11,891 research outputs found
Woo Min JOO
No full text학위논문(석사)--아주대학교 일반대학원 :산업공학과,2017. 2프로세스 마이닝은 프로세스를 분석 대상으로 삼아 프로세스를 시각화한 프로세스 모델을 도출하고, 도출된 모델을 실제 프로세스 수행과정과 비교함으로써 모델을 평가하며, 모델을 개선하는 활동을 수행해 프로세스에 대한 유용한 정보를 획득하는 것을 목표로 하는 분석 기법이다.
이러한 프로세스 마이닝에서 현실 세계의 프로세스를 잘 반영한 프로세스 모델을 도출하는 것이 가장 중요하다. 하지만 기존의 프로세스 모델들은 액티비티가 수행될 확률이나 액티비티를 수행하는 데 걸리는 시간과 같은 확률적 요소를 반영하고 있지 않은 결정적 프로세스 모델로서 이용할 수 있는 정보가 한정되어 있어 프로세스를 분석하고 이해하는 데 한계가 있다. 또한, 프로세스 모델을 도출하는 것과 더불어 프로세스 모델을 도출하기 위한 효율적인 방법론을 개발하는 것이 중요하다. 프로세스 모델 도출 방법론 중 하나인 유전자 프로세스 마이닝 알고리즘은 진화 과정을 모사한 메타 휴리스틱 알고리즘으로써 노이즈가 있는 데이터를 다룰 수 있지만, 연산 시간이 오래 걸려 대용량 데이터에 적용하기 어렵다는 한계점을 가진다.
본 논문에서는 이러한 문제점들을 개선하기 위해 확률적 프로세스 트리를 정의하고, 타부 서치-유전자 프로세스 마이닝 알고리즘을 제안하였다. 확률적 프로세스 트리는 확률적 요소를 반영하지 않은 기존의 프로세스 트리의 액티비티 노드와 연산자 노드와는 별도로 특정 액티비티 조합이 생성될 확률을 기록한 시퀀스 확률 정보 노드, 하나의 액티비티를 시작해서 마칠 때까지의 시간을 확률 분포로 표현한 단일 액티비티 수행 시간 노드, 여러 액티비티가 연속해서 실행되는 사이에 시스템의 대기 상태에 있는 시간을 확률 분포로 표현하는 인접 액티비티 간 시간 간격 노드를 정의하고 표현함으로써 확률적 요소를 반영하였다. 타부 서치-유전자 프로세스 마이닝 알고리즘은 확률적 프로세스 트리를 효율적으로 도출하기 위해 유전자 알고리즘에 타부 서치를 접목한 프로세스 모델 도출 방법론이다. 먼저, 프로세스 트리를 표현하기 위한 염색체로 2차원 배열을 정의하고, 초기 염색체 집합을 랜덤으로 생성한 뒤, 프로세스 트리의 적합도를 평가하였다. 그다음 유전 연산을 반복해 적합도를 증가시키는 방향으로 새로운 염색체 집합을 생성하였다. 유전 연산을 반복하는 과정에서 적합도가 낮은 프로세스 트리가 생성할 수 있는 액티비티를 수행 순서대로 나열한 문자열인 model trace를 타부 목록에 저장하고, 유전 연산을 통해 도출된 프로세스 트리가 생성할 수 있는 model trace와 타부 목록에 저장된 trace 간의 유사도를 비교해 이를 기준으로 프로세스 적합도가 낮은 해에 대한 중복 탐색을 방지하였다.
제안된 알고리즘의 성능을 검증하기 위해 기존 연구에서 사용된 2종의 이벤트 로그 데이터를 이용해 확률적 프로세스 트리를 도출한 뒤 모델의 적합도와 연산 시간을 비교 평가하는 실험을 수행하였다. 실험 결과 제안된 타부 서치-유전자 프로세스 마이닝 알고리즘이 적합도와 연산 시간의 측면에서 모두 기존의 유전자 프로세스 마이닝 알고리즘보다 우수한 성능을 나타냈다.제1장 서 론 1
제1절 연구 배경 및 필요성 1
제2절 연구 목적 및 접근법 6
제3절 논문 구성 7
제2장 관련 연구 분석 9
제1절 프로세스 모델 9
제2절 프로세스 모델 도출 방법론 13
제3장 타부 서치-유전자 프로세스 마이닝 알고리즘 제안 16
제1절 확률적 프로세스 트리의 정의 16
제2절 타부 서치-유전자 프로세스 마이닝 알고리즘 설계 23
제3절 타부 서치-유전자 프로세스 마이닝 알고리즘 순서도 38
제4장 실험 설계 및 성능 분석 40
제1절 실험 설계 40
제2절 실험 결과 분석 46
제5장 결 론 57
참고 문헌 59
영문 요약(Abstract) 62MasterProcess mining is an analytical technique aimed at obtaining useful information about a process by (i) extracting a process model visualizing real world process, (ii) comparing it with the actual process for evaluation, and (iii) performing activities to improve the quality of the model.
In process mining, it is most important to discover a process model that reflects real-world processes very well. However, most existing process models are deterministic because they do not include stochastic elements such as the occurrence probabilities or execution times of activities. Therefore, available information is limited, resulting in the limitations on analyzing and understanding the process. Furthermore, it is also important to develop an efficient methodology to discover the process model. Although genetic process mining algorithm is one of the methods that can handle data with noises, it has a limitation of large computation time when it is applied to data with large capacity.
In this paper, we define a stochastic process tree and propose a tabu search-genetic process mining algorithm to resolve these issues. The suggested stochastic process tree reflects stochastic characteristics by introducing two special types of nodes such as (i) the sequence probability information node representing the probability that a specific combination of activities is generated, (ii) the single activity execution time node representing the probability distribution of the activity execution time, and (iii) the time interval node between two adjacent activities expressing the probability distribution of the idle time between two possible consecutive activities.
The tabu search-genetic process mining algorithm efficiently discovers the suggested stochastic process tree by combining the genetic process mining algorithm with tabu list. In the algorithm, we first define a two-dimensional array as a chromosome to represent the process tree, an initial chromosome set is randomly generated, and the fitness values of the process trees are evaluated. Then, a new chromosome set is generated in the direction of increasing the fitness value by applying genetic operations such as crossover and mutation operations. In this procedure, a model trace, a string of activities that can be generated from the process tree, with a low fitness value is stored in the tabu list. By comparing the similarities among the model traces generated as the result of genetic operations with the traces stored in the tabu list, we can prevent duplicated searches for process trees with low fitness value being performed.
In order to verify the performance of the proposed algorithm, we performed a numerical experiment by using two kinds of event log data used in the previous research. The results show that the suggested tabu search-genetic process mining algorithm outperformed the simple genetic process mining algorithm in terms of fitness and computation time
Woo Joo Hong
No full text학위논문(박사)----아주대학교 일반대학원 :생명과학과,2007. 8Abstract 1
Table of Contents 4
List of Figures 7
I. Introduction 9
A. Skeletal muscle differentiation 9
B. PI3-kinase 18
C. Translational regulation of protein synthesis 24
D. The regulation of eEF2 phosphorylation 28
E. Extracellular matrix, integrin, and FAK 30
F. The purpose of this study 33
Ⅱ. Materials and Methods 34
A. Materials 34
B. Cell culture, cell counting and fusion index 35
C. Western blotting (Immunoblotting) 35
D. Immunoprecipitation 36
E. eEF2 Phosphorylation 37
F. Cell cycle analysis 37
G. PI3-kinase assay 38
H. Pulse metabolic labeling 38
I. Semi-quantitative RT-PCR (qRT-PCR) 39
J. Lipid preparation 39
K. Ca2+ imaging 40
Ⅲ. Results 41
A. LY294002 blocks the differentiation of L6 myoblasts 41
B. LY294002, an inhibitor of PI3-kinase, suppresses the proliferation of L6 myoblasts 45
1. LY294002 delayed the proliferation of L6 myoblasts 45
2. LY294002 inhibits the translation from G1 to S phase 49
3. LY294002 decreases the expression of cyclin D1 and cdk4 49
C. Phosphorylation of eukaryotic elongation factor 2 can be regulated by PI 3-kinase in the early stages of myoblast differentiation 53
1. PI3-kinase regulates eEF2 phosphorylation in the early stages of L6 myoblast differentiation 53
2. PI3-kinase is involved in regulation of the cell cycle by eEF2 phosphorylation 56
D. PI3-kinase can regulate expression of myogenin in post-transcriptional level 60
1. LY294002 blocks the differentiation of L6 myoblasts 60
2. LY294002 inhibits the expression of myogenin at the post-transcriptional level 63
3. LY294002 alters the phosphorylation of translational factors 65
4. LY294002 impairs the rate of global protein synthesis 67
5. Myogenin is rapidly depleted under the inhibition of mRNA translation by cycloheximide 69
E. Lipid products of PI3-kinase abrogate genistein-induced fusion inhibitioin in myoblasts 71
1. Genistein inhibits differentiation of L6 myoblasts 71
2. Genistein inhibits phosphorylation of FAK 72
3. Genistein indirectly inhibits PI3-kinase activity 77
4. Genistein as well as LY294002 impairs Ca2+ influx in L6 myoblasts 80
5. Lipid products of PI3-kinase eliminate genistein-induced inhibition of Ca2+ influx and cell fusion 82
6. Genistein does not inhibit Akt-depedent PI3-kinase pathway 88
Ⅳ. Discussion 90
A. PI3-kinase is required to transition from G1 to S during proiferation of L6 myoblasts 90
B. PI3-kinase is required to exit from cell cycle for the differentiation of L6 myoblasts 91
C. PI3-kinase activity is required to maintain a constant level of myogenin during myoblast differentiation 92
D. Products of PI3-kinase are involved in the intracellular Ca2+ influx pre-requisite for myoblast fusion 94
E. Roles of PI3-kinase during skeletal muscle differentiation 98
Ⅴ. References 100
Ⅵ. 국문 요약 118
Fig. 1. Schematic representation of skeletal muscle formation in the limb, with the different stages and genes potentially involved at each stage 10
Fig. 2. Model of satellite cell dynamics 11
Fig. 3. Scheme of skeletal muscle differentiation 16
Fig. 4. The phosphatidylinositol 3-kinase (PI3-kinase) family 19
Fig. 5. Structure and signaling of the class IA and IB PI3-kinases 20
Fig. 6. Upstream and downstream of Akt 23
Fig. 7. Regulation of translational initiation through mTOR signaling pathway 26
Fig. 8. Activation of eEF2 by insulin, GPCR agonists and other stimuli 29
Fig. 9. Structure of FAK and FAK mediated signaling 32
Fig. 10. LY294002 blocks the fusion of L6 myoblasts 43
Fig. 11. LY294002 inhibits the expression of muscle specific genes 44
Fig. 12. The effect of LY294002 on proliferation of L6 myoblasts 47
Fig. 13. LY294002 increases population doubling time 48
Fig. 14. LY294002 inhibits the transition from G1 to S phase 51
Fig. 15. LY294002 decreases the expression of cyclinD1, Cdk4 and pRb 52
Fig. 16. Phosphorylation of eEF2 during myoblast differentiation 55
Fig. 17. Phosphorylation of eEF2 and activity of PI3-kinase in synchronized cells 58
Fig. 18. LY294002 blocks the differentiation of L6 myoblasts 61
Fig. 19. Effect of LY294002 on the differentiation of L6 myoblasts when it was treated after expression of myogenin 62
Fig. 20. LY294002 decreases the protein level of myogenin but not the mRNA level 64
Fig. 21. LY294002 alters the phosphorylation of translation factors 66
Fig. 22. LY294002 impairs the rate of protein synthesis 68
Fig. 23. Cycloheximide decreases myogenin protein prior to GAPDH and b-actin 70
Fig. 24. Effect of genistein on differentiation of L6 myoblasts 73
Fig. 25. Phosphorylation of FAK 75
Fig. 26. Association of FAK with p85 78
Fig. 27. Effect of genistein on PI3-kinase activity 79
Fig. 28. Ca2+ imaging of L6 myoblasts 81
Fig. 29. Restoration of Ca2+ influx by PI-3,4,5-P3 83
Fig. 30. Restoration of cell fusion by addition of lipid products of PI 3-kinase 86
Fig. 31. Effect of genistein on the Akt phosphorylation 89
Fig. 32. Model of functions of PI3-kinase in the proliferation and the differentiation of myoblasts 99MasterMyoblast undergoes a series of events of proliferation, cell cycle exit, alignment and elongation, fusion and formation of straight muscle fiber during differentiation. Phosphatidylinositol 3-kinase (PI3-kinase) is activated by a variety of extracellular stimuli, which impacts a number of cellular process including cell growth, proliferation, differentiation, migration, and survival. This paper discusses the function of PI3-kinase in proliferation and differentiation of myoblast.
LY294002, an inhibitor of PI3-kinase, decreased the proliferation of L6 myoblasts and delayed the cell cycle progression from G1 to S phase. The expression of cyclin D1 and cdk4 were decreased by LY294002. These results suggest that PI3-kinase regulate the proliferation of L6 myoblasts by promoting the expression of cyclin D1 and cdk4.
When cultured in low growth factor-containing medium, L6 myoblasts exit cell cycle and undergo a well-defined program of differentiation. Phosphorylation of eukaryotic elongation factor 2 (eEF2) was related to the differentiation of chick embryonic muscle cells in culture. The extent of eEF2 phosphorylation declined shortly after differentiation induction of L6 myoblasts, when cells prepare for terminal differentiation by withdrawing from the cell cycle. This decrease in phosphorylation, however, was blocked by LY294002 and wortmannin, inhibitors of PI3-kinase. These inhibitors have previously been shown to strongly block the differentiation of myoblasts. Therefore, I hypothesized that PI3-kinase plays an important role in the withdrawal from the cell cycle by regulating eEF2 phosphorylation in the early stages of differentiation. To test this hypothesis, L6 myoblasts were synchronized at the G2/M phase of the cell cycle using nocodazole and then cultured in either fresh differentiation medium (DM) or growth medium (GM). Released cells accumulated in the G0/G1 phase in DM and progressed to the S phase in GM. Cyclin D1 was more rapidly degraded in cells cultured in DM than in GM. The extent of eEF2 phosphorylation was seen to decrease more prominently in DM than in GM. Inhibitors of PI3-kinase or Akt, or mTOR increased eEF2 phosphorylation, but PI3-kinase became more activated when eEF2 phosphorylation declined. These results suggest that the regulation of L6 myoblast differentiation by PI3-kinase is related to eEF2 phosphorylation.
PI3-kinase induces transcription of myogenin mRNA in myoblast differentiation. Here, I examined another mechanism regulating expression of myogenin by PI3-kinase. When LY294002 was added even after expression of myogenin, the fusion of myoblasts did not increase more than the level at the time treated with LY294002 and myogenin protein was decreased without change of its mRNA level. However, the mRNA and the protein of both GAPDH and b-actin were not changed by LY294002. LY294002 inhibited the rate of global protein synthesis through down-regulation of the activity of S6-kinase, S6, eIF4E, and eEF2. Cycloheximide, inhibitor of translation, rapidly decreased protein level of myogenin without the change of mRNA level but did not affected the level of protein and mRNA of both GAPDH and b-actin. These results shows that a half-life of myogenin protein is greatly shorter than those of GAPDH and b-actin proteins and the inhibition of protein synthesis at the translational level by LY294002 causes rapid depletion of myogenin protein which has a short half-life. This finding represents the first identification that PI3-kinase regulates expression of myogenin post-transcriptionally.
Genistein (4',5,7-trihydroxyisoflavone) is a tyrosine kinase inhibitor. Although the agent has shown to inhibit myoblast differentiation, neither intracellular target(s) as a tyrosine kinase inhibitor nor action mechanism of the agent is well known. Here I studied the effect of genistein on the differentiation of myoblasts. Genistein strongly but reversibly blocked both myoblast fusion and synthesis of the muscle-specific proteins. The agent also reversibly reduced the phosphorylation level of focal adhesion kinase (FAK), a cytoplasmic tyrosine kinase, and its interaction with p85, the regulatory subunit of PI3-kinase. In addition, genistein indirectly inhibited PI3-kinase activity and blocked calcium influx which is required for myoblast fusion. However, both genistein-induced inhibition of cell fusion and calcium influx were abrogated by the lipid products of PI3-kinase. These results demonstrate that genistein can exert their effect on the signaling pathway from FAK to calcium influx via PI3-kinase in the differentiation of myoblasts.
These data, together with the observation that PI3-kinase signaling is involved in the cell cycle progression of myoblast from G1 to S, induction of myogenin, sustaining the expression of myogenin, and cytosolic calcium elevation, suggest that PI3-kinase signaling plays important key roles in the proliferation and the differentiation of myoblasts
Stolen Land: A Marriage of Text and Music in My Country By Hyo-won Woo
Get PDFHyo-won Woo, a renowned female composer from South Korea, is highly regarded for her contributions to Korean choral music. Woo’s My Country is an 80-minute cantata consisting of 14 movements. As a full-time resident composer for the National Chorus of Korea, Woo composed this work in honor of the March 1st Revolution. She utilized her signature style of storytelling, incorporating narration and solo performers to portray the hardships of Korea during Japan’s colonial era. The cantata takes the mood of Requiem as a poignant symbol of Korea’s painful loss of their country and includes text derived from the resistance efforts of independence activists. As a result, it is an emotionally charged work for Koreans. The National Chorus of Korea has showcased multiple versions of My Country to accommodate changes in choir size and context. These versions include orchestration variations with longer and shorter arrangements, as well as different program options such as maximum, shorter, and highlight versions. Woo’s emphasis on communication with the audience through stage design is evident in this work.
This monograph presents an integrated exploration of the text and musical elements found in Woo’ cantata, My Country. Through a detailed examination of the work’s history and interviews with the composer, the author aims to offer a deeper understanding of the composer’s life and artistic style, as well as a thorough analysis of the cantata’s music, text, and context. This guide also provides valuable performance suggestions for conductors to enhance their interpretation of the work. Furthermore, the monograph emphasizes Woo’s significant projects, with a particular focus on the Ari Project. This initiative showcases Woo’s remarkable talent for rearranging her compositions to suit a diverse range of ensembles, thereby facilitating accessibility and expanding the reach of her works. Through a comprehensive examination of the historical context with meticulous detail, this dissertation enhances the knowledge and understanding of conductors and musicians who aspire to engage with Woo’s compositions. By delving into the profound musical styles of Korea, it equips them with a deep appreciation and insight necessary for meaningful interpretation and performance
케이블을 이용한 새로운 3 자유도 햅틱 인터페이스
No full text학위논문(석사) - 한국과학기술원 : 기계공학전공, 2009.2, [ 158 p. ]Cable-suspended haptic interface has advantageous features such as low inertia, high stiffness, high payload-to-weight ratio, low weight (low energy consumption), large workspace over conventional rigid-link haptic interface. Despite these many advantages, inherent major problems of cable-suspended mechanism such as redundancy of actuators, incompact & insufficient force feasible workspace, inability of manipulation with high speed and thereby low accuracy of position result in limiting the extent of application.
In this study, a novel and unique mechanism different from traditional approaches for a cable-suspended manipulator is proposed to solve those limiting factors. And a new 3 d.o.f cable-suspended haptic interface is designed and constructed considering simulation results for mechanical architecture, tension (velocity) of a spiral spring for high-speed manipulation, and workspace analysis.
Experiments are conducted into two aspects of performance, which are the position input and force output (impedance haptic display). Experimental results show the usefulness of the proposed new mechanism as a haptic interface.
Eventually, four inherent problems of conventional cable-suspended haptic interface are resolved reasonably from the developed three degrees of freedom cable-suspended haptic device and its control method.한국과학기술원 : 기계공학전공
White-light interferometry with high N.A. objective lenses for measurements of thickness profile and refractive index of thin-film structure
No full text학위논문(석사) - 한국과학기술원 : 기계공학전공, 2008.2, [ vii, 53 p. ]단일 광학계에서 셔터를 동작시킴에 따라 반사광측정계와 간섭계를 나누어 구현할 수 있는 박막의 두께형상 및 굴절률 측정용 백색광 간섭계를 제안한다. 두께 및 굴절률 측정을 위한 반사광측정법은 필터와 높은 개구수의 대물렌즈를 적용함으로써 다파장, 다중 입사각 그리고 수평 및 수직 편광방향에 대한 반사율 정보를 쉽게 획득할 수 있으며 이로부터 두께와 굴절률을 분리해낼 수 있다. 획득한 두께 및 굴절률 정보를 바탕으로 주파수 주사식 백색광 간섭계를 통해 형상정보를 구할 수 있다.한국과학기술원 : 기계공학전공
Absolute distance measurement by dispersive interferometry using a femtosecond pulse laser
Get PDFWe describe an interferometric method that enables to measure the optical path delay between two consecutive femtosecond laser pulses by way of dispersive interferometry. This method allows a femtosecond laser to be utilized as a source of performing absolute distance measurements to unprecedented precision over extensive ranges. Our test result demonstrates a non-ambiguity range of similar to 1.46 mm with a resolution of 7 nm over a maximum distance reaching similar to 0.89 m. (c) 2006 Optical Society of America.This work was supported by the Creative Research Initiatives Program of the Ministry of Science and Technology in the Republic of Korea
Intelligent Knowledge Recommendation Methods for R&D Knowledge Portals
No full textThe personalization in knowledge portals and knowledge management systems is mainly performed based on users’ explicitly specified categories and keywords. The explicit specification approach requires users’ participation to start personalization services, and has limitation to adapt changes of users’ preference. This paper suggests two implicit personalization approaches: automatic user category assignment method and automatic keyword profile generation method. The performances of the implicit personalization approaches are compared with traditional personalization approach using an Internet news site experiment. The result of the experiment shows that the suggested personalization approaches provide sufficient recommendation effectiveness with lessening users’ unwanted involvement in personalization process
Understanding collaborative filtering parameters for personalized recommendations in e-commerce
No full textCollaborative Filtering (CF) is a popular method for personalizing product recommendations for e-Commerce and customer relationship management (CRM). CF utilizes the explicit or implicit product evaluation ratings of customers to develop personalized recommendations. However, there has been no in-depth investigation of the parameters of CF in relation to the number of ratings on the part of an individual customer and the total number of ratings for an item. We empirically investigated the relationships between these two parameters and CF performance, using two publicly available data sets, EachMovie and MovieLens. We conducted three experiments. The first two investigated the relationship between a particular customers number of ratings and CF recommendation performance. The third experiment evaluated the relationship between the total number of ratings for a particular item and CF recommendation performance. We found that there are ratings thresholds below which recommendation performance increases monotonically, i.e., when the numbers of customer and item ratings are below threshold levels, CF recommendation performance is affected. In addition, once rating numbers surpass threshold levels, the value of each rating decreases. These results may facilitate operational decisions when applying CF in practice. ？ Springer Science+Business Media, LLC 2007
- …
