Development of chemo- and regioselective reactions for mixed acetals, transesterification and oxidative dehomologation using metal alkoxides

Yang, Hyesung Sungkyunkwan university 2021 국내박사

To develop efficient synthetic methods, it is necessary to examine the formulaic knowledge of organic chemistry. From this point of view, I proposed synthesis of mixed monosilyl acetals, efficient transesterification and deesterification, and oxidative dehomologation using diverse reactivity of metal alkoxides. In the first part of this thesis, a simple and selective synthesis of mixed monosilyl acetals using NaOtBu as nucleophile is introduced. Catalyst-dependent chemoselective Mukaiyama aldol reaction was established with the synthesized mixed monosilyl acetals. A plausible mechanism was probed through detection of oxocarbenium ion by HRMS. Highly stereoselective Mukaiyama aldol reaction with silyl enol ether was conducted. In the second part, I discussed transesterification and non-hydrolytic deesterification via 1,3-chelation between Na ion and ester group. The 1,3-chelation and formation of a tetrahedral intermediate were confirmed by NMR spectroscopy and ReactIR spectrometer. And nucleophilic reactivity of NaOtBu was investigated through deuterium labeling experiment. Furthermore, this protocol was successfully applied to norbornene derivative, chiral amino acid esters and oxazolidin-2-ones giving good results. In the last part, I demonstrated NaOtBu-O2 mediated oxidative dehomologation of homoallylic alcohols. The reaction afforded different carboxylic acids by selective cleavage of C(sp3)-C(sp3) bond and C(sp2)-C(sp2) bond depending on reaction solvent. Mechanistic study was performed through NMR spectroscopy and GC analysis to detect reaction intermediates. 효율적인 합성법을 개발하기 위해서는 기존에 알려진 화학적인 지식에 대해 다시 한 번 생각해 볼 필요가 있다. 본 박사학위 논문에서는 NaOtBu의 다양한 반응성을 이용한 비대칭 아세탈 (mixed acetal)의 합성법, 효율적인 에스터 교환 반응과 탈에스터화 반응 그리고 친환경적인 산화 절단법에 대해 서술하고자 한다. 논문의 첫 번째 파트는 NaOtBu를 친핵체로 이용하여 간단하고 선택적인 mixed monosilyl acetal의 합성법을 보고하였다. 합성한 mixed monosilyl acetal에서 촉매에 따라 화학 선택성을 갖는 Mukaiyama aldol 반응을 수행하였다. HRMS를 통해 oxocarbenium ion 중간체를 확인하여 반응 메커니즘을 보다 확실하게 규명하였다. 높은 입체선택성을 갖는 silyl enol ether의 Mukaiyama aldol 반응이 탐구되었다. 논문의 두 번째 파트에서는 Na ion과 에스터의 1,3-chelation을 통한 에스터 교환 반응과 물의 존재 없이 진행되는 탈에스터화 반응이 논의되었다. Na ion과 에스터의 상호 작용은 NMR spectroscopy를 통해 확인하였고, ReactIR을 통해 tetrahedral intermediate를 포착하였다. 또한 NaOtBu의 친핵성을 동위원소 실험을 통해 조사하였다. 개발된 반응은 노보넨 유도체, 키랄 아미노산 에스터 그리고 2-옥사졸리돈 유도체에 성공적으로 적용되어 의미 있는 결과를 보여주었다. 논문의 마지막 파트에서는 NaOtBu-O2 시스템에 의해 진행되는 homoallylic alcohol의 선택적인 산화적 절단을 논의하였다. 이 반응에서는 용매에 따라 한 분자 내에 있는 C(sp3)-C(sp3) 결합과 C(sp2)-C(sp2) 결합을 선택적으로 절단하여 서로 다른 구조의 카복실산을 합성한다. 반응 과정에서 생성되는 중간체들을 NMR spectroscopy와 GC 분석을 통해 확인하여 반응 메커니즘을 증명하였다.

Unit Cells for Filter Design Based on CPW Metamaterial Transmission Line

Yang, Lei 제주대학교 대학원 2013 국내석사

Laminations and connecting orbits on lattice

Yang, Zaiyong Northwestern University 2005 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

In this thesis we use the variational method to study minimal laminations and connecting orbits of lattice problem. Our setting is based on discrete variational problem as invested in X. Yang [9]. We prove the lamination structure of non-selfintersecting minimal sets by secondary invariants. With this structure we have another view of results by Yang. We also prove the existence of connecting orbits of two neighboring solutions in the lamination. Our method is based on the barrier functions as introduced in Z. Xia [12].

Customizing uEngine: An Open Source Business Process Management System

Le Yang 인제대학교 2008 국내석사

There is no doubt that Business Process Management (BPM) has added tremendous value to organizations in virtually every industry. By utilizing Business Process Management Systems (BPMS), organizations can streamline and automate tasks, reduce turnaround time, and help ensure regulatory compliance. The foremost considerations when enterprises deploy BPM solutions involve supportive system functions, friendly Graphical User Interfaces (GUI) and maximal Return on Investment (ROI), based on the specific processes required, organizational management, and working style of their particular business environment. In order to address the problems associated with BPMS customization, for the flexible coverage of various requirements, this thesis takes a three step approach. First, a generic roadmap of BPMS is presented by explaining some essential conceptions associated with BPM. Then, a comprehensive introduction of uEngine, an open source BPMS which supports flexible customization, is provided. Finally, application programming interface (API)-based and component-based customization in building a BPM application utilizing uEngine is introduced with illustrative examples.

Nonlinear Regression Based on Neural Networks: Prediction, Sufficient Dimension Reduction, and Model Compression

Yang, Wenjing ProQuest Dissertations & Theses University of Minn 2022 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

In the modern era, neural networks have achieved great success in various fields of applications. Since neural networks allow the modeling of nonlinear relationships, they have become popular and useful tools for tackling different problems such as prediction, classification, clustering, dimension reduction, visualization, computer vision, and many others. This thesis focuses on three interrelated topics in the study of neural networks under the regression setting with two primary goals. One is to incorporate traditional statistical learning methods and recent machine learning techniques, the other is to utilize statistical theory to provide theoretical justification and understanding of some neural network methods.With the rapid growth and development in modern information technology, high-dimensional data are ubiquitous in various scientific fields. Specifically, recent machine learning techniques on prediction performances under high-dimensional data frameworks have supported many applications for gene expression inferencing. We propose a deep neural network learning approach that incorporates deep learning techniques and statistical methods to effectively enhance the predictive accuracy to address a multi-task learning problem in gene expression profiling. Unlike the conventional complicated multi-task learning methods, our proposed method can effectively learn the interrelations from larger-scale tasks.As the data structure of high dimensions may raise challenges in terms of both computation and inferencing, it is often of interest to find a reduced dimensionality of the data from linear combinations of the predictors that can retain as much original information as possible in the relationship between the response and the original predictors. We propose a sufficient dimension reduction method that is able to not only obtain the structural dimension effectively without prior knowledge, but also estimate the central space. In addition, the framework is extended to the approximated case, which may be more applicable in practice.With the development of powerful extensions of classical works in deep learning, many state-of-the-art architectures today have achieved spectacular results and increasing attention in various applications. However, these works generally rely on complex architectures with tens and even hundreds of layers, which result in the need for tremendous storage space, time allocation, computation costs, and a copious amount of data. These concerns motivate research in model compression. Many empirical results have shown that when a certain model is more complex than what might be needed, different techniques can be applied to reduce model complexity and still retain the generalization power in the original model. The theoretical work, however, is limited. We examine neural network pruning from the perspective of sparse linear approximation and provide a theoretical understanding of how to characterize a network that is compressible and by how much it can be properly pruned with little prediction accuracy degradation.

Achieving High Performance Polymer Tandem Solar Cells

Yang, Jun University of California, Los Angeles 2011 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

Polymer-based organic photovoltaic systems hold the promise for a cost-effective, lightweight solar energy conversion platform, which could benefit from simple solution processing of the active layer. Nonetheless, most of the materials designed today always suffer from the inherent disadvantage of not having a broad absorption range, and relatively low mobility, which limit the utilization of the full solar spectrum. In this dissertation in order to achieve high performance, we focus on multilayered tandem structure, which serially connects two single bulk heterojunction (BHJ) cells. In addition to the requirement for photoactive materials, a serial-connected tandem cell requires an inter-connecting layer (ICL) that joins the sequential subcells. In Chapter 3 I successfully developed a novel ICL, which is resistance-free to minimize electric loss, highly transparent to minimize optical loss, and physically robust to protect the underlying layers against damage from further solution processing of the rear subcell. This layer could also tune the optical field through the device as an optical space, that can balance the photocurrent generated from the front and rear subcells, as well. The preliminary results showed that the PCE of the tandem cell reaches 7.0% under AM1.SG illumination (with up to 9% with anti-reflection coating under weak light illumination conditions). This ICL offers a general solution to the common challenging issues in the field of solution-processed tandem solar cells, organic light emitting devices and thin film transistors, enabling widely broaden the solvent selection for the secondary active layer process and ease the fabrication process. In order to develop new ICLs with other promising functions, I demonstrated plasmonic effects in an inverted tandem polymer solar cell configuration by blending Au nanoparticles (NPs) into ICL. Experimental results showed this plasmonic enhanced ICL improves both the top and bottom subcells' efficiency simultaneously by enhancing optical absorption. The presence of Au NPs did not cause electrical characteristics to degrade within the tandem cell. As a result, a 20% improvement of power conversion efficiency has been attained by the light concentration of Au NPs via plasmonic near-field enhancement. The simulated near-field distribution and experimental Raman scattering investigation support our results of plasmonic induced enhancement in solar cell performance. The results show a great potential of incorporating the plasmonic effect with conventional device structure in achieving highly efficient polymer solar cells. The final part of the thesis is try to solve a common problem from industry. because the material of PSC would degrade if post treated in the air with fluorescent light, and this would be more serious if DIO was used as a solvent additive. After a carefully exploration, the experiment results point out that the reason might come from the oxidation of the polymer chain by oxygen and high energy UV light. After the oxidation took place, the conjugating system of the polymer was destroyed and the material loss the semiconducting property. The degradation can be avoided if at least one of the above two factors (oxygen and UV light) could be eliminated. Eventually, the performance of the device fabricated in air under red light reaches 7.21%. which is the first report that fabricating device in air absolutely.

Manipulation of Magnetization Dynamics in Ferromagnetic and Antiferromagnetic Heterostructures

Cheng, Yang The Ohio State University ProQuest Dissertations & 2021 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

Spintronics is a vibrant research field that studies and utilizes the intrinsic spin properties of electrons in various materials. The dual states of spin (spin-up and spin-down) make it a natural candidate to represent the bit “0” or “1” in data storage and information processing. In the past two decades, a series of important discoveries in spintronics has been made in the labs and brought into the market. My research focus is to identify and investigate novel ferromagnetic and antiferromagnetic heterostructures that can be used to powerfully manipulate spins. This thesis is organized as follows: In Chapters 1 and 2, I briefly introduce the background knowledge in spintronics and the experimental as well as theoretical techniques that I use in my research. From Chapter 3 to 7, I present five research projects that I lead in my graduate study. In Chapter 3, we study a new nonlocal spin transport scheme. Compared with conventional nonlocal spin transport geometry where the spin current is generated through a ferromagnetic injector or spin Hall effect, in our scheme, spin current generation is through ferromagnetic resonance spin pumping in lateral Pt/Y3Fe5O12 heterostructures. This avoids the generation of high-energy thermal magnons through the spin Seebeck effect. We also observe a high spin current injection, which is more than 20 times larger than the local spin pumping in Pt/Y3Fe5O12, suggesting that the ultra-low damping bare Y3Fe5O12 is a much more efficient spin pumping source; In Chapter 4, we study the spin current transportation in Pt/NiO/Y3Fe5O12 trilayers through angular dependent ferromagnetic resonance spin pumping measurements. By systematically varying the temperatures and thicknesses of NiO, we observe spin current transport that is mediated by the antiferromagnetic NiO interlayer, which gives a non-sinusoidal angular dependence of inverse spin Hall voltage; In Chapter 5, we study current-induced Neel order switching in antiferromagnetic heterostructure Pt/α-Fe2O3. We demonstrate that the detected Hall resistance change is indeed due to the Neel order switching by carefully ruling out the thermal artifact from Pt. Our results point to a promising path toward controlling antiferromagnets using spin-orbit torque; In Chapter 6, we study the harmonic Hall response experimentally and theoretically in antiferromagnetic heterostructure Pt/α-Fe2O3. We build a theoretical model that proves the current-induced effects in antiferromagnetic heterostructures can be also detected by harmonic Hall technique like previously demonstrated in ferromagnetic heterostructures. Our experimental measurement is in excellent agreement with our modeling, which points out the important role played by the magnetoelastic effect and spin-orbit torque in current-induced switching of antiferromagnets; In Chapter 7, we make Hall measurements in antiferromagnetic heterostructure Pt/Cr2O3 bilayers and observe the high-temperature topological Hall effect. This indicates the nontrivial topological spin texture at Pt/Cr2O3 interface, which gives insights in searching skyrmions in antiferromagnets. The last chapter is a conclusion of the entire thesis.

한국 제조업에서 수출기업과 내수기업의 임금격차 분석

Yang Ling 忠南大學校 大學院 2018 국내석사

ABSTRACT A thesis submitted to the committee of Graduate School, Chungnam National University in a partial fulfillment of the requirements for the degree of Master of Arts in International Trade conferred in February 2018. An Analysis of the Wage Gap between Exporter and Non-exporter in Korean Manufacturing Industry Ling Yang Department of International Trade, Graduate School Chungnam National University (Supervised by Professor Keun Yeob Oh) In this study, the wage gap between exporter and non-exporter is analyzed using a panel data from Korean manufacturing industry. According to the theory, the wage gap can exist with the difference of productivities. It seems natural that firms can provide the higher wages when the labor productivities are higher. But this paper insists that there could be wage premium in the exporting firms even after we control the effects of the productivity differences. That may be because of some special situations for the developing countries. For example, the pressures from international organizations including ILO would make the exporting firms feel to pay higher wage. This paper analyzes the wage premium for the labors in the exporting firms panel regression methods. Wages are regressed on the exporter dummy variables, labor productivities, and other factors such as the firm size, and location of the firms. As the results, some interesting findings were obtained as follows. First, the wage gap between exporter and non-exporter was existing in the 2000s and the productivity difference was the main factor explaining the gap. Second, in addition to the productivity difference between exporter and non-exporter, some inherent factors of the exporter could also contribute to the wage gap. However, the wage gap due to these factors has been sharply decreasing recently.

Three Essays on the Relationship between ESG Strategy and Firm Performance: Applying Quantitative Text Analysis Based on Machine Learning Algorithm

YANG BYUNG MO 강원대학교 대학원 2024 국내박사

본 연구는 비정형 텍스트 자료를 활용한 환경(Environmental), 사회(Social), 지배구조(Governance)인 ESG 지수를 산출하는 새로운 방법론적 틀을 개발하는 연구가 시도되었다. 설문조사나 데이터 전문 기관의 2차 자료(Secondary Data)를 활용한 ESG 분석이 아니라 지속가능경영보고서에서 나타나는 기업의 관심 ESG 활동과 ESG 규준 키워드(ESG Standards Keyword)는 일치한다는 가정을 전제로 텍스트(Text) 속에서 유사도 검사를 통해 기업 고유의 ESG 점수를 도출하였다. 본 연구는 기업의 지속가능경영보고서내 전략적 콘텐츠를 직접 분석한 모델로써 기업의 ESG 운영 방식을 엄밀하게 해석한 텍스트 계량화 방법론을 새롭게 제시하고 있다. 따라서 본 연구에서는 머신러닝 알고리즘(Meachine Learning Algorithm) 기반 텍스트 계량화 분석에 대한 이해를 돕고자 측정 방법론 프레임워크를 제시하고 있으며, 이를 검증하기 위한 실험 결과는 다음과 같다. 첫 번째로 지속가능경영보고서와 ESG 모범 사전(Dictionary) 구축을 위한 ESG 모범규준 자료 수집과 정의하였다. 두 번째로 비정형 자연어의 전처리 과정 제시하였다. 세 번째로 워드임베딩(Word Embedding) 비지도 학습 방법인 알고리즘을 활용하여 문서들이 가진 주제어가 무엇인지를 분석하는 Word2Vec과TF-IDF의 성능 비교 분석하였다. 네 번째로 머신러닝 기반 워드임베딩 분석이 가지고 있는 하이퍼파라미터(Hyperparameters) 실험에 따른 최적화 실험 결과를 제시하고 있다. 이를 통해 최종적으로 ESG 활동별 사전과 기업별 지속가능경영보고서의 유사도 분석을 활용한 정량적 ESG 수치를 도출하고 있다. 비정형 텍스트 계량화 방법이 기존의 ESG 평가 기관에서 제시하는 ESG 지수와 비교하여 우수한 기술적 특징은 다음과 같다. 첫 번째로 연구자 단독으로 데이터 수집과 분석이 가능하고, 두 번째로 프로그래밍화된 알고리즘 계산 방식은 각종 휴먼 에러 발생을 억제하고 있다. 세 번째로, ESG 점수를 세분화하여 도출할 수 있다(기후변화, 오염, 물과 해양자원, 생물다양성과 생태계, 자원사용과 순환경제, 인력, 가치 사슬 내 근로자, 기업 활동에 영향을 받는 지역사회, 소비자 및 최종 사용자, 기업 문화 및 정책, 사업 수행의 전략, 부패 및 뇌물 사건 등의 정보를 포함한 내용으로 지배구조 리스크, 기업의 전반적 운영과 사업 수행과 관련된 정보). 네 번째로 방대한 비정형 데이터를 기업별로 3분 이내로 빠르게 ESG 수치를 산출하여 분석 속도 또한 매우 우수하다. 마지막으로 기존 결정론적 방법에서 수행하는 탐색적 방법을 넘어서는 텍스트 인과 관계를 분석하는 Word2Vec 모델을 적용함으로써 ESG 정량적 지수를 도출하는 기술적 확장을 제시하고 있다. 본 연구에서 개발한 ESG 지수를 통해 기업 성과와 ESG의 활동간의 실증분석을 수행한 결과는 다음과 같다. 특히, ESG 경영 활동과 기업성과 관계에 있어 균형적 ESG 경영전략이 상호작용 조절 관계를 통해 긍정적 영향을 미치는지 검증해 보았다. 이에 대한 대답을 찾기 위해 본 연구는 국내 KOPSI 200대 기업과 포춘에서 선정한 글로벌 500대 기업 가운데 시가총액 상위 기업들의 일련의 재무 속성 그리고 균형적 ESG 경영전략의 상호작용을 나타내는 변수로 구성된 패널 모형을 분석하였다. 한국과 글로벌 사례를 실증 분석한 결과는 다음과 같다. 기업성과와 ESG 활동 간에 직접 효과를 분석한 결과에서는 환경(E), 사회(S), 그리고 지배구조(G)의 경우 (-)부정적 관계가 관찰되었다. ESG의 추구하는 기업의 목표는 투자자의 신뢰도 확보와 투자를 확대하기 위하여 비재무적 활동을 하게 되는데 이는 오히려 비용을 수반하게되고, 재무 건정성을 악화시킬 수 있다. 이는 대리인 이론(Agency Theory) 관점에서와 같이 ESG에는 막대한 비용과 시간이 투자되기 때문에 주주를 위한 활동보다는 기업의 평판을 높이기 위한 도구로 기업의 수익성을 악화시킬 수 있다는 이론적 주장과 맥락을 같이 한다. 최근에는 기업에서 ESG 관련 과도한 투자를 지양하는 것이 재무성과에 긍정적 영향을 미친다는 이른바 안티(Anti) ESG 움직임이 나타나고 있는 상황에서 기업의 본질적 ESG 활동에 대한 의미를 다시 한번 고려해봐야 한다. ESG 활동별로 살펴보면, 환경 분야의 활동은 기업의 성과를 긍정적으로 달성하기 위해 재무적 위험을 최소화하는 방식을 고려해야 한다. ESG의 영역별 핵심 평가지표 중 환경등급을 평가하는 내용으로 탄소 배출 제로, 에너지 사용량 효율화, 폐기물 감축 등 환경 유해물질 배출 항목들이 포함되어 있다. 환경성과 개선을 위한 기업의 노력은 설비를 구조적으로 바꾸거나 친환경 효율을 증대시키기 위한 R&D 개발로, 기업이 투자한 노력 대비 재무성과에 긍정적으로 나타나기 쉽지 않다는 것이다. 즉, 기업은 현실적으로 경제성장과 환경 성장을 동시에 달성하는데 어려움이 나타나는 것으로 해석되기 때문이다. 사회적 활동은 주주 이론과 이해관계자 이론에 대한 규범적 활동으로 주주가치를 극대화한다는 신념에 기반하여 브랜드 이미지를 향상시키는 활동이다. 고객 및 지역사회와의 관계가 강화하거나 좋은 인재를 확보여 사회공헌 활동을 높이는 것이다. 지역사회 부분에서는 여러 사회구성 집단을 모두 개인 맞춤형으로 사회적 가치와 비즈니스 가치가 잘 연계되어 긍정적으로 제공하기에는 어려움이 나타나는 것으로 보인다. 우선 기업이란 이익을 최우선시하는 경향으로 인해 기업들은 주주가치 향상을 위해 고객중심의 사회활동과 이를 통한 기업 이윤 창출을 위한 노력에는 아직 긍정적인 결과가 나타나고 있지 않다. 지배구조 활동은 기업성과에 부정적 영향을 미친다는 실증분석 결과가 나타남에 따라서 최고경영자의 역할 및 거버넌스의 실행에서도 숙제가 남겨져 있다. 기업이 전략 기획이나 내부 및 외부의 환경 분석을 통해 기업에 적합한 전략을 수립했다고 할지라도, 구성원들이 전략을 제대로 실행하지 못한다면 결국 기업은 지속가능 경영에 목표한 재무성과를 달성할 수 없다. 따라서 기업이 성공적으로 ESG 경영을 수행하기 위해서는 장기적인 성과 목표에 대한 CEO의 관심과 실행 의지가 중요하고, ESG의 거버넌스에 관련된 전략 활동이 중요성을 다시 한번 제시해주고 있다. 본 연구의 실증분석 결과 글로벌 그리고 국내 기업의 전략적 관점에서 균형적 ESG 활동이 기업성과에 모두 유의한 (+)긍정적 조절 효과가 있는 것으로 확인되었다. 다시 말하면 균형적 ESG 경영전략을 잘 수행하는 기업이 균형적 ESG 경영 전략이 낮은 집단보다 기업성과에 강한 긍정 효과 발휘하는 것으로 나타났다. 직접 관계에서는 ESG 활동 수준들이 정량적 목표 달성에 의미를 중요시하는 경향으로 기업성과에 (-)부정적 영향 관계를 갖지만, 균형적 ESG 활동을 추진함으로써 활동 간 불평등 및 차별 격차를 완화함으로써 안정적인 운영을 유지하게 된다. 따라서 ESG 운영과 기업성과에 긍정적 영향을 미치기 위해서 균형적 전략 모색의 필요성은 다음과 같다. 첫째, 균형적 성장 전략은 기업의 내외부 자원의 제약적인 요소들을 통합적으로 파악함으로써 기업이 직면하고 있는 위험을 포함한 현실적인 성장 목표를 수립하는데 도움이 될 수 있다. 기업이 가진 자원의 제약을 파악하지 못하는 경우 외부자금조달 비용을 증가시켜 재무성과는 악화되지만 균형적 ESG를 통해 지속가능한 성장을 위한 전략계획을 수립하는데 중추적인 도구가 될 수 있다는 것이다. 기업이 지속 가능한 성장에 필요한 자원의 부족 현상에 대응하기 위해서 자본구조 측면에서 자본조달 정책을 어떻게 할 것인가에 대한 전략적 선택을 하게 되는데, 이때 균형적 전략 활동은 기업의 재무적 선택을 최적화 할 수 있게 도와주는 가이드라인 역할을 한다. 기존의 자원 낭비를 줄이고 조직의 운영을 과도하게 확장하지 않는 범위에서 생산성을 향상시키는 효율성의 구축을 통해 자원의 한계를 극복하고 지속가능한 성장을 도모할 수 있는 성장을 계획하는데 도움이 된다. 둘째, 균형적 관점(Balance Perspective)은 의사결정으로 인한 자원의 비효율성을 줄일 뿐만 아니라 불확실한 환경에도 조직의 운영적 문제를 최소화하여 경제적 성과에 긍정적인 영향을 미치기 때문이다. 최근 ESG 운영에 대한 비전이 기존에는 기업의 외부에서 평가에서 공공성 목표 달성이라는 핵심 가치를 제시하였다면, 최근 ESG 경영으로 인한 천문학적인 적자 규모로 이제는 기업이 구조적 최적 효율을 달성하는 방향으로 개념적 재정립을 수행해야 한다는 전략적 시사점이 도출된다. 여기서 재무적 관점이 결여되어 있는 ESG 활동은 기존의 자원과 새로운 지식 활동에 생산 비용을 수반하게 되어 단기적 관점의 경제적 성과에는 악영향이 나타날 수 있으며, 노동생산성을 악화시켜 경제적 성과를 저해할 수 있다. 하지만 균형적 전략 활동을 통해 소모적 비용을 줄이고 자원의 효율성을 높여 비용을 절감함으로써 기업의 수익성에 긍정적인 영향을 미칠 수 있다. 이와 같이 균형적 ESG 전략 활동은 전략적 도구로서 기업의 지속적인 성장을 위해 재무 및 운영 전략을 효과적으로 유지하는데 활용되고, 기업이 경쟁 우위를 확보하는데 유용하게 사용될 수 있다. 셋째, 기업성과를 향상시키는 절대 우위 전략은 없지만 균형적 ESG 전략이 가지는 성격은 최적의 전략적 메커니즘 경영을 모색하여 효과적 이익 추구를 가능하게 한다. 예를 들어 균형적 운영 전략을 추진한다는 것은 서로 다른 활동에 대해 기준을 맞춤으로써 운영 프로세스(Process)를 성과의 향상과 효율성 중심으로 반영될 수 있다. 과도하게 전략이 편향되거나 몰입되는 것은 투자한 투자로 비용이 발생할 수 있다는 점에서 균형 전략은 경제적이고 합리적인 운영 방식을 제고함으로써 경제적 비용 부담을 완화 시킬 수 있는 강점이 존재한다. 따라서 균형적 전략을 추구하는 운영은 활동 간 분석적 접근을 통해 부정적 축적에 따른 악순환 방지를 탈피하거나 문제라는 점을 인식함으로써 기업의 성과에 긍정적으로 영향을 미칠 수 있는 주요 전략 활동으로 사료된다. 한국의 실증 분석 사례에서는 ESG 경영 활동이 기업의 재무성과에 유의한 결과를 보이는 것과 달리 기업가치인 Tobin’s Q(TBQ)에는 전혀 유의한 설명력을 가지는 유형을 찾을 수 없었다. 현재 비재무적 활동의 주요 주체인 투자자와 이해관계자들은 한국의 지속가능경영보고서에서 나타나는 ESG 활동에 대해서는 전혀 반응하지 않고 있다는 것을 의미한다. 이는 비재무적 활동 효과가 시장가치에 직접적으로 반영되기 위한 논의가 우선시되어야 함을 시사하고 있으며, 현재의 논의가 단순히 정보공시 확대를 목표하기보다는 투자자와 이해관계자의 관심을 높이는 비재무적 정보가 무엇인지 판별하는 것부터 노력이 집중되어야 한다. 현재 그리고 앞으로도 금융 투자기관들은 ESG 경영에 따라 투자 범위를 확대하겠다고 밝히고 있는 사실에 주목하여 ESG 전략을 모색해야 할 것이다. 본 연구 결과를 종합하면 기업이 ESG 활동 요인들에 의한 맹목적인 기회나 이상적인 가치를 추구하기보다는, 최적의 이익을 추진하기 위한 기업의 현실적인 자원 보유 현황과 직면한 상황에 부합하는 ESG 활동 간에 균형을 추구하는 전략이 기업성과에 긍정적 영향을 미친다. 이는 ESG의 균형 활동이 자원의 활용 이점을 강화하고 부정적인 요소를 감소시킬 수 있는 상호보완적인 역할로써 지속가능한 성장을 통해 재무성과를 향상시킬 수 있다고 볼 수 있다. 균형적 ESG 활동을 통한 기업의 비재무적인 요소들의 성과 향상은 사업화 프로세스에서 운영의 비대칭성이 초래할 수 있는 부정적인 영향을 감소시키고 기업 경쟁력을 강화해 줄 수 있다는 것이다. 현재 기업들이 정부의 ESG 정책과 규제로 인하여 사업 포트폴리오 및 조직을 재편하는 등 맹목적인 ESG 성과 도출을 위해 전사적 역량을 집중하고 있는 상황에서, 앞으로 무조건적인 ESG 경영 추진하는 것보다는 처해진 상황과 환경 맞춰서 ESG 경영 활동을 고려해야만 기업 고유의 가치를 증대시킬 수 있을 것이다. Rather than relying on secondary data derived from surveys or data agencies, the primary objective of this study was to formulate an innovative methodological framework for the computation of an Environmental, Social, and Governance (ESG) index. This endeavor entailed the creation of a quantitative ESG score by means of unstructured text data analysis, employing similarity analysis techniques. The underpinning assumption in this similarity analysis was the existence of a correlation between the strategic facets of a company and the standard ESG keywords found within the ESG report. As a result, the quantified ESG scores obtained in this manner can be regarded as primary data directly extracted from the strategic content within the company's sustainability report, thereby offering a robust methodology for the comprehensive interpretation of the company's operational approach. The principal findings stemming from the text quantification process are succinctly summarized subsequent to the application of machine learning algorithms. We have delineated a comprehensive methodology for assessing a company's ESG index, with the objective of elucidating the process of ESG score derivation. This procedure encompasses several distinct stages. In the initial phase, we compile ESG standards for the establishment of sustainable management reports and the development of an ESG dictionary. Subsequently, in the second phase, we preprocess unstructured natural language data. Advancing to the third stage, we employ an unsupervised learning method based on word embedding algorithms to identify keywords within the documents. In the fourth stage, we present the outcomes of optimization experiments, specifically comparing the performance of Word2Vec and TF-IDF, both of which are machine learning-based word embedding analyses, as determined through hyperparameter experiments. Ultimately, we derive quantitative ESG scores by conducting similarity analysis utilizing the ESG dictionary for each sustainability report. In contrast to existing ESG evaluation agencies' indices, the unstructured text quantification method presented in this study exhibits several distinctive technical attributes. Firstly, it empowers researchers to autonomously amass and scrutinize data, affording a heightened level of control and customization throughout the data collection and analysis process. Secondly, the method adopts a systematic algorithmic calculation approach, mitigating the potential for human errors that may be inherent in manual processes. Moreover, the analytical flexibility inherent to this method allows for the disaggregation of ESG scores into more granular components, facilitating a more nuanced comprehension of a company's ESG performance. Additionally, the method excels in computational efficiency, with the capacity to analyze a substantial volume of data for each company in under three minutes, enabling the rapid generation of ESG metrics. Finally, this approach extends its technical prowess by delving into the analysis of causal relationships within the textual content, surpassing the exploratory techniques commonly utilized in prevailing deterministic methodologies. This extension affords the derivation of ESG quantitative indices grounded in a deeper understanding of the causal connections embedded within the text. In summary, the unstructured text quantification method empowers researchers to autonomously collect and analyze data, employs a systematic algorithmic calculation approach to minimize human errors, offers analytical flexibility, ensures expeditious calculations, and incorporates the analysis of textual causal relationships to derive ESG quantitative indices. The empirical analysis entailed an exploration of the nexus between firm performance and Environmental, Social, and Governance (ESG) factors. This investigation, in particular, sought to assess whether the implementation of a well-balanced ESG management strategy exerts a positive influence on the connection between ESG and firm performance. To conduct this examination, a panel model was employed, which incorporated variables representing the interplay among ESG factors, financial attributes, and balanced ESG scores. This analysis was centered on a sample comprising the top 200 companies listed on Korea's KOPSI, as well as Fortune 500 companies, with market capitalization serving as the key determinant for their inclusion in the study. The findings of this analysis revealed a noteworthy correlation between financial performance and Environmental (E) and Governance (G) aspects of ESG activities, indicating a negative relationship. However, no significant association was discerned concerning the Social (S) dimension. The expansion of social initiatives may prompt the integration of non-financial undertakings aimed at bolstering investor confidence and attracting additional investments. Nevertheless, such a course of action may inadvertently give rise to costs and potentially undermine financial stability. These observations align with the tenets of agency theory, which posits that significant resource allocation to ESG activities may undermine a company's profitability, primarily functioning as a reputation-enhancing tool rather than directly serving shareholder interests. Given the emergence of the anti-ESG movement, where companies are inclined to curtail excessive ESG investments to achieve a positive financial impact, it becomes imperative to formulate a strategy that enhances the core ESG activities that underpin corporate performance. To attain favorable corporate performance through environmental initiatives, companies need to strategize to mitigate financial risks. Within the various ESG evaluation criteria, the environmental rating is a crucial element, encompassing factors like carbon emissions reduction, energy efficiency enhancement, waste minimization, and the reduction of environmentally hazardous emissions. It is notable that the endeavors made by a company to enhance its environmental performance, including structural modifications to facilities or investments in research and development for eco-friendly efficiency, often do not yield a direct and positive impact on financial performance commensurate with the efforts expended. In essence, practical challenges are evident in simultaneously achieving economic growth and environmental progress. Social activities, which aim to enhance brand image under the premise that optimizing shareholder value is a normative practice in both shareholder theory and stakeholder theory, are undertaken with the intention of strengthening relationships with customers, the local community, or acquiring top talent. In the realm of community engagement, it can be challenging to concurrently deliver customized social value and business value to all social groups. Primarily, this challenge arises due to the prevailing inclination of companies to prioritize profit above all else. Consequently, companies have not yet witnessed conclusive positive outcomes in their endeavors to develop customer-centric social initiatives that align with corporate profits and contribute to enhancing shareholder value. Empirical analysis underscores that governance activities exhibit a negative influence on corporate performance, emphasizing the vital roles played by the CEO in the governance process. Practical responsibilities largely rest on the shoulders of the CEO and the effective execution of governance initiatives. Even if a company formulates a strategy well-suited to its needs through strategic planning and a comprehensive analysis of its internal and external environment, the successful attainment of sustainable management objectives hinges on the adept execution of the strategy by its members. For a company to effectively implement ESG management, the CEO's dedication to long-term performance goals and the deployment of appropriate human resource management strategies are crucial. In contrast to prior studies, which predominantly presented static perspectives on the relationship between ESG and corporate performance, this research underscores the importance of a dynamic strategy wherein corporate performance evolves based on the strategic actions undertaken by the company. This provides valuable insights into the strategic process that interconnects the domains of the environment, society, and governance and highlights the need for substantial empirical contributions. In the course of this study, an empirical analysis focusing on the strategic perspective of global and domestic companies was conducted, revealing a significant positive moderating effect on corporate performance stemming from balanced ESG activities. The findings suggest that companies effectively implementing a well-balanced ESG management strategy exhibit a notably stronger positive influence on corporate performance compared to groups with a less balanced ESG management approach. In a direct relationship, the levels of ESG activities were found to have a negative impact on corporate performance, potentially due to the prevalent emphasis on achieving quantitative targets. However, the promotion of balanced ESG activities helps alleviate disparities and discrimination gaps among these activities, thereby contributing to the sustained operational stability of companies. Hence, the imperative of formulating a balanced strategy to foster a positive impact on both ESG operations and corporate performance is underscored. To set a realistic growth objective through a balanced growth strategy, it is imperative to gain a comprehensive understanding of both the internal and external resources and the risks confronting the company. Failing to grasp the limitations of available resources can lead to heightened external financing costs and hinder overall economic performance. Nevertheless, embracing a balanced approach to Environmental, Social, and Governance (ESG) factors can prove instrumental in formulating a sustainable growth plan. To address the challenge of resource scarcity, companies strategically determine their capital financing policies, with a particular focus on capital structure. In such circumstances, active involvement in balanced strategy initiatives aids companies in optimizing their financial decisions, serving as a valuable roadmap for charting a growth trajectory that surmounts resource constraints. This approach facilitates sustainable growth by minimizing resource wastage and enhancing operational efficiency without unwarranted expansion. A balanced strategic perspective not only mitigates resource inefficiency resulting from decision-making but also exerts a positive influence on economic performance by minimizing organizational operational challenges, even in uncertain environments. While the recent ESG management vision has emphasized the core value of achieving societal objectives through external evaluation of the company, the substantial deficit associated with recent ESG management has prompted a conceptual reevaluation towards optimizing structural efficiency to benefit the company. Strategic insights are gleaned on the necessary actions to be taken. In this context, ESG activities devoid of a financial perspective may detrimentally impact short-term economic performance, given their involvement in production costs for existing resources and new knowledge initiatives, potentially hampering labor productivity and overall economic performance. In contrast, balanced strategic activities have the potential to enhance a company's profitability by reducing superfluous costs and enhancing resource efficiency. Consequently, balanced ESG strategy activities serve as a strategic instrument for effectively aligning financial and operational strategies to promote sustained company growth, thereby contributing to the acquisition of a competitive advantage. While there exists no significant association of a strategy that guarantees unequivocal superiority in enhancing corporate performance, but the adoption of a balanced Environmental, Social, and Governance (ESG) strategy serves as a facilitator in the quest for profitable outcomes via the effective management of strategic mechanisms. An illustration of balanced operational strategy promotion can be seen through the emphasis placed on performance enhancement and efficiency by adjusting standards for various activities. In contrast to excessively biased or overly committed strategies that may entail significant investments and result in economic burdens, balanced strategies confer the advantage of reducing financial strain by implementing rational and cost-effective operational methods. Consequently, the pursuit of a balanced strategy in operations is acknowledged as a pivotal strategic endeavor that exerts a positive impact on a company's performance. It aids in circumventing the pernicious cycle of negative accumulations and proactively identifies and resolves concerns through analytical approaches, thereby contributing to enhanced corporate performance. In the case of Korea, it was observed that ESG management activities had no discernible impact on the financial performance of companies, as indicated by the TBQ results. This implies that investors and stakeholders, who presently play a central role in non-financial activities, are not evidently responding to the ESG activities reported in Korea's sustainability reports. This underscores the necessity for a shift in focus towards addressing how the effects of non-financial activities can be directly reflected in market value. The ongoing discourse should prioritize discussions on the type of non-financial information that will pique the interest of investors and stakeholders, moving beyond mere information disclosure expansion. Efforts should commence with a concentrated determination to effect change. Particularly, in a landscape where ESG holds a pivotal place in management trends, prominent global financial institutions should take note of their commitments to broaden their investment scope in line with ESG management. This necessitates the pursuit of meaningful strategies for corporate value that can steer investors and stakeholders in a positive direction. In the Korean context, it is evident that ESG management activities have not yielded significant improvements in the financial performance, as reflected by the TBQ results. This implies that investors and stakeholders, who currently play pivotal roles in non-financial activities, are not noticeably responsive to the ESG initiatives outlined in Korea's sustainability reports. This underscores the need to shift the discussion towards how the impact of non-financial activities can be directly translated into market value. The current discourse should prioritize considerations regarding what types of non-financial information can engage the interest of investors and stakeholders, moving beyond mere expansion of information disclosure. Efforts should be channeled into implementing change from the outset. Particularly, in a landscape where ESG holds significant relevance as a management trend, major global financial institutions should take note of their commitments to broaden their investment scope in alignment with ESG management. This necessitates the pursuit of meaningful strategies to enhance corporate value, ultimately guiding investors and stakeholders in a positive direction. In conclusion, a balanced approach that aligns a company's actual resource capabilities and current circumstances with the pursuit of optimal profitability has a positive impact on corporate performance, as opposed to blindly pursuing opportunities or ideal values solely based on ESG activity factors. The significance of ESG balancing activities becomes evident in their role as complements, reinforcing the efficient utilization of resources and mitigating negative factors. This, in turn, promotes sustainable growth and bolsters financial performance. Hence, it is crucial to assess ESG management activities within the specific context and environment of the company to enhance its overall value. Unconditionally pursuing ESG management solely to meet ESG performance metrics, as might be imposed by government ESG policies necessitating extensive business and organizational restructuring, is misguided.

Making Data Storage Efficient in the Era of Cloud Computing

Tang, Yang Columbia University ProQuest Dissertations & These 2020 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

We enter the era of cloud computing in the last decade, as many paradigm shifts are happening on how people write and deploy applications. Despite the advancement of cloud computing, data storage abstractions have not evolved much, causing inefficiencies in performance, cost, and security.This dissertation proposes a novel approach to make data storage efficient in the era of cloud computing by building new storage abstractions and systems that bridge the gap between cloud computing and data storage and simplify development. We build four systems to address four data inefficiencies in cloud computing.The first system, Grandet, solves the data storage inefficiency caused by the paradigm shift from upfront provisioning to a variety of pay-as-you-go cloud services. Grandet is an extensible storage system that significantly reduces storage costs for web applications deployed in the cloud. Under the hood, it supports multiple heterogeneous stores and unifies them by placing each data object at the store deemed most economical. Our results show that Grandet reduces their costs by an average of 42.4%, and it is fast, scalable, and easy to use. The second system, Unic, solves the data inefficiency caused by the paradigm shift from single-tenancy to multi-tenancy. Unic securely deduplicates general computations. It exports a cache service that allows cloud applications running on behalf of mutually distrusting users to memoize and reuse computation results, thereby improving performance. Unic achieves both integrity and secrecy through a novel use of code attestation, and it provides a simple yet expressive API that enables applications to deduplicate their own rich computations. Our results show that Unic is easy to use, speeds up applications by an average of 7.58x, and with little storage overhead.The third system, Lambdata, solves the data inefficiency caused by the paradigm shift to serverless computing, where developers only write core business logic, and cloud service providers maintain all the infrastructure. Lambdata is a novel serverless computing system that enables developers to declare a cloud function's data intents, including both data read and data written. Once data intents are made explicit, Lambdata performs a variety of optimizations to improve speed, including caching data locally and scheduling functions based on code and data locality. Our results show that Lambdata achieves an average speedup of 1.51x on the turnaround time of practical workloads and reduces monetary cost by 16.5%.The fourth system, CleanOS, solves the data inefficiency caused by the paradigm shift from desktop computers to smartphones always connected to the cloud. CleanOS is a new Android-based operating system that manages sensitive data rigorously and maintains a clean environment at all times. It identifies and tracks sensitive data, encrypts it with a key, and evicts that key to the cloud when the data is not in active use on the device. Our results show that CleanOS limits sensitive-data exposure drastically while incurring acceptable overheads on mobile networks.