• World Technopolis Review
• /
• v.1 no.2
• /
• pp.92-106
• /
• 2012

This study focuses on exploring the behaviours of high-tech start-up firms in response to the policy interventions undertaken to promote regional innovation in South Korea since 1997. High-tech start-ups and their technological entrepreneurship are increasingly considered by policy makers and academics to play a crucial role in the generation of innovation and economic development. However, this study started from a basic concern of why government intervention does not necessarily result in an increase of regional innovation capacity. To explain this concern, we constructed a new conceptual framework of 'government dependency' and apply this to 'Daedeok,' a regional innovation system in South Korea, to explore the reproduction of path dependency as an impact induced by innovation policy. This conceptual framework was developed by remodeling path dependency approaches through a systemic and interactive lens. An empirical study used qualitative interviews of start-up founders to delineate the emergence of a new development path and the extent to which dependency was reproduced in the Daedeok regional innovation system. Empirical analysis suggested that 'reliance' and 'persistence' were the crucial factors in the production and reproduction of the government dependency. Some firms accepted dependency as reliance, but others regarded it as policy utilization. Thus, a critical juncture could not be clearly identified in actors' behaviour. It was also unclear if dependency had hindered innovation, but it was shown that the regional and institutional contexts strongly influenced the reproduction process. The study concludes that the construct of government dependency can also provide useful insights into policy learning as well as the success of government interventions.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.14-19
• /
• 2016

본 논문은 문장에서의 어절 간 의존관계가 성립될 때 의존소와 지배소가 어떠한 관계를 가지는지 의존 관계명을 부착하는 모델을 제안한다. 국내에서 한국어 의존구문분석에 관한 연구가 활발히 진행되고 있지만 의존 관계만을 결과로 제시하고 의존 관계명을 제공하지 않는 경우가 많았다. 따라서 본 논문에서는 의존경로(Dependency Path)와 음절의 의존 관계명 분포를 반영하는 음절 임베딩를 이용한 의존 관계명 부착모델을 제안한다. 문장에서 나올 수 있는 최적의 입력 열인 의존 경로(Dependency Path)를 순차 레이블링에서 좋은 성능을 나타내고 있는 bidirectional LSTM-CRFs의 입력 값으로 사용하여 의존 관계명을 결정한다. 제안된 기법은 자질에 대한 많은 노력 없이 의존 경로에 따라 어절 및 음절 단어표상(word embedding)만을 사용하여 순차적으로 의존 관계명을 부착한다. 의존 경로를 사용하지 않고 전체 문장의 어절 순서를 바탕으로 자질을 추출하여 CRFs로 분석한 기존 모델보다 의존 경로를 사용했을 때 4.1%p의 성능향상을 얻었으며, 의존 관계명 분포를 반영하는 음절 임베딩을 사용한 bidirectional LSTM-CRFs는 의존 관계명 부착에 최고의 성능인 96.01%(5.21%p 개선)를 내었다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.14-19
• /
• 2016

본 논문은 문장에서의 어절 간 의존관계가 성립될 때 의존소와 지배소가 어떠한 관계를 가지는지 의존 관계명을 부착하는 모델을 제안한다. 국내에서 한국어 의존구문분석에 관한 연구가 활발히 진행되고 있지만 의존 관계만을 결과로 제시하고 의존 관계명을 제공하지 않는 경우가 많았다. 따라서 본 논문에서는 의존 경로(Dependency Path)와 음절의 의존 관계명 분포를 반영하는 음절 임베딩를 이용한 의존 관계명 부착 모델을 제안한다. 문장에서 나올 수 있는 최적의 입력 열인 의존 경로(Dependency Path)를 순차 레이블링에서 좋은 성능을 나타내고 있는 bidirectional LSTM-CRFs의 입력 값으로 사용하여 의존 관계명을 결정한다. 제안된 기법은 자질에 대한 많은 노력 없이 의존 경로에 따라 어절 및 음절 단어표상(word embedding)만을 사용하여 순차적으로 의존 관계명을 부착한다. 의존 경로를 사용하지 않고 전체 문장의 어절 순서를 바탕으로 자질을 추출하여 CRFs로 분석한 기존 모델보다 의존 경로를 사용했을 때 4.1%p의 성능향상을 얻었으며, 의존 관계명 분포를 반영하는 음절 임베딩을 사용한 bidirectional LSTM-CRFs는 의존 관계명 부착에 최고의 성능인 96.01%(5.21%p 개선)를 내었다.

• Journal of KIISE:Software and Applications
• /
• v.34 no.2
• /
• pp.140-149
• /
• 2007

This paper presents a statistical model for Korean dependency-based parsing. Although Korean is one of free word order languages, it has the feature of which some word order is preferred to local contexts. Earlier works proposed parsing models using modification lengths due to this property. Our model uses headible path contexts for modification length probabilities. Using a headible path of a dependent it is effective for long distance relation because the large surface context for a dependent are abbreviated as its headible path. By combined with lexical bigram dependency, our probabilistic model achieves 86.9% accuracy in eojoel analysis for KAIST corpus, more improvement especially for long distance dependencies.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.161-177
• /
• 2019

In this paper, we study the performance improvement of the answer extraction in Question-Answering system by using sentence dependency parsing result. The Question-Answering (QA) system consists of query analysis, which is a method of analyzing the user's query, and answer extraction, which is a method to extract appropriate answers in the document. And various studies have been conducted on two methods. In order to improve the performance of answer extraction, it is necessary to accurately reflect the grammatical information of sentences. In Korean, because word order structure is free and omission of sentence components is frequent, dependency parsing is a good way to analyze Korean syntax. Therefore, in this study, we improved the performance of the answer extraction by adding the features generated by dependency parsing analysis to the inputs of the answer extraction model (Bidirectional LSTM-CRF). The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. In this study, we compared the performance of the answer extraction model when inputting basic word features generated without the dependency parsing and the performance of the model when inputting the addition of the Eojeol tag feature and dependency graph embedding feature. Since dependency parsing is performed on a basic unit of an Eojeol, which is a component of sentences separated by a space, the tag information of the Eojeol can be obtained as a result of the dependency parsing. The Eojeol tag feature means the tag information of the Eojeol. The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. From the dependency parsing result, a graph is generated from the Eojeol to the node, the dependency between the Eojeol to the edge, and the Eojeol tag to the node label. In this process, an undirected graph is generated or a directed graph is generated according to whether or not the dependency relation direction is considered. To obtain the embedding of the graph, we used Graph2Vec, which is a method of finding the embedding of the graph by the subgraphs constituting a graph. We can specify the maximum path length between nodes in the process of finding subgraphs of a graph. If the maximum path length between nodes is 1, graph embedding is generated only by direct dependency between Eojeol, and graph embedding is generated including indirect dependencies as the maximum path length between nodes becomes larger. In the experiment, the maximum path length between nodes is adjusted differently from 1 to 3 depending on whether direction of dependency is considered or not, and the performance of answer extraction is measured. Experimental results show that both Eojeol tag feature and dependency graph embedding feature improve the performance of answer extraction. In particular, considering the direction of the dependency relation and extracting the dependency graph generated with the maximum path length of 1 in the subgraph extraction process in Graph2Vec as the input of the model, the highest answer extraction performance was shown. As a result of these experiments, we concluded that it is better to take into account the direction of dependence and to consider only the direct connection rather than the indirect dependence between the words. The significance of this study is as follows. First, we improved the performance of answer extraction by adding features using dependency parsing results, taking into account the characteristics of Korean, which is free of word order structure and omission of sentence components. Second, we generated feature of dependency parsing result by learning - based graph embedding method without defining the pattern of dependency between Eojeol. Future research directions are as follows. In this study, the features generated as a result of the dependency parsing are applied only to the answer extraction model in order to grasp the meaning. However, in the future, if the performance is confirmed by applying the features to various natural language processing models such as sentiment analysis or name entity recognition, the validity of the features can be verified more accurately.

• Asia pacific journal of information systems
• /
• v.21 no.1
• /
• pp.1-27
• /
• 2011

In an attempt to cope with widespread, dynamic, and accelerating changes in both internal and external business environments, companies often utilize information technologies such as SCM(Supply Chain Management). To date, SCM research has mainly focused on the effects of dynamic factors on SCM success and emphasized adoption strategies and critical success factors. Consequently, the effects of more static factors such as interdependency between SCM partners have been largely ignored. The purpose of this study, therefore, is to examine the effects of both dynamic and static factors on SCM performance by controlling for information quality and partnership quality. The five factors examined in this study include innovative ness, mutual dependency, quality of information, partnership quality, and SCM performance. All factors were examined from the perspective of part suppliers, except the mutual dependency which was examined from two aspects: supplier's dependency on customer and customer's dependency on supplier. Data was collected through five hundred survey questionnaires distributed to the part supplier companies that have implemented SCM systems for at least one year. As a result, a total of 170 valid responses were obtained. A structural equation research model was fitted using SAS 9.1.3 and SMART-PLS 2.0. The results of this study can be summarized as follows. First, innovativeness positively affected SCM information quality. SCM partnership quality, and ultimately SCM performance. The path coefficient between innovativeness and information quality was 0.387, with a t-value of 3.528. Innovativeness also had a positively direct effect on partnership quality. The path coefficient was 0.351 with a t-value of 3.366. The total effect of innovativeness on partnership quality was significant, although its indirect effect on partnership quality by altering information quality was negligible. The total indirect effect of innovativeness on SCM performance by affecting information quality and partnership quality was significant with a p-value of 0.014. Innovativeness played an important role in determining SCM performance. Second, mutual dependency showed no significant effect on SCM information quality. This result contradicts the earlier assertion that the more dependent two companies are, the more accurate and timely the information they exchange ought to be. This study showed that this may not be the case; a partner may provide information of poor quality even when it is strongly dependent on the other. Mutual dependency showed significant effect on partnership quality. However, when the mutual dependency perceived by suppliers was divided into two parts, one being a supplier's dependency on its customer company and the other being a customer's dependency on the supplier, the latter showed a significant impact on the perceived SCM partnership quality. This result indicates that a customer company can hardly improve the partnership quality perceived by suppliers by making them more dependent. It improves only when the suppliers perceive that their partners, typically having more bargaining power, are more dependent on them. The overall effect of mutual dependency of any kind on SCM performance, however, was not significant. Although mutual dependency has been mentioned as an important static factor influencing almost every aspect of cooperation on a supply chain, its influences may not be as significant as it was initially perceived to be. Third, the correlation between information quality and partnership quality was 0.448 with a p-value of less than 0.001. Information quality had a path coefficient of 0.256 to partnership quality with a t-value of 2.940. The quality of information exchanged between partners may have an impact on their partnership quality. Fourth, information quality also had a significant impact on SCM performance with a path coefficient of 0.325 with a t-value of 3.611. In this study, SCM performance was divided into four categories: product quality, cost saving, service quality, and order fulfillment. Information quality has Significant impacts on product quality, cost saving and service quality, but not on order fulfillment. Fifth, partnership quality, as expected, had a significant impact on SCM performance. The path coefficient was 0.403 with a t-value of 3.539. Partnership quality, like information quality, had positive impacts on product quality, cost saving and service quality, but showed no impact on order fulfillment. It seemed that order fulfillment is the hardest category of performance that SCM can satisfy. One major limitation of this study is that it surveyed only the suppliers. To better understand the dual aspects of SCM, it is important to survey both suppliers and the assemblers, especially in pairs. This research, to our best knowledge, was the first attempt to study the level of dependency between the two groups by measuring the dual aspects of SCM and studying mutual dependency from the categories of suppliers and assemblers each.. In the future, a more comprehensive and precise measurement of SCM characteristics needs to be achieved by examining from both the supplier's and assembler's perspectives.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.6
• /
• pp.924-931
• /
• 2008

This paper presents a buswise transmission loss allocation algorithm utilizing the transaction strategy. We prove that whatever calculated by any transaction strategy, the total of the allocated transmission losses of each bus, including no-load loss allocation, almost equals the total loss of AC power flow algorithm and the loss is perfectly slackbus-independent. In this paper, the allocated transmission losses of each bus is calculated by the method of integrating loss sensitivities using by the load level parameter ${\lambda}$. The performance of the proposed algorithm is evaluated by the case studies carried out on the WSCC 9-bus and IEEE 14-bus systems.

• Korean Journal of Computational Design and Engineering
• /
• v.1 no.1
• /
• pp.33-44
• /
• 1996

This study deals with the development of wire path searching module as a part of automobile wire harness design system. Wire path searching module manages the free space, finds transition locations, and creates bundle paths to dramatically reduce a tedious iterative routing process which results in easy optimization of the bundle paths. A prime policy in the system configuration is to compromise between man's and computer's ability, and make it possible a designer's leading role in designing process. Human input is indispensable to cope with the special cases which were not considered in the initial design stage of the system. In this study, we improve the previous shortest-path-finding algorithm, (VGraph and RCA method) into a new method called Extended RCA. Bundles, connectors and transitions are handled as objects so one can manage and modify physical properties of the objects easily. Therefore a verification is allowed at any desired stage of design. The reuse of previous result is facilitated by using Dependency Structure, which represents the mutual relations among connectors, transitions, and bundles. Dependency Structure makes it possible the elimination of redundant calculating process, and consequently shorter routing time.

• Journal of Internet Computing and Services
• /
• v.4 no.6
• /
• pp.43-55
• /
• 2003

Workflow mining is a newly emerging research issue for rediscovering and reengineering workflow models from workflow logs containing information about workflows being executed on the workflow engine. In this paper, we proposed workcase mining which was used dependency among activities. Main purpose of this paper is to minimize discrepancies between the modeled workflow process and the enacted workflow process as it is actually being executed. That is, we can get a complete set of activity firing sequences on buildtime. Besides, we can discover from workflow logs that which path out of all reachable paths a workcase has actually followed through on runtime. For this purpose we proposed two algorithm, the one is 'Activily-Dependent Net Algorithm' and the other is 'E-Walk Series Analysis Algorithm'.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.7
• /
• pp.1577-1584
• /
• 2015

In this paper, we propose a method for generating pattern classifier based on LFSR. The proposed pattern classifier bosed on LFSR is easy to see non-reachable state, and we can obtain dependency vector by using the 0-basic path. Also, we propose a method for generating pattern classifiers based on LFSR which correspond to given dependency vector.