• International Journal of Contents
• /
• v.12 no.2
• /
• pp.24-30
• /
• 2016

Ontology is known to be one of the most important technologies in achieving semantic web. It is critical as it represents the knowledge in a machine readable state. World Wide Web Consortium (W3C) has been contributing to the development of ontology for the last several years. However, the recommendation of W3C left out HTML despite the massive amount of information it contains. Also, it is difficult and time consuming to keep up with all the technologies especially in the case of constructing ontology. Thus, we propose a module and methods that reuse HTML documents, extract necessary information from HTML tags and mapping it to OWL 2. We will be combining two kinds of approaches which will be the structural refinement for making an ontology skeleton and linguistic approach for adding detailed information onto the skeleton.

• Journal of KIBIM
• /
• v.9 no.2
• /
• pp.21-28
• /
• 2019

Recently, many scan projects are gradually increasing for maintenance, construction. The scan data contains useful data, which can be generated in the target application from the facility, space. However, modeling the scan data required for the application requires a lot of cost. In example, the converting 3D point cloud obtained from scan data into 3D object is a time-consuming task, and the modeling task is still very manual. This research proposes Scan-to-Geometry Mapping Rule Definition (S2G-MD) which maps point cloud data to geometry for irregular building plane objects. The S2G-MD considers user use case variability. The method to define rules for mapping scan to geometry is proposed. This research supports the reverse engineering semi-automatic process for the building planar geometry from the user perspective.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.5
• /
• pp.17-37
• /
• 2003

This paper presents the ASVO (Automation-considered SVO) Logic that can be used for verifying authentication and key distribution protocols. The ASVO logic was designed for automatic verification, in a way to modify the SVO logic, one of the most famous authentication logics. The ASVO logic is syntactically and semantically sound, and requires relatively simple verification steps. Also we implemented the Isabelle/ASVO system which supports semi-automated verification, by using the Isabelle/Isar system.

• International Journal of Advanced Culture Technology
• /
• v.11 no.3
• /
• pp.89-93
• /
• 2023

This study is a case study of the Ukrainian military's asymmetric combat performance method. The composition of this study is as follows. First, it presented the background for the outbreak of war in the macroscopic framework of Russia-Ukraine. Second, the Ukraine-Russia war, which broke out in February 2022, presented the justification for the study, that is, the direction of the Ukrainian military's asymmetric combat performance in terms of microscopic aspects of the study, and detailed analysis of precision strikes using commercial drones and advanced sensors. Finally, we covered in-depth the case of Ukrainian troops who attempted to attack Russian tanks using semi-automatic laser homing technology. Therefore, the Korean military organization also suggested the justification for gradually introducing and utilizing the system for the Ukrainian military's asymmetric combat performance method, and related follow-up studies should be actively conducted following this study.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.481-484
• /
• 2019

최근 한국어 정보처리를 위한 대용량 언어분석 표준 말뭉치(GS:Gold Standard Set)를 구축하고 이를 공유·확산하기 위한 국가차원의 지원이 이뤄지고 있다. 본 연구는 이러한 사업의 일환으로, 현재 국내에서 개발된 다양한 한국어 언어분석 모듈을 활용하여 공통 정답셋을 구축하기 위한 방법론을 제시하고자 한다. 특히, 대량의 학습셋을 구축하기 위해 다수의 모듈(N-modules)로부터 제시된 후보 정답을 참조, 오류 형태를 분류하여 주요 유형을 반자동으로 보정함으로써 수작업을 최소화하였다. 본 연구에서는 우선 첫 단계인 형태소 분석 모듈 적용 결과를 토대로 표준 말뭉치를 구축한 결과에 대해 논하고자 한다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.287-291
• /
• 2020

본 연구에서는 한국어 의미 표상 자원 구축과 의미 파싱 성능 향상을 위한 데이터 자동 증강 방법을 제안하고 수동 구축 결과 대비 자동 변환 정확도를 보인다. 지도 학습 기반의 AMR 파싱 모델이 유의미한 성능에 도달하려면 대량의 주석 데이터가 반드시 필요하다. 본 연구에서는 기성 언어 분석 기술 또는 기존에 구축된 말뭉치의 주석 정보를 바탕으로 Semi-AMR 데이터를 변환해내는 알고리즘을 제시하며, 자동 변환 결과는 Gold-standard 데이터에 대해 Smatch F1 0.46의 일치도를 보였다. 일정 수준 이상의 정확도를 보이는 자동 증강 데이터는 주석 프로젝트에 소요되는 비용을 경감시키는 데에 활용될 수 있다.

• Investigative Magnetic Resonance Imaging
• /
• v.17 no.3
• /
• pp.200-206
• /
• 2013

Purpose : To assess the effect of applying an automated heart model based measurements of left ventricle (LV) and compare with manual and semi-automated measurements at Cardiovascular MR Imaging. Materials and Methods: Sixty-two patients who underwent cardiac 1.5T MR imaging were included. Steady state free precession cine images of 20 phases per cardiac cycle were obtained in short axis views and both 2-chamber and 4-chamber views. Epicardial and endocardial contours were drawn in manual, automated, and semi-automated ways. Based on these acquired contour sets, the end-diastolic (ED) and end-systolic (ES) volumes, ejection fraction (EF), systolic volume (SV) and LV mass were calculated and compared. Results: In EDV and ESV, the differences among three measurement methods were not statistically significant (P = .399 and .145, respectively). However, in EF, SV, and LV mass, the differences were statistically significant (P=.001, <001, <001, respectively) and the measured value from automated method tend to be consistently higher than the values from other two methods. Conclusion: An automatic heart model-based method grossly overestimate EF, SV and LV mass compared with manual or semi-automated methods. Even though the method saves a considerable amount of efforts, further manual adjustment should be considered in critical clinical cases.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.91-98
• /
• 2023

Recently, according to development of artificial intelligence, a wide range of industry being automatic and optimized. Also we can find out some research of using supervised learning for deteceting defect of railway in domestic rail industry. However, there are structures other than rails on the track, and the fastener is a device that binds the rail to other structures, and periodic inspections are required to prevent safety accidents. In this paper, we present a method of reducing cost for labeling using semi-supervised and transfer model trained on rail fastener data. We use Resnet50 as the backbone network pretrained on ImageNet. At first we randomly take training data from unlabeled data and then labeled that data to train model. After predict unlabeled data by trained model, we adopted a method of adding the data with the highest probability for each class to the training data by a predetermined size. Futhermore, we also conducted some experiments to investigate the influence of the number of initially labeled data. As a result of the experiment, model reaches 92% accuracy which has a performance difference of around 5% compared to supervised learning. This is expected to improve the performance of the classifier by using relatively few labels without additional labeling processes through the proposed method.

• Asian-Australasian Journal of Animal Sciences
• /
• v.29 no.6
• /
• pp.879-886
• /
• 2016

As is in all economic activities, the highest yield per unit area is the main goal in animal production, while addressing the temperamental needs of animals often is ignored. Animal welfare is not only an ethical fact; it also has an economic value. Spatial environmental enrichment contributes positively to animal welfare by addressing their behavioral and mental requirements. The present study was conducted to determine the effects of weaning and spatial environmental arrangements on behaviors of goat-kids. Experimental groups were arranged in structured and unstructured spatial environments. Roughage feeder, semi-automatic concentrate feeder, bunk, bridge, and wood block were placed in the structured environment. No equipment was placed in the unstructured environment and paddock sides were enclosed with an iron sheet to prevent bipedal stance and to provide environmental isolation. In the study 10 male and 10 female Turkish Saanen goat kids were used in each group. Spatial environmental arrangements did not have significant impacts on the growth performance of kids (p>0.05). All objects in the structured group were accepted by the kids. Average use ratios of roughage feeder, semi-automatic concentrate feeder, bunk, bridge and wood block were observed as 19.3%, 14.0%, 12.6%, 3.8%, and 0.7%, respectively. There were significant differences between before- and after-weaning in use of all objects except for underneath bridge ($p{\leq}0.05$). Concentrate feed consumption, locomotion, and resting behaviors in kids showed significant differences by structural group and growth period. Roughage consumption was similar between groups, while it differed by growth period ($p{\leq}0.05$). Interaction frequency was significantly higher in structured group (p = 0.0023). Playing behavior significantly differentiated based on the growth period rather than on groups ($p{\leq}0.05$). Playing behavior significantly decreased after weaning. Abnormal oral activity was significantly higher in the structured group before weaning ($p{\leq}0.05$). Despite there being no installations facilitating climbing and bipedal stance, the kids of the unstructured group were able to exhibit 1/3 as much bipedal stance behavior as the kids of the structured group through leaning over slippery paddock wall or over their groupmates. Bipedal stance behavior of unstructured group was similar before and after weaning, while bipedal stance behavior before weaning was about 2 times that of after weaning in structured group. It was concluded that unstructured environmental arrangement limited the behavior repertoire of the goat kids.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.63-70
• /
• 2006

In general, a corpus contains lots of linguistic information and is widely used in the field of natural language processing and computational linguistics. The creation of such the corpus, however, is an expensive, labor-intensive and time-consuming work. To alleviate this problem, annotation tools to build corpora with much linguistic information is indispensable. In this paper, we design and implement an annotation tool for establishing a Korean dependency tree-tagged corpus. The most ideal way is to fully automatically create the corpus without annotators' interventions, but as a matter of fact, it is impossible. The proposed tool is semi-automatic like most other annotation tools and is designed to edit errors, which are generated by basic analyzers like part-of-speech tagger and (partial) parser. We also design it to avoid repetitive works while editing the errors and to use it easily and friendly. Using the proposed annotation tool, 10,000 Korean sentences containing over 20 words are annotated with dependency structures. For 2 months, eight annotators have worked every 4 hours a day. We are confident that we can have accurate and consistent annotations as well as reduced labor and time.