• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.327-332
• /
• 2023

데이터의 품질과 다양성은 모델 성능에 지대한 영향을 끼친다. 본 연구에서는 Topic을 활용한 데이터 전처리와 BERT 기반 MLM, T5, Random Masking을 이용한 증강으로 데이터의 품질과 다양성을 높이고자 했으며, 이를 KoBERT 기반 에세이 자동 평가 모델에 적용했다. 데이터 전처리만 진행했을 때, Quadratic Weighted Kappa Score(QWK)를 기준으로 모델이 에세이의 모든 평가 항목에 대해 베이스라인보다 더욱 높은 일치도를 보였으며 평가항목별 일치도의 평균을 기준으로 0.5368029에서 0.5483064(+0.0115035)로 상승했다. 여기에 제안하는 증강 방식을 추가 할 경우 MLM, T5, Random Masking 모두 성능 향상 효과를 보였다. 특히, MLM 데이터 증강 방식을 추가로 적용하였을 때 최종적으로 0.5483064에서 0.55151645(+0.00321005)으로 상승해 가장 높은 일치도를 보였으며, 에세이 총점으로 QWK를 기준으로 성능을 평가하면 베이스라인 대비 0.4110809에서 0.4380132(+0.0269323)로의 성능 개선이 있었다.

• Science of Emotion and Sensibility
• /
• v.11 no.1
• /
• pp.69-80
• /
• 2008

The scheme to approach emotional design developed in various way of research without defining criterion of emotional way. A aim of this research to derive a new way of approaching system and construct the process in emotional design. In first experiments, analyze meaning of important factors in cooking kits to construct the model of emotional approaching system by the analysis of quantification theory type 3. This shows that four emotional factors as shape, atmosphere, functional matter, information are deeply related with emotional thinking and design. Consequently, the way of emotional approaching in cooking kit have several conditions essentially as following these; 1.Is it have functional advantages to cook? 2. Is it good looking in shape? 3. Is it possible to make some of mood on cooking? 4. How much information of cook we get in? These conditions were classified in detail to establish emotional design system.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.04a
• /
• pp.161-165
• /
• 2006

본 연구의 목적은 한국기업을 대상으로 한국 국가품질상(The Korean National Quality Award) 평가도구의 타당성을 검증하고, 이러한 평가 항목들 간에 어떠한 인과관계가 있는지를 Pilot test를 통해 분석하고자 한다. 이를 위해 미국의 말콤볼드리지 국가품질상(The Malcolm Bladrige National Quality Award)의 평가기준인 리더십, 전략계획, 고객과 시장중시, 정보와 분석, 인적자원 중시, 프로세스 관리, 사업성과의 7개 항목과 21개의 세부항목을 이용하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.1
• /
• pp.99-112
• /
• 2014

This paper addresses an assessment model for selecting PL(Private Level) product companies. The proposed model extracts the weights of evaluation elements based on AHP(Analytic Hierarchy Process). These evaluation elements are from real-world instances in a domestic hyper market. Especially, the model points at food products which constitute a large portion of entire profits in the market. In this model, we first classify the 54 evaluation elements into 4 layers and secondly carry out a survey with relevant specialists based on them. We also estimate the weights of evaluation elements according to pairwise comparisons from the survey, and propose them as a quantitative alternative which can be applied in real-world problems. Finally, the pilot-study is conducted to compare the proposed model with the existing simple summation method. From this study, HACCP system assessment and Review(0.279281), Transportation(0.117706) and Fundamental law observance(0.066392) are presented as the key evaluation elements for selecting PL product companies. The proposed model facilitates the company selection among those candidate companies which is not easy to determine the superiority by reflecting the importance of evaluation elements.

• Therapeutic Science for Rehabilitation
• /
• v.10 no.3
• /
• pp.83-96
• /
• 2021

Objective : This study aimed to determine Korea-type family function factors that affect individual growth and social activities, and examine check the appropriate contents and domestic culture components based on the McMaster model, a family function evaluation tool, through the Delphi technique. Methods : The Delphi technique was applied to 12 expert panels in fields related to family function. The period lasted 9 weeks, from May to June 2020. The Delphi survey was conducted twice. In the first survey, the domestic culture and appropriate contents of the McMaster model were selected and localized, and expert' opinions about the components were collected using closed and open-ended questions. In the second survey, the fitness and importance of the components were investigated. Results : As a result of the first Delphi investigation, 18 items were deleted from the 53 items presented. After adding 11 items and excluding any overlapping items, a total of 40 items were selected. Subsequently, sentences that were difficult to understand were revised to familiar vocabulary. A second survey was constructed, with an example sentence. In the second Delphi investigation, 33 items were selected. The average content validity ratio for the final selected component was 0.76, and the stability was 0.28. Conclusion : Family function and the factors influencing domestic family function identified though this study can be used to conduct family function evaluations and interventions in clinical sites or relevant research studies.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.7
• /
• pp.3245-3250
• /
• 2011

This study introduces a threat level assessment model adapting Fuzzy theories in order to help make decisions for better covering quantitative factors and qualitative ones together. The threat is classified into three major categories - one resulting from navigational condition, another from target vessel specification and the other from external decision environment. The threat levels by each category are examined by a fuzzy inference, and its corresponding weights are assigned via fuzzy measures. Finally the high level threat measures become integrated via a Choquet Fuzzy Integral method into ultimate threat level indicators.

• Journal of Korea Technology Innovation Society
• /
• v.3 no.1
• /
• pp.55-67
• /
• 2000

본 고는 국내에 적용할 수 있는 기술등급평가모델을 제시하고 이에 입각하여 이루어진 실제의 평가결과를 제시하였다 이 평가는 중소기업청에 의해 전국규모로 1998년과 1999년 두해에 이루어졌다. 등급평가는 기술의 권리적 측면 환경적측면 등 세가지항목으로 구성된 평점모델을 제시하였다. 이평가결과는 변별력이 약해 평가요소와 실제 경쟁력간의 인과관계등에 대한 보다 집중적인 연구가 필요함을 보여준다.

• Journal of KIISE:Software and Applications
• /
• v.30 no.7_8
• /
• pp.618-641
• /
• 2003

Various quality models using quality factor, quality criteria and metrics have been proposed in order to evaluate quality of software products. However, a customized quality model which is specific to the characteristics of software component is required. In this paper, we propose external quality prediction techniques enable us to predict what external quality the final software product will have by using metrics as with internal attributes of software in development. We also propose a model not only for measuring quality by using metrics but also for applying internal attributes of ISO 9126 into artifacts of software component development.

• Tunnel and Underground Space
• /
• v.14 no.5
• /
• pp.353-362
• /
• 2004

Prediction of fragmentation in bench blasting is one of the most important factors to establish the production plan. It is widely accepted that fragmentation could be accurately predicted using the Kuz-Ram model in bench blasting. Nevertheless, the model has an ambiguous or subjective aspect in evaluating the model parameters such as joint condition, rock strength, density, burden, explosive strength and spacing. This study proposes a new method to evaluate the parameters of Kuz-Ram model, and the predicted mean fragment sizes using the proposed method are examined by comparing the measured sizes in the field. The results show that the predictions using Kuz-Ram model with the proposed method coincide with field measurements, but Kuz-Ram model does not reflect the in-situ rock condition and hence needs to be improved.

• Journal of Software Assessment and Valuation
• /
• v.15 no.2
• /
• pp.73-83
• /
• 2019

Voice Recognition has recently been improved with AI(Artificial Intelligence) and has greatly improved the false recognition rate and provides an effective and efficient Human Machine Interface (HMI). This trend has also been applied in the defense industry, particularly in the aviation, F-35. However, for the quality evaluation of Voice Recognition, the defense industry, especially the aircraft, requires measurable quantitative models. In this paper, the quantitative evaluation model is proposed for applying Voice Recognition to aircraft. For the proposal, the evaluation items are identified from the Voice Recognition technology and ISO/IEC 25000(SQuaRE) quality attributes. Using these two perspectives, the quantitative evaluation model is proposed under aircraft operation condition and confirms the evaluation results.