영어어문교육
• /
• 제15권2호
• /
• pp.169-184
• /
• 2009

The purpose of this study is to identify the most important factor(s) that contribute(s) to the overall oral proficiency and to examine whether the factor(s) play(s) a different role depending on the proficiency level of learners. Learners were divided into novice group and advanced group and were asked to produce an oral recording of a story based on a comic strip. The recordings were transcribed and graded by three raters. According to the results, the raters attained high inter-rater reliability when assessing advanced learners. However, the reliability level became considerably lower in the assessment of novice learners. The best predictor of oral proficiency among novice learners was sociolinguistic competence and fluency, while grammatical competence and fluency were the strongest predictors for advanced learners. The results suggest the need for a separate assessment tool for different proficiency levels and the need for a different focus in the classroom depending on the learners' proficiency level.

• 한국보육지원학회지
• /
• 제12권4호
• /
• pp.111-130
• /
• 2016

This study examined the reliability and validity evidence of the Korean version of Wechsler Primary & Preschool Scale for Intelligence(K-WPPSI-IV), which was standardized recently. Following the stratified sampling for Korean population in terms of region and maternal education level, 1,700 children aged 2.6~7.7 years were recruited for Korean norm development. The split-half reliability (N=1,700) was high(Fisher's z(.75-.89)) and test-retest reliability(N=100) was satisfactory ranging from .95 to .68. The inter-rater reliability(N=110) was very high(.96-1.0). Discriminant and convergent validities were revealed by correlational patterns among subtests and indices. Future studies are needed to extend both reliability and validity of K-WPPSI-IV by collecting data with a various sample group and a variety of instruments.

• 말소리와 음성과학
• /
• 제9권2호
• /
• pp.1-7
• /
• 2017

The purpose of this study is to investigate the relations between phonetic differences among Korean learners' production of English front vowels, pronunciation intelligibility and speaking proficiency test score. To do so, thirty Korean university students were asked (1) to read English text book paragraphs and (2) describe a picture. Two English native raters and one Korean rater evaluated Korean subjects' English pronunciation intelligibility and speaking. In addition, subjects' English vowel productions were acoustically analyzed(F0, F1, F2, vowel duration, intensity). The results of the study show that the vowel quality and pitch of the unstressed vowels and lax vowel are related to the pronunciation intelligibility. In addition, the scores of pronunciation intelligibility and speaking are highly related.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 B
• /
• pp.801-803
• /
• 2003

Faults in induction motors can be categorized into mechanical faults and electrical faults, and most mechanical faults result from inferiority or damage of the bearing, while most electrical faults derive from insulation faults of stator windings and rotor bar cracks. When a crack appears on the rotor bar, its efficiency decreases, which increases energy consumption and temperature, reducing the life span of the motor. This kind of fault can only be sensed by the protection relay after the condition has worsened to a certain degree, bringing massive economic loss. This paper will deal with the diagnosis method of rotor bar faults through the load current analysis method of the motor used during operation.

• 한국IT서비스학회:학술대회논문집
• /
• 한국IT서비스학회 2003년도 추계학술대회
• /
• pp.223-232
• /
• 2003

추천시스템을 설계하는 방법에는 크게 Content-Based Filtering 기법과 Collaborative Filtering 기법이 있다. 이 중 Collaborative Filtering 기법은 사용자가 아직 평가하지 못한 상품에 대한 예측값을 계산할 때, 나와 유사한 상품선호를 갖고 있는 사람들이 그 상품에 대해 평가한 점수를 활용하는 방법이다. 하지만 순수한 Collaborative Filtering 방법은 일반적으로 알려진 Data Sparsity의 문제, First Rater의 문제뿐만 아니라 예측값의 부정확성과 기하급수적 계산량의 증가로 실제구현이 어렵다는 문제점을 가지고 있다. 본 연구에서는 이러한 'Collaborative filtering' 시스템의 문제들 중 예측의 부정확성과 실제 구현의 어려움을 해결할 수 있는 방법으로 군집분석을 적용해 보았다. 특히 본 연구에서는 군집을 나눌 때, 실제 추천이 이루어지는 상품 도메인이 아닌, 그 상품도메인과 비슷한 선호의 기준을 가지고 선택하게 되는 '선택의 상관관계'가 높은 '이웃 상품도메인'에서 사용자들의 군집을 나누고 이를 실제 추천이 이루어지는 상품도메인에 적용하는 방식을 사용하였다.

• Nuclear Engineering and Technology
• /
• 제26권4호
• /
• pp.570-577
• /
• 1994

The loss of water in the carbon steel balls and water region of the end shield for CANDU 6 reactor could lead to significant temperature gradient through the end shield structure which amy result in the excessive deformation. With an assumed end shield drained scenario, the heat deposition rates were calculated through the end shield associated with the central fuel channel during full power operation as an initial step to thermal stress analysis. The drained case was compared with that of water present normal case in therms of heat deposition rater and the total heating throughout the end shield regions. The compared results show that the heat deposition and the total heating remain almost the same between the two cases. It was found that the change of volume integrated flux in the end shield regions due to the loss of water contribute a negligible effect on the heat deposition in this region.

• 대한치과의사협회지
• /
• 제57권8호
• /
• pp.448-455
• /
• 2019

This study aims to apply the Generalizability Theory (G-theory) for estimation of reliability of evaluation scores between raters on Patient Dentist Interaction. Selecting a number of raters as multiple error sources, this study was analyzed the error sources caused by relative magnitude of error variances of interaction between the factors and proceeded with D-study based on the results of G-study for optimal determination of measurement condition. The estimated outcomes of variance component for accuracy among the Patient Dentist Interaction evaluation with G-theory showed that impact of error was the biggest influence factor in students. The second influence was the item effect, and the rater effect was relatively small. The Generalizability coefficients for case1 and case2 which were estimated through the D- study were calculated relatively low.

• 산업융합연구
• /
• 제3권1호
• /
• pp.125-143
• /
• 2005

This study investigates that the relationship of performance appraisal justice and employee's job satisfaction and whether appraisal outcome perception can significantly take the moderate role on this relationship. We conduct empirical study with one dimension of distributive justice and 5 dimensions of procedural justice; appraisal system, participation of employee, feedback, knowledge of rater, and appraisal standard. The result shows that distributive and procedural justice significantly explain employees' job satisfaction. The result, also, shows that moderate role of appraisal outcome perception is significant. This means if the employee who regards appraisal outcome as high perceive appraisal justice, she/he will experience more job satisfaction than the employee who regards appraisal outcome as low.

• Journal of information and communication convergence engineering
• /
• 제20권2호
• /
• pp.113-124
• /
• 2022

News in the form of web data generates increasingly large amounts of information as unstructured text. The capability of understanding the meaning of news is limited to humans; thus, it causes information overload. This hinders the effective use of embedded knowledge in such texts. Therefore, Automatic Knowledge Extraction (AKE) has now become an integral part of Semantic web and Natural Language Processing (NLP). Although recent literature shows that AKE has progressed, the results are still behind the expectations. This study proposes a method to auto-extract surface knowledge from English news into a machine-interpretable semantic format (triple). The proposed technique was designed using the grammatical structure of the sentence, and 11 original rules were discovered. The initial experiment extracted triples from the Sri Lankan news corpus, of which 83.5% were meaningful. The experiment was extended to the British Broadcasting Corporation (BBC) news dataset to prove its generic nature. This demonstrated a higher meaningful triple extraction rate of 92.6%. These results were validated using the inter-rater agreement method, which guaranteed the high reliability.

• 말소리와 음성과학
• /
• 제15권2호
• /
• pp.13-20
• /
• 2023

This study explores the potential of automated speech recognition (ASR) in assessing English learners' pronunciation. We employed ASR technology, acknowledged for its impartiality and consistent results, to analyze speech audio files, including synthesized speech, both native-like English and Korean-accented English, and speech recordings from a native English speaker. Through this analysis, we establish baseline values for the word error rate (WER). These were then compared with those obtained for human raters in perception experiments that assessed the speech productions of 30 first-year college students before and after taking a pronunciation course. Our sub-group analyses revealed positive training effects for Whisper, an ASR tool, and human raters, and identified distinct human rater strategies in different assessment aspects, such as proficiency, intelligibility, accuracy, and comprehensibility, that were not observed in ASR. Despite such challenges as recognizing accented speech traits, our findings suggest that digital tools such as ASR can streamline the pronunciation assessment process. With ongoing advancements in ASR technology, its potential as not only an assessment aid but also a self-directed learning tool for pronunciation feedback merits further exploration.