• Journal of the Korea Computer Industry Society
• /
• v.2 no.12
• /
• pp.1581-1590
• /
• 2001

This paper asserted more developmental model of WSD(word sense disambiguation) than J. Hur(2000)'s WSD model. This model suggested an improved statistical homonym disambiguation Model based on Bayes Theory. This paper using semantic information(co-occurrence data) obtained from definitions of part of speech(POS) tagged UMRD-S(Ulsan university Machine Readable Dictionary(Semantic Tagged)). we extracted semantic features in the context as nouns, predicates and adverbs from the definitions in the korean dictionary. In this research, we make an experiment with the accuracy of WSD system about major nine homonym nouns and new seven homonym predicates supplementary. The inner experimental result showed average accuracy of 98.32% with regard to the most Nine homonym nouns and 99.53% for the Seven homonym predicates. An Addition, we save test on Korean Information Base and ETRI's POS tagged corpus. This external experimental result showed average accuracy of 84.42% with regard to the most Nine nouns over unsupervised learning sentences from Korean Information Base and ETRI Corpus, 70.81 ％ accuracy rate for the Seven predicates from Sejong Project phrase part tagging corpus (3.5 million phrases) too.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.539-542
• /
• 2004

배낭 문제는 조합 최적화 문제로서, 다항 시간(polynomial time)에 풀리지 않는 NP-hard 문제이다 이 문제를 해결하기 위해 기존에는 DNA 컴퓨팅 기법과 GA 등을 사용하여 해결하였다. 하지만 기존의 방법들은 DNA의 정확한 특성을 고려하지 않아, 실제 실험과의 결과 차이가 발생하고 있다. 본 논문에서는 DNA 컴퓨팅 실험 과정에서 발생하는 DNA 조작 오류를 최소화하고, 보다 정확한 예측을 위해 함수 언어인 Haskell을 이용한 코드 최적화 DNA-Haskell을 제안한다. 코드 최적화 DNA-Haskell은 배낭 문제 중 (0,1)-배낭 문제에 적용하였고, 그 결과 기존의 DNA 컴퓨팅 방법보다 실험적 오류를 최소화하였으며, 또한 적합한 해를 빠른 시간 내에 찾을 수 있었다.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1021-1026
• /
• 2017

Mixture experiments are commonly encountered in many fields including food, chemical and pharmaceutical industries. In mixture experiments, measured response depends on the proportions of the components present in the mixture and not on the amount of the mixture. Statistical analysis of the data from mixture experiments has mainly focused on a continuous response variable. In the example of quantal response data in mixture experiments, however, the tumor incidence data have been analyzed in Chen et al. (1996) to study the effects of 3 dietary components on the expression of mammary gland tumor. In this paper, we compared the logistic regression models with linear predictors such as second degree Scheffe polynomial model, Becker model and Akay model in terms of classification accuracy.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2009.05a
• /
• pp.67-70
• /
• 2009

본 논문에서는 생리신호를 이용한 감성인식의 정확도를 높이기 위한 개인화 방안에 대해 연구하였다. 이전연구 결과인 TDP(Time Dependent Parameters)를 이용한 감성인식방법은 자율신경계 반응을 4 단계로 세분화하여 감성인식의 정확도를 높일 수 있었다. 하지만 평균 정확도는 향상된 반면 개인별로 정확도의 개인차가 발생하였다. 본 연구는 개인차를 줄이기 위해서, 개인의 반응에 따라 감성인식을 위한 룰베이스가 변화하는 적응적 TDP 알고리즘을 개발하였다. 시각자극을 이용한 감성유발 실험결과를 분석하여 감성인식 개인차가 감소하였는지 확인하였다. 실험은 4 명을 대상으로 하였으며 한 명당 24번의 시각 자극을 제시하여 96개의 데이터가 수집되었다. 데이터는 자율신경계 반응과 주관적 감성을 측정하였나 TDP 를 이용한 분석과 적응적 TDP 분석방법으로 감성을 인식한 결과를 비교한 결과 평균 정확도는 증가하지 않았지만 전반적인 정확도 수준은 상승하는 것을 확인하였다. 그러므로 적응적 TDP를 이용할 경우 개인차를 줄일 수 있다는 것을 확인하였다.

• The Journal of the Korea Contents Association
• /
• v.16 no.9
• /
• pp.117-126
• /
• 2016

This study examined the relative effects of different feedback delivery method (face-to-face vs. e-mail) and reinforcement history on work performance. Participants were asked to work on a simulated mobile phone assembly task. They performed for 30 minutes per session and attended 4 sessions. The dependents variable was the percentage of correctly completed work tasks. Of 100 participants recruited, 50 had a reinforcement history and another 50 had no reinforcement history with the feedback provider in this study. The participants in each group were randomly assigned into two experimental conditions: face-to-face feedback and e-mail feedback. The results showed that for the participants who had reinforcement history, the two feedback delivery methods did not produce a significant difference in the percentage of correctly completed work tasks. However, for those who had no reinforcement history, the two feedback methods did produce a significant difference.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.4
• /
• pp.501-506
• /
• 2008

Statistical POS tagging is prone to error, because of the inherent limitations of statistical data, especially single source of data. Therefore it is widely agreed that the possibility of further enhancement lies in exploiting various knowledge sources. However these data sources are bound to be inconsistent to each other. This paper shows the possibility of using maximum entropy model to Korean language POS tagging. We use as the knowledge sources n-gram data and trigger pair data. We show how perplexity measure varies when two knowledge sources are combined using maximum entropy method. The experiment used a trigram model which produced 94.9% accuracy using Hidden Markov Model, and showed increase to 95.6% when combined with trigger pair data using Maximum Entropy method. This clearly shows possibility of further enhancement when various knowledge sources are developed and combined using ME method.

• Journal of the Korean Institute of Gas
• /
• v.25 no.5
• /
• pp.88-96
• /
• 2021

Advanced Internet of Things (IoT) technology is being used in various ways for the safety of the energy industry. At the center of safety measures, drones play various roles on behalf of humans. Drones are playing a role in reaching places that are difficult to reach due to large-scale facilities and space restrictions that are difficult for humans to inspect. In this study, the accuracy and completeness of movement of dangerous facilities were tested using drone images, and it was confirmed that the movement recognition accuracy was 100%, the average data analysis accuracy was 95.8699%, and the average completeness was 100%. Based on the experimental results, a future-oriented facility risk analysis system combined with ICT technology was implemented and presented. Additional experiments with diversified conditions are required in the future, and ICT convergence analysis system implementation is required.

• The Journal of Korean Academy of Prosthodontics
• /
• v.59 no.1
• /
• pp.27-35
• /
• 2021

Purpose: The purpose of this study is to compare the accuracy of digital scans of implants according to different shapes of scanbodies, and to compare the accuracy of library merging according to different oral exposure height. Materials and methods: A master model with a single tooth edentulous site was prepared. For the first experiment, three types of intraoral scanbodies were prepared, divided into three groups, and the following experiments were conducted for each group: An internal hex implant was placed. The master model with the scanbody connected was scanned with a model scanner, and a master reference file (control group) was created. 10 files (experimental group) were created by performing 10 consecutive scans with an intraoral scanner. After superimposing the control and experimental groups, the following values were calculated: 1) Distance deviation of a designated point on the scanbody 2) Angle deviation of the major axis of the scanbody. For the second experiment, the scanbody scan data were prepared in 6 different heights. Library files were merged with each of the scan data. The distance and angular deviation were calculated using the 7 mm scan data as control group. Results: In the first experiment, there were no significant differences between A and B (P=.278), B and C (P=.568), and C and A (P=.711) in the distance deviations. There were no significant differences between A and B (P=.568), B and C (P=.546), and C and A (P=.112) in the angular deviations. Also, the scanbody showed significantly higher library merging accuracy in the groups with high oral exposure height (P<.5). Conclusion: There were no significant differences in scan accuracy according to the different shapes of scanbodies, and the accuracy of library merging increased according to exposure height of the scanbody in the oral cavity.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.2
• /
• pp.141-146
• /
• 2016

In this paper, we apply limited area mask operation and active contour model to accurately detect tongue area outline in tongue diagnosis system. To accurately analyze the properties of the tongue, first, the tongue area to be detected. Therefore an effective segmentation method for detecting the edge of tongue is very important. It experimented with tongue image DB consists of 20~30 students 30 people. Experiments on real tongue image show the good performance of this method. Experimental results show that the proposed method extracts object boundaries more accurately than existing methods without mask operation.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.163-170
• /
• 2001

웹에서 정보를 찾고자 하는 사용자들을 돕기 위해서는 조직화된 방법으로 검색 결과들을 제시하는 것이 바람직하다. 이러한 목적을 위해, 문서 클러스터링 기법들이 제안되었다. 문서 클러스터링은 사용자들이 관심의 대상이 되는 문서들을 더욱 쉽게 배치할 수 있게 하고, 검색된 문서집합에 대한 개관을 손쉽게 얻을 수 있게 한다. 클러스터링 결과로 주어지는 각 클러스터의 주제를 사용자들이 빠르게 파악할 수 있게 하려면 클러스터 제목을 표현하는 문제가 중요시 된다. 본 연구에서는, 웹 디렉토리의 계층적 구조를 사용하여 자동으로 클러스터 제목을 생성하는 방법을 제안한다. 이 방법은 대상이 되는 클러스터에 있는 문서들의 내용과 부합되는 계층상의 노드를 계층구조 상에서 찾아내어, 계층구조의 루트로부터 그 노드에 이르는 경로명을 클러스터의 제목으로 사용자에게 제시하도록 한다. 본 연구에서 제안한 모델은 '야후' 디렉토리를 사용하여 실험되었다. 실험 결과, 실험대상 클러스터의 본래 제목과 정확하게 일치하는 제목을 찾을 수 있는 경우의 정확률이 57.5% 의미적으로 본래 제목에 부합되는 제목을 찾을 수 있는 경우의 정확률이 대략 90%에 이른다는 것을 알 수 있었다.