• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2018년도 제30회 한글 및 한국어 정보처리 학술대회
• /
• pp.557-558
• /
• 2018

언어를 구사하는 능력은 연령별, 개인별로 차이가 있다. 이 능력은 평가 분류에 따라 수치적으로 평가될 수 있다. 그러나 수치로 평가하는 것 보다 비슷한 연령의 평균적인 능력과 비교를 통하여 능력을 평가하는 것이 일반적이다. 언어 병리학에서 평가를 하는 보편적인 방법은 언어 병리 전문가가 하는 것이다. 그러나 사람을 통한 평가 방법은 시간과 비용이 많이 소요될 뿐만 아니라 객관적이지 못한 평가라 할 수 있다. 따라서 이번 연구에서는 한국어 발화에 대한 분석을 자동화 하는 시스템을 제안한다.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2018년도 제30회 한글 및 한국어 정보처리 학술대회
• /
• pp.559-560
• /
• 2018

순환 신경망(RNN) 기반의 Long Short-Term Memory(LSTM)는 자연어처리 분야에서 우수한 성능을 보이는 모델이다. 음성을 문자로 변환해주는 Speech to Text (STT)를 이용해 자막을 생성하고, 생성된 자막을 다른 언어로 동시에 번역을 해주는 서비스가 활발히 진행되고 있다. STT를 사용하여 자막을 추출하는 경우에는 마침표가 없이 전부 연결된 문장이 생성되기 때문에 정확한 번역이 불가능하다. 본 논문에서는 영어자막의 자동 번역 시, 정확도를 높이기 위해 텍스트를 문장으로 분할하여 마침표를 생성해주는 방법을 제안한다. 이 때, LSTM을 이용하여 데이터를 학습시킨 후 테스트한 결과 62.3%의 정확도로 마침표의 위치를 예측했다.

• Journal of information and communication convergence engineering
• /
• 제6권1호
• /
• pp.59-63
• /
• 2008

The methods or algorithms which can accelerate simulation speed became of great importance, as the modeling and simulation methodology for discrete event systems is used in many areas such as model validation/verification and performance evaluation. This paper proposes a tool named, DEVSIF composer. The tool is made of an automated compiled simulation technology and it builds a new composed model which can be executed much fast by composing the component models together. Models are described by our new specification language DEVSIF, which is compatible with object-oriented language and supports representation of a hierarchical model structure. Experimental results demonstrates that DEVSIF composer enhances the simulation speed of a transformed DEVS model 5 times faster than that of the original ones in average.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2014년도 제26회 한글 및 한국어 정보처리 학술대회
• /
• pp.119-121
• /
• 2014

커뮤니티 질의응답 시스템(cQA)은 기존에 구축된 '질문-답' 쌍에서 사용자의 질문과 비교하여 유사도 순으로 결과를 보여주는 시스템이다. 본 논문에서는 '국립국어원'의 질의응답 게시판에 적용 가능한 '커뮤니티 질의응답 시스템'을 소개하고, 국립국어원 질의응답 게시판의 질문 특성을 분석하여 cQA의 성능 향상을 위한 자질 추출 방법을 제시한다.

• Journal of Information Processing Systems
• /
• 제15권6호
• /
• pp.1378-1391
• /
• 2019

In recent years, automatic document summarization has been widely studied in the field of natural language processing thanks to the remarkable developments made using deep learning models. To decode a word, existing models for abstractive summarization usually represent the context of a document using the weighted hidden states of each input word when they decode it. Because the weights change at each decoding step, these weights reflect only the local context of a document. Therefore, it is difficult to generate a summary that reflects the overall context of a document. To solve this problem, we introduce the notion of a general context and propose a model for summarization based on it. The general context reflects overall context of the document that is independent of each decoding step. Experimental results using the CNN/Daily Mail dataset show that the proposed model outperforms existing models.

• 디지털융복합연구
• /
• 제5권1호
• /
• pp.69-76
• /
• 2007

This paper presents a study on the introduction of User Model-Based Dialogue System. Also we present a plan-based Korean dialogue system as a natural language database user interface for product search. The system can be characterized by its support for mixed initiative to give user more control over dialogue, employment of user model to reflect user's preferences, alternative solution suggestion if there is no product matched exactly to user's requirements, handling circumlocution which frequently occurs in dialogues. The user modeling shell system BGP-MS is adapted for the system. The system provides for a user-friendly database user interface by managing dialogue intelligently. By its implementation and test, it has been shown that the user model-based dialogue system can be utilized effectively for product search.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2013년도 제25회 한글 및 한국어 정보처리 학술대회
• /
• pp.107-109
• /
• 2013

주식 시장에서 거래되고 있는 증권은 MACD(Moving Average Convergence Divergence), Stochastic 등의 보조 지표를 이용하는 기술적 분석을 통하여 매수/매도 시점을 결정한다. 주식 시장의 객관적인 자료를 통하여 분석하는 기술적 분석 방법은 주식 시장 외적인 요소를 반영하는데 있어 한계점이 존재한다. 본 논문에서는 기술적 분석 방법에 기사를 종목별로 분류하고 기사의 긍정 및 부정을 판별하는 문서 분류 기법을 적용하여 주식 외적인 요소를 반영하는 시스템을 제안한다.

• Journal of information and communication convergence engineering
• /
• 제17권4호
• /
• pp.234-238
• /
• 2019

Existing web applications and services have historically been implemented using JavaScript. However, new technologies such as artificial intelligence, the Internet of Things, and Big Data are being developed as part of the Fourth Industrial Revolution. With the definition of the HTML5 web standard, services (such as the technologies mentioned above) that were previously not available through the Web become available. These services, however, need to have the same performance as native applications, and implementing these services will require new technologies. Therefore, additional tools that can work on the Web with native performance are needed. In this paper, a system for branching processing was established using JavaScript and WebAssembly, a language that can operate on the Web. This system performs user requests in advance, and requests are branched in a language that produces faster results. Therefore, a service capable of quick response times can be implemented.

• Journal of information and communication convergence engineering
• /
• 제17권4호
• /
• pp.239-245
• /
• 2019

Recently, effort to obtain various information from the vast amount of social network services (SNS) big data generated in daily life has expanded. SNS big data comprise sentences classified as unstructured data, which complicates data processing. As the amount of processing increases, a rapid processing technique is required to extract valuable information from SNS big data. We herein propose a system that can extract human sentiment information from vast amounts of SNS unstructured big data using the naïve Bayes algorithm and natural language processing (NLP). Furthermore, we analyze the effectiveness of the proposed method through various experiments. Based on sentiment accuracy analysis, experimental results showed that the machine learning method using the naïve Bayes algorithm afforded a 63.5% accuracy, which was lower than that yielded by the NLP method. However, based on data processing speed analysis, the machine learning method by the naïve Bayes algorithm demonstrated a processing performance that was approximately 5.4 times higher than that by the NLP method.

• Journal of information and communication convergence engineering
• /
• 제17권2호
• /
• pp.117-127
• /
• 2019

In software engineering, to ensure reliability between systems, describing both system architecture and assurance arguments between system elements is considered necessary. There are proposals for system architecture assurance, but use of these traditional methods often requires development of different diagrams using different editors. Because the visual sense of the traditional methods is inadequate, errors readily occur when manipulating different diagrams. Therefore, it is essential that the assurance of dependability between components and systems is visualized and easy to understand. In this paper, an integrated approach to describe the relationship between system actors and system architecture is proposed, and this approach is clarified using an enterprise architecture modeling language. A case study is carried out and comparison to the traditional approach $d^*$ framework is explained. The comparison results show that the proposed approach is more suitable for ensuring dependability in system architecture.