• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.555-562
• /
• 2016

Semantic Role Labeling (SRL) is to determine the semantic relation of a predicate and its argu-ments in a sentence. But Korean semantic role labeling has faced on difficulty due to its different language structure compared to English, which makes it very hard to use appropriate approaches developed so far. That means that methods proposed so far could not show a satisfied perfor-mance, compared to English and Chinese. To complement these problems, we focus on suffix information analysis, such as josa (case suffix) and eomi (verbal ending) analysis. Korean lan-guage is one of the agglutinative languages, such as Japanese, which have well defined suffix structure in their words. The agglutinative languages could have free word order due to its de-veloped suffix structure. Also arguments with a single morpheme are then labeled with statistics. In addition, machine learning algorithms such as Support Vector Machine (SVM) and Condi-tional Random Fields (CRF) are used to model SRL problem on arguments that are not labeled at the suffix analysis phase. The proposed method is intended to reduce the range of argument instances to which machine learning approaches should be applied, resulting in uncertain and inaccurate role labeling. In experiments, we use 15,224 arguments and we are able to obtain approximately 83.24% f1-score, increased about 4.85% points compared to the state-of-the-art Korean SRL research.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.527-534
• /
• 2016

The emergent information society requires the talent for creative thinking based on problem-solving skills and comprehensive thinking rather than simple memorization. Therefore, the Korean curriculum has also changed into the direction of the creative thinking through increasing short-answer questions that can determine the overall thinking of the students. However, their scoring results are a little bit inconsistency because scoring short-answer questions depends on the subjective scoring of human raters. In order to alleviate this point, an automated scoring system using a machine learning has been used as a scoring tool in overseas. Linguistically, Korean and English is totally different in the structure of the sentences. Thus, the automated scoring system used in English cannot be applied to Korean. In this paper, we introduce an automated scoring system for Korean short-answer questions using predictability and unanimity. We also verify the practicality of the automatic scoring system through the correlation coefficient between the results of the automated scoring system and those of human raters. In the experiment of this paper, the proposed system is evaluated for constructed-response items of Korean language, social studies, and science in the National Assessment of Educational Achievement. The analysis was used Pearson correlation coefficients and Kappa coefficient. Results of the experiment had showed a strong positive correlation with all the correlation coefficients at 0.7 or higher. Thus, the scoring results of the proposed scoring system are similar to those of human raters. Therefore, the automated scoring system should be found to be useful as a scoring tool.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.2
• /
• pp.113-118
• /
• 2013

The current GPU virtualization techniques incur large overheads when executing application programs mainly due to the fine-grain time-sharing scheduling of the GPU among multiple Virtual Machines (VMs). Besides, the current techniques lack of portability, because they include the APIs for the GPU computations in the VM monitor. In this paper, we propose a low overhead and high performance GPU virtualization approach on a heterogeneous HPC system based on the open-source Xen. Our proposed techniques are tailored to the bio applications. In our virtualization framework, we allow a VM to solely occupy a GPU once the VM is assigned a GPU instead of relying on the time-sharing the GPU. This improves the performance of the applications and the utilization of the GPUs. Our techniques also allow a direct pass-through to the GPU by using the IOMMU virtualization features embedded in the hardware for the high portability. Experimental studies using microbiology genome analysis applications show that our proposed techniques based on the direct pass-through significantly reduce the overheads compared with the previous Domain0 based approaches. Furthermore, our approach closely matches the performance for the applications to the bare machine or rather improves the performance.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.1
• /
• pp.9-18
• /
• 2018

Recently, the use of image stitching technology has been increasing as the number of contents based on virtual reality increases. Image Stitching is a method for matching multiple images to produce a high resolution image and a wide field of view image. The image stitching is used in various fields beyond the limitation of images generated from one camera. Image Stitching detects feature points and corresponding points to match multiple images, and calculates the homography among images using the RANSAC algorithm. Generally, corresponding points are needed for calculating conversion relation. However, the corresponding points include various types of noise that can be caused by false assumptions or errors about the conversion relationship. This noise is an obstacle to accurately predict the conversion relation. Therefore, RANSAC algorithm is used to construct an accurate conversion relationship from the outliers that interfere with the prediction of the model parameters because matching methods can usually occur incorrect correspondence points. In this paper, we propose an algorithm that extracts more accurate inliers and computes accurate transformation relations by using correspondence point relation information used in RANSAC algorithm. The correspondence point relation information uses distance ratio between corresponding points used in image matching. This paper aims to reduce the processing time while maintaining the same performance as RANSAC.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.11
• /
• pp.367-376
• /
• 2020

The fourth industrial revolution will greatly contribute to the entry of objects into an intelligent society through technologies such as big data and an artificial intelligence. Through the revolution, we were able to understand human behavior and awareness, and through the use of an artificial intelligence, we established ourselves as a key tool in various fields such as medicine and science. However, the fourth industrial revolution has a negative side with a positive future. In this study, an analysis was conducted using text mining techniques based on unstructured big data collected through social media. We wanted to look at keywords related to the fourth industrial revolution by year (2016, 2017 and 2018) and understand the meaning of each keyword. In addition, we understood how the keywords related to the Fourth Industrial Revolution changed with the change of the year and wanted to use R to conduct a Keyword Analysis to identify the recognition flow closely related to the Fourth Industrial Revolution through the keyword flow associated with the Fourth Industrial Revolution. Finally, people's perceptions of the fourth industrial revolution were identified by looking at the positive and negative feelings related to the fourth industrial revolution by year. The analysis showed that negative opinions were declining year after year, with more positive outlook and future.

• International Journal of Highway Engineering
• /
• v.12 no.4
• /
• pp.79-86
• /
• 2010

This study deals with the roundabout installed at the diamond interchange. The goal is to analyze the relative effectiveness of roundabout to signalized intersection. In pursuing the above, this study gives the particular attentions to comparing the performances using VISSIM software. The main results analyzed are as follows. First, the single and double roundabouts are analyzed to have the rapid change of delay in the case of total entering vehicles of more than 2,400pcph for directional rate 10:5:5, 2,800pcph for 10:8:2, and 4,400pcph for 10:2:8. Second, the roundabouts are evaluated to be more effective than signalized intersection in the case of total entering vehicles of less than 4,000pcph for directional rate 10:5:5, and 4,400pcph for 10:2:8. In the case of directional rate 10:8:2, double roundabout of total entering vehicles less than 5,600pcph is analyzed to be more effective than signalized intersection. Finally, the performance of double roundabout is analyzed to be very similar to that of single roundabout in the case of total entering vehicles less than 4,400pcph. However, the double roundabout is evaluated to be more effective than single in the case of total entering vehicles more than 4,400pcph.

• Proceedings of the KSRS Conference
• /
• 2000.04a
• /
• pp.166-171
• /
• 2000

밀티빔 음향 측심기 (Multibeam Echo Sounder)는 탐사선에 수직방향으로 해저면을 주사(Swath)하여, 한번의 송수신(Ping)으로 다중의 빔자료를 얻을 수 있는 측심기로, 해저면에 반사되어 되돌아오는 음파의 음압을 기록하고, 사이드 스캔 소나 자료도 동시에 취득하는 기능을 가지고 있으므로, 측심된 해저 지형(Bathymetry)과 해저 지형을 덮고 있는 해저면의 퇴적 상황(Sediment Environment)도 동시에 얻을 수 있는 다목적 측심기이다. 본 논문에서는 L3사의 Sea Beam 2100 멀티빔 음향 측심기를 통해 얻은 자료를 처리하여, 3차원 공간 데이터인 DEM(Digital Elevation Model)을 생성하고, VRML을 이용한 웹상에서의 해저 지형 가시화를 통해, 세계 어느 곳에서나 웹을 통하여 쉽게 정보를 공유할 수 있는 3차원 해저 지리 정보 시스템의 구현을 목적으로 한다. 멀티빔 음향 측심기를 통해 얻어진 자료는 항해 자료 보정, 음속 보정, 빔 좌표 계산과 분리, 오측심 자료 제거, 조석 보정 등의 단계를 거쳐 측심자료의 정확도 및 신뢰도를 높이는 과정을 거치게 된다. 보정된 멀티빔 음향 측심자료는 무작위 점 사상(Point Topology)으로 산재 되어 있는 빔 자료를 임의의 단위영역으로 변환하는 과정을 거쳐야 하는데, 이 과정을 격자화라고 한다. 자료의 격자화를 통해 3차원 공강 데이터인 DEM 파일을 제작하고, 이 DEM 파일과 음압 영상을 이용해 웹상에서의 3차원 해저 지형의 가시화를 실현한다. 웹상에서의 3차원 지형 가시화에서 방대한 양의 지형 데이터는 데이터 전송 시간과 렌더링 시간에 치명적인 문제이다. 따라서, 렌더링 시간과 데이터 전송 시간을 단축시키기 위한, 지형 자료의 LOD(Level of Detail)를 통해, VRML을 이용한 보다 효과적인 웹상에서의 3차원 해저 지형의 가시화를 실현한다.면 기업은 고객으로 공간적인 제약으로 인한 불신을 불식시키는 신뢰감을 주게 된다. 이러한 고객서비스 향상과 물류비용 절감은 사이버 쇼핑몰이 전국 어디서나 우리의 안방에서 자연스럽게 점할 수 있는 상황을 만들 것이다.SP가 도입되어, 설계업무를 지원하기위한 기본적인 시스템 구조를 구상하게 된다. 이와 함께 IT Model을 구성하게 되는데, 객체지향적 접근 방법으로 Model을 생성하고 UML(Unified Modeling Language)을 Tool로 사용한다. 단계 4)는 Software Engineering 관점으로 접근한다. 이는 최종산물이라고 볼 수 있는 설계업무 지원 시스템을 Design하는 과정으로, 시스템에 사용될 데이터를 Design하는 과정과, 데이터를 기반으로 한 기능을 Design하는 과정으로 나눈다. 이를 통해 생성된 Model에 따라 최종적으로 Coding을 통하여 실제 시스템을 구축하게 된다.the making. program and policy decision making, The objectives of the study are to develop the methodology of modeling the socioeconomic evaluation, and build up the practical socioeconomic evaluation model of the HAN projects including scientific and technological effects. Since the HAN projects consists of 18 subprograms, it is difficult In evaluate all the subprograms simultaneously. Despite, each program is being performed under the category of HAN projects, so the common soci

• Tunnel and Underground Space
• /
• v.15 no.2 s.55
• /
• pp.119-128
• /
• 2005

Although the evaluation of the mechanical properties and behavior of jointed rock masses is very important for the design of tunnel and underground openings, it has always been considered the most difficult problem. One of the difficulties in describing the rock mass behavior is the selection of the appropriate constitutive model. This limitation may be overcome with the progress in discrete element software such as PFC, which does not need the user to prescribe a constitutive model for rock mass. In this paper, a 30\;m\;\times\;30\;m\;\times\;30\;m m jointed rock mass of road tunnel site was analyzed. h discrete fracture network was developed from the joint geometry obtained from core logging and surface survey. Using the discontinuities geometry from the DFN model, PFC simulations were carried out, starting with the intact rock and systematically adding the joints and the stress-strain response was recorded for each case. With the stress-strain response curves, the mechanical properties of jointed rock masses were determined. As expected, the presence of joints had a pronounced effect on mechanical properties of the rock mass. More importantly, getting the mechanical response of the PFC model doesn't require a user specified constitutive model.

• International Journal of Highway Engineering
• /
• v.18 no.2
• /
• pp.73-82
• /
• 2016

PURPOSES: The national highways and expressways in Korea constitute a total length of 17,951 km. Of this total length of pavement, the asphalt pavement has significantly deteriorated, having been in service for over 10 years. Currently, hot in-place recycling (HIR) is used as the rehabilitation method for the distressed asphalt pavement. The deteriorated pavement becomes over-heated, however, owing to uncontrolled heating capacity during the pre-heating process of HIR in the field. METHODS: In order to determine the appropriate heating method and capacity of the pre-heater at the HIR process, the heating temperature of asphalt pavement is numerically simulated with the finite element software ABAQUS. Furthermore, the heating transfer effects are simulated in order to determine the inner temperature as a function of the heating system (IR and wire). This temperature is ascertained at $300^{\circ}C$, $400^{\circ}C$, $500^{\circ}C$, $600^{\circ}C$, $700^{\circ}C$, and $800^{\circ}C$ from a slab asphalt specimen prepared in the laboratory. The inner temperature of this specimen is measured at the surface and five different depths (1 cm, 2 cm, 3 cm, 4 cm, and 5 cm) by using a data logger. RESULTS: The numerical simulation results of the asphalt pavement heating temperature indicate that this temperature is extremely sensitive to increases in the heating temperature. Moreover, after 10 min of heating, the pavement temperature is 36%~38% and 8%~10% of the target temperature at depths of 25 mm and 50 mm, respectively, from the surface. Therefore, in order to achieve the target temperature at a depth of 50 mm in the slab asphalt specimen, greater heating is required of the IR system compared to that of the gas. CONCLUSIONS : Numerical simulation, via the finite element method, can be readily used to analyze the appropriate heating method and theoretical basis of the HIR method. The IR system would provide the best heating method and capacity of HIR heating processes in the field.

• Journal of KIISE:Software and Applications
• /
• v.28 no.3
• /
• pp.224-238
• /
• 2001

본 논문은 ATM(Abstract Timed Machine)으로 명세된 실시간 시스템을 검증하기 위한 방법을 기술한다. ATM은 임무 위급 시스템인 실시간 시스템을 명세, 분석, 검증하기 위한 정형기법이다. ATM은 모드와 전이, 포트로 구성되어 있으며 모드는 머신의 압축된 상태를 표현한다. 전이는 하나의 모드에서 다른 모드로의 전환을 나타내며 조건과 이벤트로 구성되어 있다. 포트는 ATM간의 상호작용을 위한 진입을 표현한다. 다른 정형기법과 비교하여 ATM은 소프트웨어의 순환공학 과정에서 사용하기 위해 설계되었다. 역공학 측면에서 볼 때 ATM은 계산 논리뿐만 아니라 실시간 시스템의 실제 소스코드에 있는 설계나 환경정보를 표현할 수 있다. 이러한 목적을 위해 ATM의 모드는 계산모드, 추상화 모드, 주제모드로 구분된다. 계산 모드는 코드 상에서의 논리와 계산을 나타내며 추상화 모드는 모드와 전이의 블록을 하나의 ATM으로써 표현한다. 대개의 경우, 이것은 코드 상에서의 블록을 ATM내 하나의 모드로 나타낼 때 사용한다. 주제 모드는 예외나 주기적 동작 등과 같은 다수의 ATM의 주제를 표현한다. 실시간 시스템을 검증하기 위해 시스템의 소스 코드는 역명세 과정을 통하여 ATM으로 표현된다. 검증은 ATM에 대한 도달성 그래프를 생성하는 것에 의해 수행된다. 도달성 그래프는 상태와 시간을 추상화되고 압축된 형태로 표현할 수 있으며 그 결과 시간 속성을 지닌 상태 공간을 감소시킬 수 있다. 또한 시스템의 교착상태를 쉽게 발견할 수 있다. 본 논문은 ATM과 실행 모델, 도달성 그래프, 검증을 위한 속성 등을 기술하며 이들을 다른 정형 방법들과 예제를 통하여 비교한다.수 있다. 모피우스는 헤더나 광고와 같은 불필요한 정보들을 제거하는 별도의 단계를 거치지 않으므로 wrapper를 빠르게 생성한다. 궁극적으로 모피우스는 새로운 웹 상점을 사용자가 자유롭게 추가, 삭제할 수 있는 환경을 제공한다.X>와 반응시킬 경우에는 반응식 c에 의거 진행됨을 예측할 수 있었다.의거 진행됨을 예측할 수 있었다.이 거의 산화되지 않았고, $700^{\circ}C$에서도 ZnS와 ZnO 상이 공존한 것으로 보아 SnO$_2$코팅이 ZnS의 산화를 억제하는 것으로 나타났다.pplied not only to the strike system in the RSC circle, but also to the logistics system in the SLC circle. Thus, the RSLC model can maximize combat synergy effects by integrating the RSC and the SLC. With a similar logic, this paper develops "A Revised System of Systems with Logistics (RSSL)" which combines "A New system of Systems" and logistics. These tow models proposed here help explain several issues such as logistics environment in future warfare, MOE(Measure of Effectiveness( on logistics performance, and COA(Course of Actions)