• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Korean Journal of Social Welfare
• /
• v.69 no.1
• /
• pp.33-57
• /
• 2017

This study examines the effect of working conditions for care workers on the care quality in long-term care facilities, particularly the coexisting perspective on publicness and the marketization of Long-term care services in South Korea brings about. Prior studies have not identified a causal relationship between working conditions and the care quality, only explained cause of a low-wage labor market and low productivity of social services. Theoretical relevance of working conditions and service quality on Long-term care in Korea is to view from a integrated care model by Daly and Lewis(2002). A nonproportional stratified sampling procedure was used to consider Long-term care facility's ownership. A merged dataset combining surveys from 248 Long-Term Care facilities and online resources from NHIC administrative was used and analyzed by multiple regression. The analysis results is showed as follows. Overall, organizations with better working conditions, having higher wage, having greater a fringe benefit, being skills development and training are likely to have good care quality in each area. This research shows that the working conditions, rewards and support to care workers of organizational culture in the normative dimension beyond the minimum standard on labor market policy and evaluation system by government regulations have a positive impact on Long-term care quality.

• Journal of The Korean Association For Science Education
• /
• v.38 no.2
• /
• pp.135-145
• /
• 2018

In this study, we investigated the argumentations of group and classroom discussions in socioscientific issues (SSI) discussion classes. Twenty-seven high school students participated in the SSI discussion classes on nuclear power generation. We observed and recorded the classes and also conducted semi-structured interviews. For the analyses, we revised a previous framework that was developed to analyze dialogic argumentations in the context of SSI. The analyses of the results indicated that there were more discourse schemes in the classroom discussions than the group discussions which are related to awareness and openness to multiple perspectives, evidence based reasoning, and on-going inquiry and skepticism. And there were few discourse schemes related to moral and ethical sensitivity in the group and classroom discussions. Various grounds, data, and information were presented in the classroom discussions. Students concentrated on carrying their claims and were not able to sympathize with and accept other opinions. Therefore, there were few discourse schemes to reach consensus. In addition, they perceived classroom discussions as competitive and actively rebutted other claims or grounds. The levels of argumentation were also high in the classroom discussions. The group discussions were held in relaxed atmosphere, and they asked the opponents more for clarification or additional information and evidences. However, classroom discussions were held in serious atmosphere, and they actively queried the validity of the claims or grounds. Based on the results, some suggestions to implement SSI discussion classes were discussed.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1995.10a
• /
• pp.289-295
• /
• 1995

구조물의 동적부하에 대한 동적변형 응답을 정확히 예측하고, Over Design이나 Under Design이 아닌 합리적인 설계방안의 개발은 중요한 과제이다. 동적강도해석이나 소음 승차감과 같은 진동 및 충격에 기인하는 제반 문제를 복잡한 구조물을 대상으로 합리적으로 처리하기 위한 Dynamic Design Analysis는 높은 신뢰성의 추구와 더불어 필요불가결한 기술이 되고 있다. 동적해석 방법으로는 현재 유한요소법이 널리 사용되고 있으며 여러 종류의 범용 프로그램들이 보급되어 있는 실정이다. 그러나 특히 동적문제에 있어서는 형상이나 거동이 복잡한 구조물의 경우, 또는 차량의 차체와 같이 많은 장착물이 부착된 경우에는 유한요소법의 적용이 곤란하여, 지금까지 대처할 수 있는 유용한 방법이 없었다. 따라서 비교적 용이하고 간단하게 적용가능한 진동실험을 기초로 한 구조물의 동적 응답해석 및 설계 방안의 개발이 필요하다. 본 연구에서는 진동시험으로 얻어진 부분구조물의 응답특성과 결합특성으로부터 결합 후의 응답특성을 예측할 수 있는 방법을 전달함수합성이론을 기초로하여 프로그래밍 package화 한다. 그리고 평판구조물에 대하여 진동시험과 컴퓨터 시뮬레이션을 통하여 개발된 방법의 타당성을 검증한다. 또한 실제 차량에서 차체만의 진동시험과 엔진의 자유진동시험에 의한 시험데이터로부터 차체와 엔진이 마운트 결합된 후의 진동특성을 예측한다. 진동시험시에 입력과 출력에 노이즈가 필연적으로 혼입되어 주파수응답함수의 크기(magnitude)와 위상(phase)을 왜곡시킨다. 특히 위상의 왜곡은 복소수연산을 하는 전달함수합성법의 결과에 중요한 영향을 미치게 된다. 본 연구에서는 데이타 획득시 입력과 출력의 시간지연으로 생기는 위상왜곡을 보정하는 방법을 제시하고, 그 개선 정도를 조사한다.는 소견의 확실도로서 가능성을 표현한 것이다. 예를 들면, 진동진폭 스펙트럼상에 2X 성분이 상당히 크게 나타나 정렬불량의 가능성이 0.7 정도라고 판정하는 것 등은 이러한 수치적진리치를 이용하는 방법이다. 그러나 상기의 수치적 표현만으로는 확실도를 한개의 수치로서 대표하게 하는 것은 진단의 정밀도에 문제가 있을 것으로 생각된다. 따라서 언어적진리치가 도입되어 [상당히 확실], [확실], [약간 확실] 등의 언어적인 표현을 이용하여 애매성을 표현하게 되었다. 본 논문에서는 간이진단 결과로부터 추출된 애매한 진단결과중에서 가장 가능성이 높은 이상원인을 복수로 선정하고, 여러 종류의 수치화할 수 없는 언어적(linguistic)인 정보ㄷㄹ을 if-then 형식의 퍼지추론으로 종합하는 회전기계의 이상진단을 위한 정밀진단 알고리즘을 제안하고 그 유용성을 검토한다. 존재하여도 모우드 변수들을 항상 정확하게 구할 수 있으며, 또한 알고리즘의 안정성이 보장된 것이다.. 여기서는 실험실 수준의 평 판모델을 제작하고 실제 현장에서 이루어질 수 있는 진동제어 구조물에 대 한 동적실험 및 FRS를 수행하는 과정과 동일하게 따름으로써 실제 발생할 수 있는 오차나 error를 실험실내의 차원에서 파악하여 진동원을 있는 구조 물에 대한 진동제어기술을 보유하고자 한다. 이용한 해마의 부피측정은 해마경화증 환자의 진단에 있어 육안적인 MR 진단이 어려운 제한된 경우에만 실제적 도움을 줄 수 있는 보조적인 방법으로 생각된다.ofile whereas relaxivity at high field is not affected by τS. On the other hand, the change in τV does not affect low field profile but strongly in fluences on bot

• Journal of the Korean earth science society
• /
• v.28 no.6
• /
• pp.673-685
• /
• 2007

The purpose of this study was to investigate Earth Systems Understandings (Mayer, 1991) and Earth Systems meaning generation reported by science-talented students who participated in a geological field trip. The eight (4 female and 4 male students) field trip reports were randomly selected among all the reports written by twenty eighth-grade students who joined Shiwha-Lake field trip in Korea. The three-step program, including preparation, field trip, and summary, was provided to the students in order to facilitate meaningful learning through outdoor teaming activities. Seven Earth Systems Understandings and thematic types (Keys, 1999) were used to analyze the reports. The results of this study indicated thai aesthetic views and stewardship toward the Earth, which were the most distinguishing characteristics in Earth Systems Education, were reflected on most of the reports. The results also showed that the students tried to represent their understandings in such a type as meaning extension, meaning enhancement, or meaning elaboration. Overall, many students used 'knowledge-telling' process with a long list of observations and facts, whereas a few students used higher-order 'knowledge-transforming' process by coordinating their findings with interpretations and reasoning in their writings.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2000.05a
• /
• pp.303-308
• /
• 2000

용언은 그 어간이 여러 문법소와 결합하면서 자동적 음운 변동을 제외한 형태 변동이 있는가, 없는가에 의해 규칙 용언/불규칙 용언으로 구분할 수 있다. 이러한 불규칙 용언은 심성 어휘집에 어떤 형태로 저장되어 있으며, 규칙 용언과는 어떠한 관계가 있는지, 나아가 실어증 환자의 경우에는 정상인에 비해 어떤 행동장애를 보이며, 장애가 있다면 어느 경로의 손상으로 인한 장애인지를 알아보는 것이 본 연구의 목적이다. 이를 위해 이해성 실어증 환자 한 명과 음어적 실행증 현상을 동반한 경미한 정도의 실어증 환자를 피험자로 하였다. 실험 과제는 단어 채워 넣기 과제(word completion task)를 사용하였다. 즉 주어진 기본형 용언을 검사 문장의 문맥에 맞게 활용하여 채워 넣는 것이다. 실험 결과에 의하면 환자들은 규칙용언의 활용(예. 먹다/먹는)과 불규칙 용언 중 형태를 유지한 채로 활용하는 경우(예. 줍다/줍고)에는 거의 오류가 없었으나, 불규칙 용언이 형태 변화를 겪어야 할 경우(예. 줍다/주워)에는 대부분 오류를 보였다. 또 이때는 기본형(basic form)을 그대로 유지하는 오류 방향성을 관찰할 수 있었다. 이는 그간 문법으로 구분되어 오던 규칙 용언/불규칙 용언의 정보 처리보다는 형태 유지/형태 변화 정보 처리의 영향이 크다는 것을 알 수 있다. 특히 이해성 실어증 환자는 전체적인 오류율이 매우 높았는데, 규칙 용언의 경우에도 오류를 보였다. 이때, 용언의 어간에 해당하는 부분에는 오류가 없고, 뒤에 따르는 내용과의 관계를 파악해야 하는 문법 기능소, 즉 연결 어미에서 오류를 보여 정보의 유지, 통합에 문제가 있다는 기존의 연구와도 일치하는 결과를 나타냈다.환자는 시제 선어말 어미를 선택하는데도 어려움을 보임이 확인되었다. 실험 3 역시 실험 1과 실험2에서와 동일하게 처리의 어려움을 보였다. 이러한 실험 결과들은 국어의 존칭과 시제 선어말 어미가 통사부에서 구(XP)와 결합하여 새로운 구를 형성하는 통사적 접사로 해석할 수 있으며 Grodzinsky의 가설을 지지하는 결과를 보여 줌으로서 국어에서도 AgrP, TP, CP 사이의 통사적 위계가 있음을 뒷받침하는 증거가 된다.전처리한 Group 3에서는 IL-2와 IL-4의 수준이 유의성있게 억제되어 발현되었다 (p < 0.05). 이러한 결과를 통하여 T. denticola에서 추출된 면역억제 단백질이 Th1과 Th2의 cytokine 분비 기능을 억제하는 것으로 확인 되었으며 이 기전이 감염 근관에서 발견되는 T. denticola의 치수 및 치근단 질환에 대한 병인기전과 관련이 있는 것으로 사료된다.을 보였다. 본 실험 결과, $Depulpin^{\circledR}은{\;}Tempcanal^{\circledR}와{\;}Vitapex^{\circledR}$에 비해 높은 세포 독성을 보여주공 있으나, 좀 더 많은 임상적 검증이 필요할 것으로 사료된다.중요한 역할을 하는 것으로 추론할 수 있다.근관벽을 처리하는 것이 필요하다고 사료된다.크기에 의존하며, 또한 이러한 영향은 $(Ti_{1-x}AI_{x})N$ 피막에 존재하는 AI의 함량이 높고, 초기에 증착된 막의 업자 크기가 작을 수록 클 것으로 여겨진다. 그리고 환경의 의미의 차이에 따라 경관의 미학적 평가가 달라진 것으로 나타났다.corner$적 의도에 의한 경관구성의 일면을 확인

• Science of Emotion and Sensibility
• /
• v.10 no.4
• /
• pp.511-521
• /
• 2007

It is very important that design based on preference of consumers who continuously change. Therefore, the method that can decide on the design concept which a consumer can prefer in future points of time that a design is released is necessary. There may be various ways to decide a design concept, but trend analysis is one of the best ways to be able to satisfy consumer preference. The purpose of this study is to provide a process that can give a direction of MP3 player design oriented consumer emotion. For the purpose, we considered about trend analysis as the ways that can present the design direction that can grasp a change of continuous preference, and a consumer can prefer with early bases in future points of time of a consumer. In this empirical research, we decided on design elements and levels of the elements after collecting 228 MP3 players released from 2000 to 2007, and carried out trend analysis through homogeneity analysis by SPSS program. In the result, we knew that future consumers also will regard emotional experience consumption as important. So, MP3 player design will be developed into consumer emotion-oriented design. We predict 4 trends for a future MP3 player design. 1. Development of high-priced MP3 player with various multimedia functions. 2. Development of MP3 player with basic functions. 3. Development of new convergence products with MP3 function. 4. Development of new MP3 player based on flash memory. If designers can infer a future MP3 player design from this design trend results, the probability that can occupy competitive advantage in their competitions will be high. Therefore this study can be useful.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.29 no.3
• /
• pp.272-279
• /
• 2013

The concept of implant stability was basically originated from the relative condition of bone-implant interface and has some meanings for evaluation of that interface. In addtion, it has been used for the investigation of initial bone healing process after fixture installation because a degree of micromotion around interface can affect unfavorable clinical results. The purpose of this study is to investigate the mode of initial bone healing from fixture installation through prospective trial. Thirty fixtures were consecutively installed in mandibles of 26 patients with single tooth loss area and then healing abutment were secured for one-stage surgery meothod. Resonance frequency analysis was performed with one week interval during 12 weeks and periapical radiographs were taken at each month. Although marginal bone level change was not shown through observation period (P>0.05), statistical difference of implant stability was shown through 4 and 6 week (P<0.05) and was not shown after 6 week (P>0.05) according to the bone quality. Initial bone healing process is a successive process of bone resorption and favorable bone healing result might be postulated at 4 week interval after installation through RFA.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.3
• /
• pp.77-84
• /
• 2015

The key objective of this research is to review the effectiveness of spatial regression to identify the influencing factors of spatial distribution patterns of floating population. To this end, global and local spatial autocorrelation test were performed using seoul floating population survey(2014) data. The result of Moran's I and Getis-Ord $Gi^*$ as used in the analysis derived spatial heterogeneity and spatial similarities of floating population patterns in a statistically significant range. Accordingly, Geographically Weighted Regression was applied to identify the relationship between land use attributes and population floating. Urbanization area, green tract of land of micro land cover data were aggregated in to $400m{\times}400m$ grid boundary of Seoul. Additionally public transportation variables such as intersection density transit accessibility, road density and pedestrian passage density were adopted as transit environmental factors. As a result, the GWR model derived more improved results than Ordinary Least Square(OLS) regression model. Furthermore, the spatial variation of applied local effect of independent variables for the floating population distributions.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.