• Journal of Digital Convergence
• /
• v.15 no.6
• /
• pp.505-511
• /
• 2017

The purpose of this study is to identify the characteristic of the length of sentence of sports news articles compared with other sections, other newspaper and sports events. In order to achieve this purpose, 4 daily newspapers(Chosun Ilbo, Dong-A Ilbo, Kyunghyang Shinmun and MKsports) were selected as research sources. The length of sentence of collected articles was analyzed by One way ANOVA and Scheffe as post-hoc. The results of these analyses are summarized into three as follows: First, the length of sentence of news articles has significant difference by sections. To put it concretely, the sentences in politics section were the longest, and then economy, editorials comes. The sports section has the shortest sentences. Second, the length of sports articles has significant difference by newspapers. In particular, MK sports which is a sports-oriented newspaper has the shortest in the length of sentence of sports articles than other newspapers such as Chosun, Dong-A and Kyunghyang. Third, the length of sentence of sports articles has significant differences by the sports events. More specifically golf is the longest, and then basketball, soccer and baseball follows.

• Korean journal of communication and information
• /
• v.79
• /
• pp.140-164
• /
• 2016

What is an appropriate length of an article in the mobile era producing new tastes of contents combining computing and mobile communication? Is it still valid to have a lengthy article as high level journalism even in the mobile era? Is there any possibility to have a short article combination to give us readability and the amount of information? This study aims to find some answers to these questions. The length of articles were controlled with a field experiment; from an article of 346 syllables(including spaces between syllables) which does not need finger scrolling on a mobile phone to articles of 633, 1033, 1368 syllables(including spaces between syllables) which frequently appear in newspapers and broadcasting news program. All the main themes were same for 4 articles which have different length. Three hundred and eighty four students viewed one of the 4 articles through mobile phone or newspaper. Each participant checked their preference and evaluated quality on the article, were asked to recall contents of the article. In newspaper group, articles with 346 or 1033 syllables were highly evaluated. Mobile group seemed to prefer articles of 346 or 633 syllables. In conclusion, we need to consider various strategies shortening the length of an article into 346 or 633 syllables as a basic format to meet the neEds of mobile era.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.495-500
• /
• 2020

뉴스 기사는 반드시 객관적이고 넓은 시각으로 정보를 전달하지 않는다. 따라서 뉴스 기사를 기존의 추천 시스템과 같이 개인의 관심사나 사적 정보를 바탕으로 선별적으로 추천하는 것은 바람직하지 않다. 본 논문에서는 최대한 객관적으로 다양한 시각에서 비슷한 사건과 인물에 대해서 판단할 수 있도록 유사도 기반의 기사 추천 모델을 제시한다. 길이가 긴 문서 사이의 유사도를 측정하기 위해 GPT2 [1]언어 모델을 활용했다. 이 과정에서 단방향 디코더 모델인 GPT2 [1]의 단점을 추가 학습으로 개선했으며, 저장 공간의 효율과 핵심 문단 추출을 위해 BM25 [2]함수를 사용했다. 그리고 준 지도 학습 [3]을 통해 유사도 레이블링이 되어있지 않은 최신 뉴스 기사에 대해서도 자가 학습을 진행했으며, 이와 함께 길이가 긴 문단에 대해서도 효과적으로 학습할 수 있도록 문장 길이를 기준으로 3개의 단계로 나누어진 커리큘럼 학습 [4]방식을 적용했다.

• Journal of the Korean Society for information Management
• /
• v.18 no.3
• /
• pp.159-178
• /
• 2001

This paper presents an automatic text summarization model which selects representative sentences from sentence clusters to create a summary. Summary generation experiments were performed on two sets of test documents after learning the optimum environment from a training set. Centroid clustering method turned out to be the most effective in clustering sentences, and sentence weight was found more effective than the similarity value between sentence and cluster centroid vectors in selecting a representative sentence from each cluster. The result of experiments also proves that inverse sentence weight as well as title word weight for terms and location weight for sentences are effective in improving the performance of summarization.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.361-363
• /
• 2001

본 논문은 신문 기사문에 특정적인 정보 추출의 내용과 방법을 제안한다. 신문 기사에서 이용자가 원하는 정보 추출의 내용으로 육하원칙을 중심으로 한 다섯 가지 정보를 제시하였으며, 이를 추출하기 위해 통계적인 기법을 주로 이용하고 부분적으로 언어적 지식을 이용하였다. 본 논문에서는 비교적 문서의 길이가 짧은 신문기사문을 요약 대상으로 하므로 단락이나 문장이 아닐 절 이하 단위로 추출하며, 중심절을 추출한 뒤 그 절과의 관계를 통해 나머지 정보들을 추출함으로써 추출되는 내용이 유사하거나 산만하지 않기 때문에 이 추출 정보로 요약문을 생성할 경우에 긴밀한 요약문을 생성할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.869-872
• /
• 2018

온라인 뉴스 소비의 확산과 함께 댓글은 여론 형성에 큰 역할을 담당한다. 그러나 아직 댓글에 영향을 미치는 형식 요소에 대한 실증 데이터 기반의 연구는 미흡하다. 본 연구는 이의 시작으로 온라인 뉴스 기사 소비의 두 가지 중요 요소 즉, 헤드라인과 댓글의 관계에 대해 다루고자 한다. 이를 위해, 헤드라인의 논조 유무에 따른 댓글의 논쟁 활성화 정도 차를 확인하고자 댓글의 수와 길이를 분석하였다. '이세돌, 알파고 바둑대결', '최저임금', '북미회담' 기사로 총 537건의 해드라인과 약 85만개의 댓글을 수집하였다. 그 결과 논쟁 활성화 측면에서 논조가 있는 헤드라인일때 댓글의 수가 많고 길이가 길어 논쟁이 더 활발한 것을 할 수 있었다. 또, 댓글의 논쟁 주제도 차이가 있어 헤드라인의 논조가 있는 경우에 의견이나 감정을 표출하는 토픽이 더 많았다. 본 연구는 실증 데이터를 통해, 헤드라인의 논조 유무가 댓글의 논쟁의 활성화 정도와 주제에 영향을 주는 요소임을 밝힘으로써 댓글 소비에 대한 새로운 관점을 제시하고, 헤드라인의 형식 요소의 연구의 중요성을 확인한 데 그 의의가 있다.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.505-515
• /
• 2004

In a newspaper, 5W1H information is the most fundamental and important element for writing and understanding articles. Focusing on such a relation between a newspaper article and the 5W1H, we propose a summarization method based on the activation degree of 5W1H. To overcome problems of the lead-based and the title-based methods, both of which are known to be the most effective in newspaper summarization, sufficient 5W1H information is extracted from both a title and a lead sentence. Moreover, for each sentence, its weight is computed by considering various factors, such as activation degree of 5W1H, the number of 5W1H categories, and its length and position. These factors make a great contribution to the selection of more important sentences, and thus to the improvement of readability of the summarized texts. In an experimental evaluation, the proposed method achieved a precision of 74.7% outperforming the lead-based method. In sum, our 5W1H approach was shown to be promising for automatic summarization of newspaper articles.

• The Journal of the Korea Contents Association
• /
• v.22 no.4
• /
• pp.159-171
• /
• 2022

This study examined the problem of utilizing celebrity SNS in online news, which have increased by an average of 745 every year since 2010, reaching about 10,000 in 2021. 40 online newspapers were selected and 202,730 news articles produced by these newspapers in July 2021 were analyzed. As a result, 1.27% (2,582) of all articles were found to be using celebrity SNS as a source. This indicates that on average, online newspapers produce 2.08 celebrity SNS-utilized articles per day and 64.7 articles per month. Specifically, entertainer SNS (53.7%) was used the most compared to SNS of politician(39.8%) and influencer(6.5%). Instagram(69.1%, 57.1%) was utilized the most for entertainer and influencer and this were mostly related to personal information. On the other hand, Facebook(70.4%) was cited the most for politician, mostly related to opinions on social/political issues. The average length of SNS-based articles was 536 characters. The problem with news articles utilizing SNS is that most articles simply copy the SNS content without additional coverage(88.4%), and 14% of the articles did not disclose the exact source. Implication of the research on 40 online news agency is discussed.

• Bulletin of the Korean Space Science Society
• /
• 1994.09a
• /
• pp.56-60
• /
• 1994

• The KIPS Transactions:PartB
• /
• v.12B no.6 s.102
• /
• pp.691-696
• /
• 2005

In this paper, we present a pilot system that tn compress a Korean sentence automatically using knowledge extracted from news articles and their headlines. A sot of compressed sentences can be presented as an abstraction of a document. As a compressed sentence is of headline-style, it could be easily displayed on small devices, such as mobile phones and other handhold devices. Our compressing system has shown to be promising through a preliminary experiment.