• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.191-203
• /
• 2022

In this paper, we developed a system that intelligently identifies skin image data from big data collected from social media Instagram and extracts standardized skin sample data for skin condition diagnosis and management. The system proposed in this paper consists of big data collection and analysis stage, skin image analysis stage, training data preparation stage, artificial neural network training stage, and skin image identification stage. In the big data collection and analysis stage, big data is collected from Instagram and image information for skin condition diagnosis and management is stored as an analysis result. In the skin image analysis stage, the evaluation and analysis results of the skin image are obtained using a traditional image processing technique. In the training data preparation stage, the training data were prepared by extracting the skin sample data from the skin image analysis result. And in the artificial neural network training stage, an artificial neural network AnnSampleSkin that intelligently predicts the skin image type using this training data was built up, and the model was completed through training. In the skin image identification step, skin samples are extracted from images collected from social media, and the image type prediction results of the trained artificial neural network AnnSampleSkin are integrated to intelligently identify the final skin image type. The skin image identification method proposed in this paper shows explain high skin image identification accuracy of about 92% or more, and can provide standardized skin sample image big data. The extracted skin sample set is expected to be used as standardized skin image data that is very efficient and useful for diagnosing and managing skin conditions.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.21-31
• /
• 2018

In Hyper-Connected Society, IoT-based Lifelog data is used throughout the Internet and is an important component of customized services that reflect user requirements. Also, Users are using social network services to easily express their interests and feelings, and various life log data are being accumulated. In this paper, Intelligent characters using IoT based lifelog data have been developed and qualitative/quantitative data are collected and analyzed in order to systematically grasp emotions of users. For this, qualitative data through the social network service used by the user and quantitative data through the wearable device are collected. The collected data is verified for reliability by comparison with the persona through esnography. In the future, more intelligent characters will be developed to collect more user life log data to ensure data reliability and reduce errors in the analysis process to provide personalized services.

• Science of Emotion and Sensibility
• /
• v.26 no.1
• /
• pp.101-114
• /
• 2023

This research aimed to recognize the determinant characteristics affecting consumers' clothing fit evaluation by employing text mining and social network analysis. For this aim, we first extracted text data linked to clothing fit from 2,000 consumer reviews collected from social network services and conducted semantic network examination and CONCOR analysis. As a result, we reported that "pants" and "skirts" were the most commonly associated clothing items with consumers' clothing fit evaluation. And the length of clothing was most commonly investigated. Then, the "waist" and "hip" were the most critical body parts affecting consumers' perception of clothing fit. Further, the four keywords including "wide," "large," "short," and "long" were the most employed ones in consumer reviews when evaluating clothing fit. This study is meaningful in that it specifically recognized the structural relationship and semantic meanings of keywords relevant to consumers' evaluation of clothing fit, which could bring empirical reference information for advanced clothing fit.

• The Journal of the Korea Contents Association
• /
• v.20 no.11
• /
• pp.110-118
• /
• 2020

Social media data serves as a proxy indicator to understand the problems and the future of public opinion in Korean society. This research used 109,015 news data from 2016 to 2018 to analyze the sensitivity of the elderly and employment in Korean society, and explored the possibility of expanding the labor force in Korean society, which is facing a cliff between the elderly and the population. Topic keywords for employment of the elderly include "elderly*employment", "elderly*employment", and "elderly*wage". As a result of the analysis, positive sensitivity prevails for most of the period, and it is possible to expand the working-age population. Positive feelings about expanding employment opportunities for the elderly and negative feelings about low wages have brought to light the reality of the elderly who are still poor despite their work. In this study, social big data was used to analyze the perceptions and sensibilities of Korean society related to the elderly and employment through hierarchical crowd analysis and related text mining analysis.

• 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.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.23-34
• /
• 2019

Recently, the crime that utilizes the digital platform is continuously increasing. About 140,000 cases occurred in 2015 and about 150,000 cases occurred in 2016. Therefore, it is considered that there is a limit handling those online crimes by old-fashioned investigation techniques. Investigators' manual online search and cognitive investigation methods those are broadly used today are not enough to proactively cope with rapid changing civil crimes. In addition, the characteristics of the content that is posted to unspecified users of social media makes investigations more difficult. This study suggests the site-based collection and the Open API among the content web collection methods considering the characteristics of the online media where the infringement crimes occur. Since illegal content is published and deleted quickly, and new words and alterations are generated quickly and variously, it is difficult to recognize them quickly by dictionary-based morphological analysis registered manually. In order to solve this problem, we propose a tokenizing method in the existing dictionary-based morphological analysis through WPM (Word Piece Model), which is a data preprocessing method for quick recognizing and responding to illegal contents posting online infringement crimes. In the analysis of data, the optimal precision is verified through the Vote-based ensemble method by utilizing a classification learning model based on supervised learning for the investigation of illegal contents. This study utilizes a sorting algorithm model centering on illegal multilevel business cases to proactively recognize crimes invading the public economy, and presents an empirical study to effectively deal with social data collection and content investigation.

• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1136-1144
• /
• 2014

Twitter is a social network service that generates about 140 million contents a day. Contents of Twitter contain a variety of information and many researchers research those in various fields. In this research, we propose a method for evaluating the importance of content based on characteristics of Twitter. We have found that number of follower means user's popularity and Re-tweet that means the popularity of content. We perform experiments about proposed method using real Twitter data for proving effectiveness of proposed method. Also, we found information providers in Twitter are public user who represent a company or a representative of a specific group.

• The Journal of the Korea Contents Association
• /
• v.20 no.6
• /
• pp.460-469
• /
• 2020

This study analyzes online news articles and cafe articles on the '5-day Rotation Mask Distribution System', which is emerging as a recent issue due to the COVID-19 incident, to identify the mass media and social media agendas containing media and public reactions. This study figured out the difference between mass media and social media. For analysis, we collected 2,096 full text articles from Naver and 1,840 posts from Naver Cafe, and conducted word frequency analysis, word cloud, and LDA topic modeling analysis through data preprocessing and refinement. As a result of analysis, social media showed real-life topics such as 'family members' purchase', 'the postponement of school opening', ' mask usage', and 'mask purchase', reflecting the characteristics of personal media. Social media was found to play a role of exchanging personal opinions, emotions, and information rather than delivering information. With the application of the research method applied to this study, social issues can be publicized through various media analysis and used as a reference in the process of establishing a policy agenda that evolves into a government agenda.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.577-580
• /
• 2018

지난 몇 년간 우리는 소셜 검색에 몰두하여 연관검색 및 소비자의 만족을 위해 빅데이터 분석을 하였다. 최근에는 빅데이터 분석이라는 흐름에 맞춰 기업 및 기관별 본연의 정보를 통합하여 효율적인 검색을 할 수 있도록 하는 솔루션을 대거 도입하고 있다. 또한 기업 및 기관에서 가지고 있는 정보는 기존 비정형 데이터로 방대하여 기존의 방법이나 도구로 수집 및 저장 분석이 어려운 실정이다. 이에 공공기관 및 민간기업 등에서는 키워드 중심의 다양한 검색엔진을 개발하거나 도입하고 있으며, 정보 분류의 확대, 메타데이터의 활용, 태그정보의 제공, 개인 맞춤형 서비스 등 고객의 만족도를 제고하기 위한 다양한 방법을 시도하고 있다. 본 연구에서는 기관의 교통 연구와 관련한 일련의 작업 중 행정문서, 연구정보, 유관기관 게시물 등의 통합 빅데이터를 가지고 검색시스템을 구현하였다. 이와 더불어 사용자 사전 및 동의어 사전을 통한 검색 키워드를 데이터베이스에 저장하여 검색 효율성을 제고하는 방안을 제시한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.276-277
• /
• 2010

Web2.0의 고도화와 소셜미디어의 진화에 따라 웹 기반 콘텐츠에 대한 다양한 활용이 시도되고 있다. 또한, IPTV 부분에서는 기존의 TV기반에서 Web기반, Mobile기반 서비스로 확산되고 있다. 이러한 멀티디바이스 환경과 웹기반 콘텐츠의 IPTV 활용 서비스를 위해서는 콘텐츠에 대한 정의와 이를 표현하기 위한 메타데이터에 대한 정의가 필요하다. 이를 위해서 관련 표준화 단체를 분석한 후 이들이 제공하는 콘텐츠 메타데이터의 연관관계를 비교 분석하였다. IPTV 서비스를 위한 콘텐츠 메타데이터는 TV AnyTime Phase1 을 바탕으로 각 표준기구의 네트워크 상황과 서비스 방식에 따라 선택적으로 적용하고 있으며, VOD 서비스를 위해서는 CableLab의 ADI 메타데이터를 적용하고 있다. 향후 멀티디바이스 환경에서 IPTV 서비스를 위해서는 메타데이터에 대한 확장 연구가 필요하다.