• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2008.10b
• /
• pp.575-582
• /
• 2008

본 논문은 시맨틱 웹 기술이 어떻게 국가 표준(KS) 정보 검색 서비스 내 정보들을 연계시키고 사용자 접근성을 향상시키는 데 도움을 줄 수 있는지를 실증적으로 보여준다. 기존 표준 정보 검색 서비스는 용어 검색의 유연성이 부족하여 표준 정보에서 사용된 용어와 사용자 용어 간의 괴리를 해소하지 못했으며 표준, 기관, 인력 등 상호 관련성을 가진 개체 정보들을 개별적으로 서비스하였다. 이러한 상황은 사용자의 표준 정보 검색 서비스 접근성을 떨어뜨리는 요인으로 작용한다. 본 연구에서는 유의어, 관련어를 중심으로 한 표준 용어 사전 구축을 통해 사용자 용어와 표준 정보 내 용어 간의 원활한 매칭을 지원하며, 표준 관련 개체들을 온톨로지와 추론을 통해 연계시키는 방안을 제시한다. 개선된 표준 정보 검색 서비스는 개선된 표준 정보 검색 서비스는 개체 중심적 통합 검색 결과 제공 방식으로 관련 정보들을 단일 웹 페이지 내에서 확인할 수 있도록 해준다. 예를 들어, 특정 KS 표준 검색 결과 페이지에서는 기존에 DB 접근이나 검색 엔진을 통해 바로 획득할 수 없었던 정부 표준들, 기관들의 해당 KS 표준 인용 현황, 해당 KS 표준 전문가들, 부합화를 위해 참조된 국제 표준들, 해당 KS 표준 전문가들, 부합화를 위해 참조된 국제 표준들, 해당 KS 표준 전문가 네트워크, 해당 KS 표준 내 표준 용어 사전 정보 등 다양한 관련 정보들을 조합하여 서비스한다. 본 연구를 위해 모델링된 온톨로지와 시맨틱 웹기반 서비스 프레임워크인 OntoFrame 상에서 추론 작업이 표준 정보 적재 시에 전방 추론 (Forward-chaining) 방식으로 수행되었으며, 표준 온톨로지 질의 언어인 SPARQL (SPARQL Protocol and RDF Query Language)을 이용해 일반 검색 서비스 수준의 속도로 서비스될 수 있었다.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.721-732
• /
• 1995

The IPSE(Integrated Project Support Environment) tool can be seen as a result of an attempt to synthesize the key aspects of language-centered, specific methodology-based and toolkit oriented environments, which are current CASE tools into an organic whole. The IPSE approach based on a process model is regarded as an effective way to implement integrated CASE. The PM-CASE(Process Model based CASE) tool is currently a prototype which draw diagrams describing processes by using a new modeling technique. Attributes related with a task of withen the process model should be defined an saved the database. These attributed are used to retrieve the information of products, and to call the tool related which the task. In this paper, TSEE(Process centered Software Engineering Environment) tools are compared and analyzed. By describing the basic concept, architecture and design of PM-CASE tool, a method of building an process model-based CASE tool is proposed be support an effect software development and management.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.2
• /
• pp.130-142
• /
• 2008

Different from other ontology languages, TopicMap is capable of integrating numerous amount of heterogenous information resources using the locational information without any information transformation. Although many conventional editors have been developed for topic maps, they are standalone-type only for writing XTM documents. As a result, these tools request too much time for handling large-scale data and provoke practical problems to integrate with legacy systems which are mostly based on relational database. In this paper, we model a large-scale topic map structure based on XTM 1.0 into RDB structure to minimize the processing time and build up the ontology in legacy systems. We implement a topic map platform called X-TOP that can enhance the efficiency of ontology construction and provide interoperability between XTM documents and database. Moreover, we can use conventional SQL tools and other application development tools for topic map construction in X-TOP. The X-TOP is implemented to have 3-tier architecture to support flexible user interfaces and diverse DBMS. This paper shows the usability of X-TOP by means of the comparison with conventional tools and the application to healthcare cancer ontology management.

• Phonetics and Speech Sciences
• /
• v.12 no.2
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.1
• /
• pp.268-277
• /
• 2019

Flight test systems should monitor various conditions in real time during flight tests and take safety measures in an emergency. The importance of ensuring test safety increases in more complicated and wider test environments. Also, due to the transition of wartime operational authority, many guided missile systems must be developed simultaneously. Early deployment and budget reduction by shortening the development and T&E periods are also necessary. Consequently, the risk of flight tests under the circumstance of inefficient test resources is increasing. To address this deficiency, a flight test system model using SysML was proposed in this study. The method of designing and verifying the test system is based on the agile shift left testing methodology of advanced T&E labs and utilizing a system reference model in the aerospace field. Through modeling and simulation analysis, early identification and correction of faults resulting from inconsistent test requirements can mitigate the risk of delays during the T&E phase of flight tests. Also, because the flight test system model was constructed using SysML, it can be applied to test various guided missile systems.

• KIPS Transactions on Software and Data Engineering
• /
• v.13 no.1
• /
• pp.35-49
• /
• 2024

Video captioning technology, as a significant outcome of the integration between computer vision and natural language processing, has emerged as a key research direction in the field of artificial intelligence. This technology aims to achieve automatic understanding and language expression of video content, enabling computers to transform visual information in videos into textual form. This paper provides an initial analysis of the research trends in deep learning-based video captioning and categorizes them into four main groups: CNN-RNN-based Model, RNN-RNN-based Model, Multimodal-based Model, and Transformer-based Model, and explain the concept of each video captioning model. The features, pros and cons were discussed. This paper lists commonly used datasets and performance evaluation methods in the video captioning field. The dataset encompasses diverse domains and scenarios, offering extensive resources for the training and validation of video captioning models. The model performance evaluation method mentions major evaluation indicators and provides practical references for researchers to evaluate model performance from various angles. Finally, as future research tasks for video captioning, there are major challenges that need to be continuously improved, such as maintaining temporal consistency and accurate description of dynamic scenes, which increase the complexity in real-world applications, and new tasks that need to be studied are presented such as temporal relationship modeling and multimodal data integration.