• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.309-313
• /
• 2017

개체명은 인명, 지명, 조직명 등 문서 내에서 중요한 의미를 가지므로 질의응답, 요약, 기계번역 분야에서 유용하게 사용되고 있다. 개체명 인식은 문서에서 개체명에 해당하는 단어를 찾아 개체명 범주를 부착하는 작업을 말한다. 개체명 인식 연구에는 개체명 범주가 부착된 개체명 말뭉치를 사용한다. 개체명의 범주는 연구 분야에 따라 다양하게 정의되므로 연구 분야에 적합한 개체명 말뭉치가 필요하다. 하지만 이런 말뭉치를 구축하는 일은 시간과 인력이 많이 필요하다. 따라서 본 논문에서는 개체명 사전 기반의 반자동 말뭉치 구축 도구를 제안한다. 제안하는 도구는 크게 전처리, 사용자 태깅, 후처리 단계로 나뉜다. 전처리 단계는 자동으로 개체명을 찾는 단계이다. 약 11만 개의 개체명을 기반으로 하여 트라이(trie) 구조의 개체명 사전을 구축한 후 사전을 이용하여 개체명을 자동으로 찾는다. 사용자 태깅 단계는 사용자가 수동으로 개체명을 태깅하는 단계이다. 전처리 단계에서 찾은 개체명 중 오류가 있는 개체명들은 수정하거나 삭제하고, 찾지 못한 개체명들은 사용자가 추가로 태깅하는 단계이다. 후처리 단계는 태깅한 결과로부터 사전 정보를 갱신하는 단계이다. 제안한 말뭉치 구축 도구를 이용하여 752개의 뉴스 기사에 대해 개체명을 태깅한 결과 7,620개의 개체명이 사전에 추가되었다. 제안한 도구를 사용한 결과 사용하지 않았을 때 비해 약 57.6% 정도 태깅 횟수가 감소했다.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.285-290
• /
• 2018

논지가 높은 설득력을 갖기 위해서는 충분한 지지 근거가 필요하다. 논지 내의 주장을 논리적으로 지지할 수 있는 근거 자료 추출의 자동화는 자동 토론 시스템, 정책 투표에 대한 의사 결정 보조 등 여러 어플리케이션의 개발 및 상용화를 위해 필수적으로 해결되어야 한다. 하지만 웹문서로부터 지지 근거를 추출하는 시스템을 위해서는 다음과 같은 두 가지 연구가 선행되어야 하고, 이는 높은 성능의 시스템 구현을 어렵게 한다: 1) 논지의 주제와 직접적인 관련성은 낮지만 지지 근거로 사용될 수 있는 정보를 확보하기 위한 넓은 검색 범위, 2) 수집한 정보 내에서 논지의 주장을 명확하게 지지할 수 있는 근거를 식별할 수 있는 인지 능력. 본 연구는 높은 정밀도와 확장 가능성을 가진 지지 근거 추출을 위해 다음과 같은 단계적 지지 근거 추출 시스템을 제안한다: 1) TF-IDF 유사도 기반 관련 문서 선별, 2) 의미적 유사도를 통한 지지 근거 1차 추출, 3) 신경망 분류기를 통한 지지 근거 2차 추출. 제안하는 시스템의 유효성을 검증하기 위해 사설 4008개 내의 주장에 대해 웹 상에 있는 845675개의 뉴스에서 지지 근거를 추출하는 실험을 수행하였다. 주장과 지지 근거를 주석한 정보에 대하여 성능 평가를 진행한 결과 본 연구에서 제안한 단계적 시스템은 1,2차 추출 과정에서 각각 0.41, 0.70의 정밀도를 보였다. 이후 시스템이 추출한 지지 근거를 분석하여, 논지에 대한 적절한 이해를 바탕으로 한 지지 근거 추출이 가능하다는 것을 확인하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.7
• /
• pp.285-292
• /
• 2014

A named entity recognition method is used to improve the performance of information retrieval systems, question answering systems, machine translation systems and so on. The targets of the named entity recognition are usually PLOs (persons, locations and organizations). They are usually proper nouns or unregistered words, and traditional named entity recognizers use these characteristics to find out named entity candidates. The titles of books, movies and TV programs have different characteristics than PLO entities. They are sometimes multiple phrases, one sentence, or special characters. This makes it difficult to find the named entity candidates. In this paper we propose a method to quickly extract title named entities from news articles and automatically build a named entity dictionary for the titles. For the candidates identification, the word phrases enclosed with special symbols in a sentence are firstly extracted, and then verified by the SVM with using feature words and their distances. For the classification of the extracted title candidates, SVM is used with the mutual information of word contexts.

• Annual Conference on Human and Language Technology
• /
• 2000.10d
• /
• pp.323-328
• /
• 2000

국내에서 영한 자동번역 시스템을 1985 년부터 개발한 지 벌써 15년이 흐르고 있다. 15 년의 영한 자동번역 기술개발에도 불구하고 아직도 영한 자동번역 시스템의 번역품질은 40%를 넘지 못하고 있다. 이렇게 번역품질이 낮은 이유는 다음과 같이 요약할 수 있을 것이다. o 입력문에 대해 파싱할 때 오른쪽 경계를 잘못 인식함으로써 구조적 모호성의 발생문제: 예를 들어 등위 접속절에서 오른쪽 등위절이 등위 접속절에 포함되는 지의 모호성. o 번역 단위로써 전체 문장을 대상으로 한 번역패턴이 아닌 구나 절과 같은 부분적인 번역패턴으로 인한 문장 전체의 잘못된 번역 결과 발생. o 점차 증가하는 대용량 번역지식의 구축과 관련해 새로 구축되는 번역 지식과 기구축된 대용량 번역지식들 간의 상호 충돌로 인한 번역 품질의 저하. 이러한 심각한 원인들을 극복하기 위해 본 논문에서는 문틀에 기반한 새로운 영한 자동번역 방법론을 소개하고자 한다. 이 문틀에 기반한 영한 자동번역 방법론은 현재 CNN뉴스 방송 자막을 대상으로 한 영한 자동번역 시스템에서 실제 활용되고 있다. 이 방법론은 기본적으로 data-driven 방법론에 속하다. 문틀 기반 자동번역 방법론은 규칙기반 자동번역 방법론보다는 낮은 단계에서 예제 기반 자동번역 방법론보다는 높은 단계에서 번역을 하는 번역방법론이다. 이 방법론은 영한 자동번역에 뿐만 아니라 다른 언어쌍에서의 번역에도 적용할 수 있을 것이다.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.773-780
• /
• 2016

Semantic information and features are very important for Semantic Role Labeling(SRL) though many SRL systems based on machine learning mainly adopt lexical and syntactic features. Previous SRL research based on semantic information is very few because using semantic information is very restricted. We proposed the SRL system which adopts semantic information, such as named entity, word sense disambiguation, filtering adjunct role based on sense, synonym cluster, frame extension based on synonym dictionary and joint rule of syntactic-semantic information, and modified verb-specific numbered roles, etc. According to our experimentations, the proposed present method outperforms those of lexical-syntactic based research works by about 3.77 (Korean Propbank) to 8.05 (Exobrain Corpus) F1-scores.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1880-1889
• /
• 2022

The online portal platform provides online news with online comments, but the anonymity of comments causes incivility, and online comments are considered social problems. While there are many foreign language-based incivility detection studies, in-depth research is not being conducted in Korea since there has not been implemented Korean language dataset which is labeled detailed criteria of incivility. In this study, the incivility notation of comments was conducted in a total of 13 items, uncivil words were summarized. Furthermore, Attention algorithm was applied to each comment and summary to extract embedding vectors. 2-d CNN followed at the end to detect incivility in given data. As a result, we showed that the proposed algorithm is useful for anti-citizen detection such as name-calling and offensive tones. This study is expected to contribute to the formation of a healthy online comment culture by detecting uncivil comments which hinder democratic discourse.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.4
• /
• pp.65-76
• /
• 2023

In the field of technology entrepreneurship and startups, the development of Artificial Intelligence(AI) has emerged as a key topic for business model innovation. As a result, venture firms are making various efforts centered on AI to secure competitiveness(Kim & Geum, 2023). The purpose of this study is to analyze the relationship between the development of GenAI technology and the startup ecosystem by analyzing domestic news articles to identify trends in the technology startup field. Using BIG Kinds, this study examined the changes in GenAI-related news articles, major issues, and trends in Korean news articles from 1990 to August 10, 2023, focusing on the emergence of ChatGPT before and after, and visualized the relevance through network analysis and keyword visualization. The results of the study showed that the mention of GenAI gradually increased in the articles from 2017 to 2023. In particular, OpenAI's ChatGPT service based on GPT-3.5 was highlighted as a major issue, indicating the popularization of language model-based GenAI technologies such as OpenAI's DALL-E, Google's MusicLM, and VoyagerX's Vrew. This proves the usefulness of GenAI in various fields, and since the launch of ChatGPT, Korean companies have been actively developing Korean language models. Startups such as Ritten Technologies are also utilizing GenAI to expand their scope in the technology startup field. This study confirms the connection between GenAI technology and startup entrepreneurship activities, which suggests that it can support the construction of innovative business strategies, and is expected to continue to shape the development of GenAI technology and the growth of the startup ecosystem. Further research is needed to explore international trends, the utilization of various analysis methods, and the possibility of applying GenAI in the real world. These efforts are expected to contribute to the development of GenAI technology and the growth of the startup ecosystem.

• The KIPS Transactions:PartB
• /
• v.12B no.1 s.97
• /
• pp.81-90
• /
• 2005

In this paper, we propose a novel approach which models multilingual story link detection by adapting the features such as timelines and multilingual spaces as weighting components to give distinctive weights to terms related to events. On timelines term significance is calculated by comparing term distribution of the documents on that day with that on the total document collection reported, and used to represent the document vectors on that day. Since two languages can provide more information than one language, term significance is measured on each language space and used to refer the other language space as a bridge on multilingual spaces. Evaluating the method on Korean and Japanese news articles, our method achieved $14.3{\%}\;and\;16.7{\%}$ improvement for mono- and multi-lingual story pairs, and for multilingual story pairs, respectively. By measuring the space density, the proposed weighting components are verified with a high density of the intra-event stories and a low density of the inter-events stories. This result indicates that the proposed method is helpful for multilingual story link detection.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.125-132
• /
• 2022

Sentence compression is a natural language processing task that generates concise sentences that preserves the important meaning of the original sentence. For grammatically appropriate sentence compression, early studies utilized human-defined linguistic rules. Furthermore, while the sequence-to-sequence models perform well on various natural language processing tasks, such as machine translation, there have been studies that utilize it for sentence compression. However, for the linguistic rule-based studies, all rules have to be defined by human, and for the sequence-to-sequence model based studies require a large amount of parallel data for model training. In order to address these challenges, Deleter, a sentence compression model that leverages a pre-trained language model BERT, is proposed. Because the Deleter utilizes perplexity based score computed over BERT to compress sentences, any linguistic rules and parallel dataset is not required for sentence compression. However, because Deleter compresses sentences only considering perplexity, it does not compress sentences by reflecting the linguistic information of the words in the sentences. Furthermore, since the dataset used for pre-learning BERT are far from compressed sentences, there is a problem that this can lad to incorrect sentence compression. In order to address these problems, this paper proposes a method to quantify the importance of linguistic information and reflect it in perplexity-based sentence scoring. Furthermore, by fine-tuning BERT with a corpus of news articles that often contain proper nouns and often omit the unnecessary modifiers, we allow BERT to measure the perplexity appropriate for sentence compression. The evaluations on the English and Korean dataset confirm that the sentence compression performance of sentence-scoring based models can be improved by utilizing the proposed method.

• The Journal of Bigdata
• /
• v.9 no.1
• /
• pp.75-88
• /
• 2024

Recently, there have been cases reported in the news of individuals experiencing symptoms of food poisoning after consuming raw beef purchased from online platforms, or reviews claiming that cherry tomatoes tasted bitter. This suggests the potential for analyzing food reviews on online platforms to detect food hazards, enabling government agencies, food manufacturers, and distributors to manage consumer food safety risks. This study proposes a classification model that uses sentiment analysis and large language models to analyze food reviews and detect negative ones, multi-labeling key food safety hazards (food poisoning, spoilage, chemical odors, foreign objects). The sentiment analysis model effectively minimized the misclassification of negative reviews with a low False Positive rate using a 'funnel' model. The multi-labeling model for food safety hazards showed high performance with both recall and accuracy over 96% when using GPT-4 Turbo compared to GPT-3.5. Government agencies, food manufacturers, and distributors can use the proposed model to monitor consumer reviews in real-time, detect potential food safety issues early, and manage risks. Such a system can protect corporate brand reputation, enhance consumer protection, and ultimately improve consumer health and safety.