• Proceedings of the Korean Society for Language and Information Conference
• /
• 1996.02a
• /
• pp.137-146
• /
• 1996

Our research goal has been the development of a domain independent natural language processing (NLP) system suitable for information retrieval. As part of that research, we have investigated ways to automatically extend the semantics of a lexicon derived from machine-readable lexical sources. This paper details the extraction of thematic roles derived from lexical patterns in a machine-readable dictionary.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.141-149
• /
• 1992

본 논문에서는 자연언어 처리 시스템에서 사용할 수 있는 단어의 개념 정보에 대해 연구한다. 그러기 위해 언어의 실생활에서의 쓰임 - 연세대학교 사전 편찬실의 연세 말뭉치 -을 바탕으로 한국어의 동사와 명사에 대해 개념을 조사, 분류하였으며 그 개념들이 한국어 문장 분석기에 어떻게 이용되는가를 보인다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.254-256
• /
• 2020

자연언어처리 문제에서 딥러닝 모델이 좋은 성능을 보이고 있고 딥러닝 결과는 구조화된 결과를 내놓는 경우가 많다. 딥러닝 모델 결과가 구조적인 형태를 가지는 경우 후처리 통해 특정 구조에 맞는 제약을 가해주는 경우가 일반적이다. 본 논문에서는 이러한 제약을 규칙에 기반하지 않고 직접 학습을 통해 얻고자 하였다.

• Journal of Internet Computing and Services
• /
• v.1 no.2
• /
• pp.49-59
• /
• 2000

The majority of work done to date on natural language processing has focused on analysis and understanding of language, thus natural language generation had been relatively less attention than understanding, And people even tends to regard natural language generation CIS a simple reverse process of language understanding, However, need for natural language generation is growing rapidly as application systems, especially multi-language machine translation systems on the web, natural language interface systems, natural language query systems need more complex messages to generate, In this paper, we propose an algorithm to generate more flexible and natural sentence using lexical functions of Igor Mel'uk (Mel'uk ＆ Zholkovsky, 1988) and systemic grammar.

• Journal of Society of Preventive Korean Medicine
• /
• v.17 no.2
• /
• pp.179-187
• /
• 2013

Objectives : In many cases, patient's symptoms have been recorded on EMR in natural language instead of medical terminologies. It is possible to build a database by analyzing the symptoms of Korean Medicine(KM) that indicates patient's symptoms in natural language. Using the database, when doctors record patient's symptoms on EMR in natural language, conversely it'll be also possible to extract the symptoms of KM from those natural language. The database will enhance the value of EMR as a medical data. Methods : In this study, we aimed to make data structure of the terminologies that represent the symptoms of KM. The data structure is combinations of smallest unit in natural language. We made the database by analyzing morpheme of the natural language to express the symptoms of KM. Results & Conclusions : By classifying the natural language in 15 features, we made the structure of concept and the data available for morphological analysis.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.300-304
• /
• 2022

본 논문은 한국어 법률 텍스트 처리를 위해 세 가지 서로 다른 사전 학습 모델을 미세 조정하여 그 성능을 평가하였다. 성능을 평가하기 위해 타겟 판결 요지에 대한 판결 요지 후보를 추출하여 판결 요지 간의 유사도를 계산하였다. 또한 유사도를 바탕으로 추출된 판결 요지가 실제 법률 전문가와 일반 언어학자의 직관에 부합하는지 판단하기 위해 정성적 평가를 진행하였다. 그 결과 법률 전문가가 법률 전문 지식이 없는 일반 언어학자에 비해 판결 요지 간 유사도를 낮게 평가하였는데 법률 전문가가 법률 텍스트의 유사성을 판단하는 기준이 기계와 일반 언어학자와는 달라 전문가 자문에 기반한 한국어 법률 AI 모델 개발의 필요성을 확인하였다. 최종 연구 결과로 한국어 법률 AI 프레임워크를 제안하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.2E
• /
• pp.38-46
• /
• 1998

In this paper, we propose a remote speech input device, a new method of user-friendly speech input in spontaneous speech recognition system. We focus the user friendliness on hands-free and microphone independence in speech recognition applications. Our method adopts two algorithms, the automatic speech detection and the microphone array delay-and-sum beamforming (DSBF)-based speech enhancement. The automatic speech detection algorithm is composed of two stages; the detection of speech and nonspeech using the pitch information for the detected speech portion candidate. The DSBF algorithm adopts the time domain cross-correlation method as its time delay estimation. In the performance evaluation, the speech detection algorithm shows within-200 ms start point accuracy of 93%, 99% under 15dB, 20dB, and 25dB signal-to-noise ratio (SNR) environments, respectively and those for the end point are 72%, 89%, and 93% for the corresponding environments, respectively. The classification of speech and nonspeech for the start point detected region of input signal is performed by the pitch information-base method. The percentages of correct classification for speech and nonspeech input are 99% and 90%, respectively. The eight microphone array-based speech enhancement using the DSBF algorithm shows the maximum SNR gaing of 6dB over a single microphone and the error reductin of more than 15% in the spontaneous speech recognition domain.

• Journal of Internet Computing and Services
• /
• v.13 no.2
• /
• pp.99-108
• /
• 2012

Many methods had been developed to improve the accuracy of extracting information from a vast amount of data. This paper combined a number of natural language processing methods such as NER (named entity recognition), sentence extraction, and part of speech tagging to carry out text analysis. The data source is comprised of texts obtained from the web using a domain-specific data extraction agent. A framework for the extraction of information from unstructured data was developed using the aforementioned natural language processing methods. We simulated the performance of our work in the extraction and analysis of texts for the detection of organizational structures. Simulation shows that our study outperformed other NER classifiers such as MUC and CoNLL on information extraction.

• Journal of the Korean Society for information Management
• /
• v.23 no.2
• /
• pp.167-183
• /
• 2006

This paper describes a metadata extraction technique based on natural language processing (NLP) which extracts personalized information from email communications between financial analysts and their clients. Personalized means connecting users with content in a personally meaningful way to create, grow, and retain online relationships. Personalization often results in the creation of user profiles that store individuals' preferences regarding goods or services offered by various e-commerce merchants. We developed an automatic metadata extraction system designed to process textual data such as emails, discussion group postings, or chat group transcriptions. The focus of this paper is the recognition of emotional contents such as mood and urgency, which are embedded in the business communications, as metadata.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.6 no.1
• /
• pp.29-35
• /
• 2012

Most Augmentative and Alternative Communication(AAC) use exclusive equipment or studied desktop, tablet PC based windows. Besides, the preceding study offers proper noun dictionary so, henceforward study has problem to innumerable proper noun processing. This paper suggests a method of proper noun processing using a mobile smart equipment. And via the button with virtual keyboard input method and the errors that can occur is also proposing a complementary way. AAC system to check availability for application on Android has been implemented. Experimental results, depending on user location and selection of proper nouns in the around could produce a sentence is derived.