• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.2
• /
• pp.379-384
• /
• 2021

With people's continuous research and exploration in the field of artificial intelligence, more relatively mature artificial intelligence technology is applied in people's daily life. Mirrors are the most commonly used daily necessities in life, and it should be applied to artificial intelligence. The research results of this paper show that the smart mirror designed based on the raspberry pi displays weather, temperature, greetings, and has a human-mirror interaction function. The research method of this paper uses the Raspberry pi 3B + as the core controller and Google Assistant as the intelligent control. When connected to the network via Raspberry Pi's own WiFi, the mirror can automatically display and update time, weather and news information features. You can wake up the Google Assistant using keywords, then control the mirror to play music, remind the time, It implements the function of smart mirror voice interaction. Also, all the hardware used in this study is modular assembly. Later, it is convenient for user to assemble by himself later. It is suitable for market promotion at an affordable price.

• Fashion & Textile Research Journal
• /
• v.21 no.2
• /
• pp.179-188
• /
• 2019

Domestic apparel products labeled as 'Made in Korea' in the Chinese market are recognized as a high quality products due to the influence of the Korean Wave (Intergen Consulting Group, 2007). This study analyzes the patterns and order arrangement types of a fashion brand company commissioned to produce apparel in Seoul, Korea in order to rebuild a network of small sewing factories scattered in Korea, reorganize operations, and to find the possibility of regenerating the Korean sewing industry by establishing contact points with domestic sewing factories. We surveyed 100 apparel brand companies in Seoul listed in the 2014/2015 Korea Fashion Brand Annual (Apparel News, 2014) and conducted a questionnaire survey on the company's general management status, type of fabric materials dealt with, and major contact points and methods of production handling. The frequency analysis indicated that the main production material with cloth type was woven fabric with ladies' clothes. The Planning MD team has the highest rate of ordering production with delivery method to the production factory after purchasing fabric and trims. Most respondents answered that they would select a production factory based on recommendations from acquaintances. This was due to a lack of no objective indicator provided by the sewing factory at present and the absence of objectively proceeded communication with brand companies. In this study, we analyze various conditions and measurements for production arrangements from a fashion brand company to revitalize sewing factories in Korea.

• Knowledge Management Research
• /
• v.20 no.3
• /
• pp.155-171
• /
• 2019

In the entertainment industry which has great uncertainty, it is essential to predict public preference first. Thanks to various mass media channels such as cable TV and internet-based streaming services, the reality audition program has been getting big attention every day and it is being used as a new window to new entertainers' debut. This phenomenon means that it is changing from a closed selection process to an open selection process, which delegates selection rights to the public. This is characterized by the popularity of the public being reflected in the selection process. Therefore, this study aims to implement a machine learning model which predicts the winner of , which has recently been popular in South Korea. By doing so, this study is to extend the research method in the cultural industry and to suggest practical implications. We collected the data of winners from the 1st, 2nd, and 3rd seasons of the Produce 101 and implemented the predictive model through the machine learning method with the accumulated data. We tried to develop the best predictive model that can predict winners of by using four machine learning methods such as Random Forest, Decision Tree, Support Vector Machine (SVM), and Neural Network. This study found that the audience voting and the amount of internet news articles on each participant were the main variables for predicting the winner and extended the discussion by analyzing the precision of prediction.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1053-1065
• /
• 2020

With the recent development of hardware performance and artificial intelligence technology, sophisticated fake videos that are difficult to distinguish with the human's eye are increasing. Face synthesis technology using artificial intelligence is called Deepfake, and anyone with a little programming skill and deep learning knowledge can produce sophisticated fake videos using Deepfake. A number of indiscriminate fake videos has been increased significantly, which may lead to problems such as privacy violations, fake news and fraud. Therefore, it is necessary to detect fake video clips that cannot be discriminated by a human eyes. Thus, in this paper, we propose a deep-fake detection model applied with Bidirectional Convolution LSTM and Attention Module. Unlike LSTM, which considers only the forward sequential procedure, the model proposed in this paper uses the reverse order procedure. The Attention Module is used with a Convolutional neural network model to use the characteristics of each frame for extraction. Experiments have shown that the model proposed has 93.5% accuracy and AUC is up to 50% higher than the results of pre-existing studies.

• Journal of Information Technology Services
• /
• v.19 no.6
• /
• pp.1-13
• /
• 2020

Recently, 80% of big data consists of unstructured text data. In particular, various types of documents are stored in the form of large-scale unstructured documents through social network services (SNS), blogs, news, etc., and the importance of unstructured data is highlighted. As the possibility of using unstructured data increases, various analysis techniques such as text mining have recently appeared. Therefore, in this study, topic modeling technique was applied to the Korea Highway Corporation's voice of customer (VOC) data that includes customer opinions and complaints. Currently, VOC data is divided into the business areas of Korea Expressway Corporation. However, the classified categories are often not accurate, and the ambiguous ones are classified as "other". Therefore, in order to use VOC data for efficient service improvement and the like, a more systematic and efficient classification method of VOC data is required. To this end, this study proposed two approaches, including method using only the latent dirichlet allocation (LDA), the most representative topic modeling technique, and a new method combining the LDA and the word embedding technique, Word2vec. As a result, it was confirmed that the categories of VOC data are relatively well classified when using the new method. Through these results, it is judged that it will be possible to derive the implications of the Korea Expressway Corporation and utilize it for service improvement.

• The Journal of the Korea Contents Association
• /
• v.21 no.5
• /
• pp.139-148
• /
• 2021

The study examines public attitude and reaction regarding administrative integration in Daegu and Gyeongbuk area. Specifically, it employs social big data including textual comments on online news articles and YouTube video clips. The collected data are analyzed in order to compare two periods, that is, before and after the inauguration of the Public Opinion Committee for One Daegu-Gyeongbuk. As a result, we have found that people's favorable response to administrative integration has gradually increased since the launch of the Committee. However, it still lacks specific administrative procedures and discussion topics among the frequently used words in the collected data. Thus, the Committee needs to provide a variety of information and materials related to administrative integration.

• Journal of Information Science Theory and Practice
• /
• v.11 no.1
• /
• pp.14-30
• /
• 2023

The purpose of this study is to compare research trends in the information and communication technology (ICT) field between North and South Korea and analyze the differences by using data mining. Frequency analysis, clustering, and network analysis were performed using keywords from seven South Korean and two North Korean ICT academic journals published for five years (2015-2019). In the case of South Korea (S. Korea), the frequency of research on image processing and wireless communication was high at 16.7% and 16.3%, respectively. North Korea (N. Korea) had a high frequency of research, in the order of 18.2% for image processing, 16.9% for computer/Internet applications/security, and 16.4% for industrial technology. N. Korea's natural language processing (NLP) sector was 11.9%, far higher than S. Korea's 0.7 percent. Student education is a unique subject that is not clustered in S. Korea. In order to promote exchanges between the two Koreas in the ICT field, the following specific policies are proposed. Joint research will be easily possible in the image processing sector, with the highest research rate in both Koreas. Technical cooperation of medical images is required. If S. Korea's high-quality image source is provided free of charge to N. Korea, research materials can be enriched. In the field of NLP, it calls for proposing exchanges such as holding a Korean language information conference, developing a Korean computer operating system. The field of student education encourages support for remote education contents and management know-how, as well as joint research on student remote evaluation.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.113-119
• /
• 2023

Despite the lifting of the mandatory wearing of masks, which was the main tool of the COVID-19 quarantine policy, we paid attention to the fact that some people are still wearing masks, and we wanted to clarify why people do not take off their masks. Through a survey in this regard, we were able to ascertain why some people continue to wear masks in a broader context. In this article, we directly and indirectly confirm the hidden side of citizens' continued wearing of masks by analyzing how the lifting of the mask-wearing obligation was reported in media articles that have a significant impact on citizens' behavior and attitude. Through this, it was confirmed that citizens continue to wear masks to protect themselves in an uncertain situation where the COVID-19 endemic has not been declared, despite the quarantine authorities' announcement of lifting the mandatory wearing. In a situation where crises such as COVID-19 are expected to repeat frequently in the future, it was concluded that it is important to build trust in the quarantine authorities.

• Smart Media Journal
• /
• v.12 no.5
• /
• pp.17-27
• /
• 2023

Recently, automatic text summarization, which automatically summarizes only meaningful information for users, is being studied steadily. Especially, research on text summarization using Transformer, an artificial neural network model, has been mainly conducted. Among various studies, the GSG method, which trains a model through sentence-by-sentence masking, has received the most attention. However, the traditional GSG has limitations in selecting a sentence to be masked based on the degree of overlap of tokens, not the meaning of a sentence. Therefore, in this study, in order to improve the quality of text summarization, we propose SbGSG (Semantic-based GSG) methodology that selects sentences to be masked by GSG considering the meaning of sentences. As a result of conducting an experiment using 370,000 news articles and 21,600 summaries and reports, it was confirmed that the proposed methodology, SbGSG, showed superior performance compared to the traditional GSG in terms of ROUGE and BERT Score.

• Asian Journal for Public Opinion Research
• /
• v.12 no.1
• /
• pp.53-74
• /
• 2024

The proliferation of health misinformation gained momentum amidst the outbreak of the novel coronavirus disease 2019 (COVID-19). People stuck in their homes, without work pressure, regardless of health concerns towards personal, family, or peer groups, consistently demanded information. People became engaged with misinformation while attempting to find health information content. This study used the content analysis method and analyzed 1,154 misinformation stories from four prominent signatories of the International Fact-Checking Network during the pandemic. The study finds the five main categories of misinformation related to the COVID-19 pandemic. These are 1) the severity of the virus, 2) cure, prevention, and treatment, 3) myths and rumors about vaccines, 4) health authorities' guidelines, and 5) personal and social impacts. Various sub-categories supported the content characteristics of these categories. The study also analyzed the emotional valence of health misinformation. It was found that misinformation containing negative sentiments got higher engagement during the pandemic. Positive and neutral sentiment misinformation has less reach. Surprise, fear, and anger/aggressive emotions highly affected people during the pandemic; in general, people and social media users warning people to safeguard themselves from COVID-19 and creating a confusing state were found as the primary motivation behind the propagation of misinformation. The present study offers valuable perspectives on the mechanisms underlying the spread of health-related misinformation amidst the COVID-19 outbreak. It highlights the significance of discerning the accuracy of information and the feelings it conveys in minimizing the adverse effects on the well-being of public health.