• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.576-583
• /
• 2020

This paper proposes a technique to determine the spam comments on YouTube, which have recently seen tremendous growth. On YouTube, the spammers appeared to promote their channels or videos in popular videos or leave comments unrelated to the video, as it is possible to monetize through advertising. YouTube is running and operating its own spam blocking system, but still has failed to block them properly and efficiently. Therefore, we examined related studies on YouTube spam comment screening and conducted classification experiments with six different machine learning techniques (Decision tree, Logistic regression, Bernoulli Naive Bayes, Random Forest, Support vector machine with linear kernel, Support vector machine with Gaussian kernel) and ensemble model combining these techniques in the comment data from popular music videos - Psy, Katy Perry, LMFAO, Eminem and Shakira.

• JOURNAL OF ELECTRICAL WORLD
• /
• no.2 s.146
• /
• pp.2-10
• /
• 1989

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.17 no.1
• /
• pp.44-49
• /
• 2003

The textual comments on Shanghanlun and Jinguiyaolue were found in Hyangyakjipsung-bang, the representative medical book in the early period of Choson Dynasty. In all 57 chapters of the book, 17 chapters are related to those comments, and only one comment is quoted from all chapters except the chapter of 'Shanghanlun' and 'Jinguiyaolue'. As classified the comments by citation order, Jinguifang had 14 comments, Zhangzhongjing had 7 comments, Zhangzhongjing had 4 comments, and Jinguiyuhan had 1 comment. Comparing to the present version, 16 comments were qouted from Jinguiyaolue and 7 comments were quoted from Shanghanlun and 1 comment was quoted from Jinguiyuhanjing, but the source of 2 comments were not identified. Especially the 1 comment from Jinguiyuhanjing not only shows the importing date of the book into Korea, but also proofs the importance of the book which can refute the supposed source of the book as a reprint by Chenshijie in China. This results showed that Zhangzhongjing's books, which has imported before the early period of Chosun Dynasty, had an influence on Korean Medicine. As a result, further research on the medical books in the early period of Chosun Dynasty excepting Hyangyakjipsung-bang will be necessary.

• The Journal of the Korea Contents Association
• /
• v.9 no.7
• /
• pp.45-56
• /
• 2009

There are many internet users who collect the public opinions and express their opinions for internet news or blog articles through the replying comment on online community. But, it is hard to search and explore useful messages on web blogs since most of web blog systems show articles and their comments to the form of sequential list. Also, spam and malicious comments have become social problems as the internet users increase. In this paper, we propose a clustering and visualizing system for responding comments on large-scale weblogs, namely 'Daum AGORA,' using similarity analysis. Our system shows the comment clustering result as a simple screen view. Our system also detects spam comments using Needleman-Wunsch algorithm that is a well-known algorithm in bioinformatics.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.10-22
• /
• 2022

This study analyzed the effects of YouTube comments on belated hit in music chart by focusing on VIDITOR's YouTube channel video called 'BraveGirs_Rollin_comment_Collection'. For this, first of all, it investigated how the comments used by the channel master could achieve storytelling. Moreover, the reason for its popularity hit was researched by analyzing the comments made by users on this video through the 'Tipping point' rules. As a result, it was seen that this video created storytelling by connecting Brave Girls to the story of the Korean army. At the 'Tipping point,' besides, it was revealed that 'the law of the few' is the military, and 'the stickiness factor' is the good image left through Korea's island, 'Baengnyeongdo,' performance. Finally, the rule of 'the power of context' was indicated that it is a difficult reality situation because of COVID-19 and the good situation due to direct communication through the comments of Brave Girls' members and the company.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.61-79
• /
• 2022

This study investigated the effect of cognitive dissonance that people experience during online communication on face-to-face communication. When people communicate with others on Interne using the comment system, they know that all the people including themselves equally participate in the discussion as one of the many commenters. At the same time, they somewhat distrust other commenters' attitudes because of the Internet anonymity. We name this seemingly contradicting beliefs as cognitive dissonance and examine how these beliefs affect the intention to communicate in face-to-face. Also, the proposed research model includes other factors such as curiosity and the differences in attitudes between online and offline. To verify the hypotheses, a total of 323 comment system users were recruited and show that most of the hypotheses were supported. This study emphasized its implications by examining the reasons when and why people prefer direct communication rather than comment based communication.

• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.23-28
• /
• 2020

A comment system is essential for communication on the Internet. However, there are also malicious comments such as inappropriate expression of others by exploiting anonymity online. In order to protect users from malicious comments, classification of malicious / normal comments is necessary, and this can be implemented as text classification. Text classification is one of the important topics in natural language processing, and studies using pre-trained models such as BERT and graph structures such as GCN and GAT have been actively conducted. In this study, we implemented a comment classification system using BERT, GCN, and GAT for actual published comments and compared the performance. In this study, the system using the graph-based model showed higher performance than the BERT.

• Journal of Internet Computing and Services
• /
• v.21 no.1
• /
• pp.119-125
• /
• 2020

Based on an online experiment, this research examined how uncivil expressions made by participants from the same political partisan group (in-group) influenced the emotional and behavioral intentions of other in-group members, especially when the incivility was supported by social recommendations such as "recommendations." As predicted, results showed that a higher level of vicarious shame was felt when participants perceived higher levels of incivility. However, no significant effects of social recommendations were found regarding levels of vicarious shame. That is, the level of shame was not significantly different between participants who were exposed to an in-group uncivil comment that received recommendations and participants who were exposed to in-group uncivil comment without recommendations. Findings further found two types of coping strategies -situation-reparation and situation-avoidance - among participants exposed to in-group uncivil comments. Yet no significant effects were found regarding coping strategies in response to the presence of social recommendations. Participants' feelings of shame were positively correlated with both types of coping strategies, supporting findings of previous studies. Implications of this study are further discussed.

• East Asian mathematical journal
• /
• v.36 no.4
• /
• pp.475-492
• /
• 2020

This study is to establish the items on class critiques based on the mathematical competencies according to the mathematics curriculum revised in 2015. Namely, this study deals with the items on how pre-service or in-service teachers understand and comment on mathematics instruction on their own instruction or peers' instruction. To accomplish this, first of all the draft items on instructional reviews was developed by researchers of this study on the basis of the previous study(Hwang, 2018). In order to revise and develop the draft items, the experimental study was executed. The experimental study was done by the subject of 11 groups who are undergraduate students in the educational college of C University. The subject was supposed to watch an in-service middle school mathematics teacher's excellent instruction(study) video and to comment the instruction video on the draft items on class critiques. While analyzing the comments of the subject, the revised items on the class critiques were to be develop. Based on this study, from now on, the final and ideal items on the class critiques would be establish to reflect and comment teachers' instruction.

• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.329-342
• /
• 2020

In recent years, emotional text classification is one of the essential research contents in the field of natural language processing. It has been widely used in the sentiment analysis of commodities like hotels, and other commentary corpus. This paper proposes an improved W-LDA (weighted latent Dirichlet allocation) topic model to improve the shortcomings of traditional LDA topic models. In the process of the topic of word sampling and its word distribution expectation calculation of the Gibbs of the W-LDA topic model. An average weighted value is adopted to avoid topic-related words from being submerged by high-frequency words, to improve the distinction of the topic. It further integrates the highest classification of the algorithm of support vector machine based on the extracted high-quality document-topic distribution and topic-word vectors. Finally, an efficient integration method is constructed for the analysis and extraction of emotional words, topic distribution calculations, and sentiment classification. Through tests on real teaching evaluation data and test set of public comment set, the results show that the method proposed in the paper has distinct advantages compared with other two typical algorithms in terms of subject differentiation, classification precision, and F1-measure.