• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.680-682
• /
• 2015

최근 스마트폰의 보급으로 소셜 네트워크 서비스를 이용하는 사용자들이 급증하였다. 그 중 트위터는 정보의 빠른 전파력과 확산성으로 인해 현실에서 발생한 이벤트를 탐지하는 도구로 활용하는 것이 가능하다. 따라서 트위터 사용자 개개인을 하나의 센서로 가정하고 그들이 작성한 트윗 텍스트를 분석한다면 이벤트 탐지의 도구로써 활용할 수 있다. 이와 관련된 연구들은 이벤트 발생 위치를 추적하기 위해 GPS좌표를 이용하지만 트위터 사용자들이 위치정보 공개에 회의적인 점을 감안하면 명확한 한계점으로 제시될 수 있다. 이에 본 논문에서는 트위터에서 제공하는 위치정보를 이용하지 않고, 트윗 텍스트에서 위치정보를 추적하는 방법을 제시하였다. 트윗 텍스트에서 키워드간의 관계를 고려하여 이벤트의 사실여부를 결정하였으며, 실험을 통해 기존 매체들보다 빠른 탐지를 보임으로써 제안된 시스템의 필요성을 보였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.5
• /
• pp.27-35
• /
• 2013

Recently, as Social Network Services(SNS) are becoming more popular, much research has been doing on analyzing public opinions from SNS. One of the most important tasks for solving such a problem is to separate opinion(subjective) documents from others(e.g. objective documents) in SNS. In this paper, we propose a new method of retrieving the opinion documents from Twitter. The reason why it is not easy to search or classify the opinion documents in Twitter is due to a lack of publicly available Twitter documents for training. To tackle the problem, at first, we build a machine-learned model for sentiment classification using the external documents similar to Twitter, and then modify the model to separate the opinion documents from Twitter. Experimental results show that proposed method can be applied successfully in opinion classification.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.613-621
• /
• 2021

This study conducted a network analysis to determine how COVID-19 vaccine-related issue spread on Twitter during the introduction stage of the COVID-19. Issue diffusion tendency is analyzed according to the time period: phase 1 (initiation of vaccine introduction: March 7 - April 3, 2021), phase 2 (stagnant period of vaccination: April 4 - April 22, 2021), and phase 3 (increase of vaccination: April 23 - May 5, 2021). NodeXL was used for data collection and analysis. Daily Twitter network data were collected by entering search terms highly related to the COVID-19 vaccine. This study found that side effects-related opinions were repeatedly formed throughout the analysis period. As the vaccination rate increased and death cases were reported on media, death-related issues also emerged on Twitter. On the other hand, vaccine safety did not receive much attention on Twitter. The results of this study highlight the role of social media as a channel of issue diffusion when a national disaster strikes. We emphasize the need for the government to monitor public opinions on social media and reflect them in crisis communication strategies.

• Journal of Internet Computing and Services
• /
• v.14 no.6
• /
• pp.49-57
• /
• 2013

As 'Big data' has been one of challenging issues, development of new services using Social Network Service (SNS) which is its typical example became active. SNS has developed as a media where everyone communicates at real time and the number of SNS opinion analyzing services is increasing. Meanwhile, new approach to acquire and analyze twitter data becomes necessary in TV advertisement system. This paper proposes LiveAD system, which store and analyze big data such as twitter data as well as analyze TV advertising effect based on twitter data. As a proof of concept, the proposed system has been implemented collecting and analyzing twitter data using Hadoop. The result of collected information over the system increases the chance of analyzing TV advertising effect on twitter in real-time.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.8
• /
• pp.377-384
• /
• 2016

Our goal is to build the system which collects tweets from Twitter, analyzes the sentiment of each tweet, and helps users build a sentiment tagged corpus semi-automatically. After collecting tweets with the Twitter API, we analyzes the sentiments of them with a sentiment dictionary. With the proposed system, users can verify the results of the system and can insert new sentimental words or dependency relations where sentiment information exist. Sentiment information is tagged with the JSON structure which is useful for building or accessing the corpus. With a test set, the system shows about 76% on the accuracy in analysing the sentiments of sentences as positive, neutral, or negative.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.311-314
• /
• 2018

최근 '미투운동'이 활발히 진행되면서 새로운 페미니즘의 물결을 맞이하였다. 이전의 페미니즘 운동과의 차이점은 SNS 를 통해 익명으로 활동하며 전파속도가 굉장히 빠르다는 것이다. 본 연구는 미투운동의 이러한 특성을 고려하여 실제 트위터 데이터에서 주요 키워드를 파악하고, 해당 키워드의 연관성 및 네트워크 분석으로 사회적 맥락을 알아본다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1636-1637
• /
• 2015

소셜 네트워크 서비스(SNS)에서는 사용자들의 친한 관계를 나타내는 여러 가지 특징을 발견할 수 있다. 본 논문에서는 트위터 이고-네트워크(ego-network) 데이터를 이용한 분석 실험을 통해 유저 간 친밀한 정도를 나타내는 여러 특징들과 관심사 유사도의 상관관계를 밝힌다.

• Journal of Korea Multimedia Society
• /
• v.16 no.6
• /
• pp.667-677
• /
• 2013

With the huge success of Social Network Services, studies on social network analysis to extract the current issues or to track the symptoms of epidemic disease are being carried out actively. On Twitter, tweets reflect people's reaction to an event and users' individual status well, so it is possible to detect an event regarding a tweet as a sensory value. Recently, social signals are used to detect the spread of illness like the flu as well as the occurrence of disaster event like an earthquake in early stages. In this paper, we set up a cold as a target event and regarded tweets as Cold Signals. To evaluate the reliability of Cold Signals, we analyzed correlations between weather factors and the cold index provided by Korea Meteorological Administration.

• Journal of Korean Society of Archives and Records Management
• /
• v.19 no.1
• /
• pp.167-194
• /
• 2019

This study aims to examine the change of awareness on presidential archives and records before and after impeachment by analyzing user frames. To achieve the goal of this study, prior studies of frame analysis were reviewed, and tweets of presidential archives and records before and after the impeachment were collected. This study conducted an analysis of Twitter and news extracted from Twitter using user frames and determined the differences between each frame over time. Afterward, five frames were set up to be used for the research through prior research and Twitter network analysis; changes in frames over time were examined by analyzing Twitter and news extracted from Twitter. Through such frame analysis, changes in the frame of presidential archives and records before and after the impeachment were examined, changes in public perception of presidential archives and records were identified, and areas of interest were determined. This study is significant as it identified changes in the public perception of presidential archives and records as well as in the areas of interest for the general public.

• Journal of Digital Convergence
• /
• v.10 no.6
• /
• pp.83-92
• /
• 2012

The purpose of this current study is to examine CEO image presented in messages of CEO twitter and perceived by CEO twit followers. To conduct this study, we selected two CEOs, ChanJin Lee and HyeonMyeong Pyo who are top ranking based on a number of followers in Korea. We investigated each CEO's Tweet, RT, Reply, and RT+Reply using three CEO image factors: CEO personality, CEO quality, and CEO outward characteristics. Also, followers were asked what image mostly they have based on those three image factors. The results showed that for ChanJin Lee, CEO quality were more presented in twitter than CEO personality and CEO outward characteristics, and CEO Lee's followers also perceived that Lee has more CEO quality image than others. Pyo, HyeonMyeong has CEO quality image in twitter than others, and his followers thought that he had more CEO outward characteristics image than others.