• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.305-318
• /
• 2022

Abusive behaviors have become a common issue in many online social media platforms. Profanity is common form of abusive behavior in online. Social media platforms operate the filtering system using popular profanity words lists, but this method has drawbacks that it can be bypassed using an altered form and it can detect normal sentences as profanity. Especially in Korean language, the syllable is composed of graphemes and words are composed of multiple syllables, it can be decomposed into graphemes without impairing the transmission of meaning, and the form of a profane word can be seen as a different meaning in a sentence. This work focuses on the problem of filtering system mis-detecting normal phrases with profane phrases. For that, we proposed the deep learning-based framework including grapheme and syllable separation-based word embedding and appropriate CNN structure. The proposed model was evaluated on the chatting contents from the one of the famous online games in South Korea and generated 90.4% accuracy.

• Journal of Creative Information Culture
• /
• v.7 no.2
• /
• pp.91-98
• /
• 2021

As the amount of time that elementary school students spend online increased due to Corona 19, the amount of posts, comments, and chats they write increased, and problems such as offending others' feelings or using swear words are occurring. Netiquette is being educated in elementary school, but training time is insufficient. In addition, it is difficult to expect changes in student behavior. So, technical support through natural language processing is needed. In this study, an experiment was conducted to automatically filter profanity sentences by applying them to a pre-trained language model on sentences written by elementary school students. In the experiment, chat details of elementary school 4-6 graders were collected on an online learning platform, and general sentences and profanity sentences were trained through a pre-learned language model. As a result of the experiment, as a result of classifying profanity sentences, it was analyzed that the precision was 75%. It has been shown that if the learning data is sufficiently supplemented, it can be sufficiently applied to the online platform used by elementary school students.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.181-188
• /
• 2024

Traditional profanity detection methods have limitations in identifying intentionally altered profanities. This paper introduces a new method based on Named Entity Recognition, a subfield of Natural Language Processing. We developed a profanity detection technique using sequence labeling, for which we constructed a dataset by labeling some profanities in Korean malicious comments and conducted experiments. Additionally, to enhance the model's performance, we augmented the dataset by labeling parts of a Korean hate speech dataset using one of the large language models, ChatGPT, and conducted training. During this process, we confirmed that filtering the dataset created by the large language model by humans alone could improve performance. This suggests that human oversight is still necessary in the dataset augmentation process.

• Journal of the Korean Society of School Health
• /
• v.31 no.2
• /
• pp.88-97
• /
• 2018

Purpose: The purpose of this study was to investigate the effects of a profanity intervention application on interpersonal relationship ability, aggression, and empathic ability among female middle school students. Method: The subjects of the present study were freshmen students at a girls' middle school in Gyeonggi-do. The profanity intervention application was implemented from November 20th to 27th, 2017. The pre- and post-tests were conducted one week before and two weeks after the intervention, respectively. The study was designed as a randomized controlled study and the collected data were analyzed using SPSS/WIN 23.0 (Chicago, IL). A homogeneity test between the two groups was performed using $x^2$ test, Fisher's exact test, independent t-test, and ANCOVA. Results: The two groups were found to be homogeneous in the pre-test. After the implementation of the profanity intervention application, aggression significantly decreased (F=20.17, p<.001) and interpersonal reactivity (F=5.92, p=.023) and empathy (F=11.10, p=.003) significantly increased in the experimental group compared to the control group. Among the factors related with aggression, verbal aggression, anger, and hostility significantly decreased in the experimental group. Meanwhile, among the factors of interpersonal reactivity, empathic concern and perspective-taking significantly increased and, among the factors of empathy, cognitive empathy significantly increased in the experimental group compared to the control group. Conclusion: Based on the findings of the present study, various intervention programs that can consistently improve the moral and ethical sensitivity of students should be presented starting from elementary school to prevent verbal violence in adolescents.

• International Journal of Advanced Culture Technology
• /
• v.10 no.4
• /
• pp.153-159
• /
• 2022

This research investigates the accessibility of Korean modern values at screens from the perspectives of North American audiences while consuming Korean films and TV dramas. In order to decode Korean cultural values at screens, this study collects the data of 266 Korean films and TV dramas distributed in North America from IMDb platform, unlocking that the success of box office for Korean projects in North America is positively correlated with the characteristics of Violence & Gore and Profanity. Meanwhile, the audiences' positive responses to the violent project mainly depends on the degree of "Sex and Nudity", the plots of "social groups". The current success of Korean cultural and creative industry as well as the future global markets strengths are analyzed in this research. Stuart Hall's cultural studies approach and cultural psychological account theory are applied to discuss the success of Korean entertainment industry in the transnational cultural communication.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.498-499
• /
• 2023

비속어 탐지 기법으로 주로 사용되는 비속어 데이터베이스 활용 방식 혹은 문장 자체를 혐오, 비혐오로 분류하는 방식은 변형된 비속어 탐지에 어려움이 있다. 본 논문에서는 자연어 처리 태스크 중 하나인 개체명 인식 방법에서 착안하여 시퀀스 레이블링 기반의 비속어 탐지 방법을 제안한다. 한국어 악성 댓글 중 비속어 부분에 대해 레이블링 된 데이터셋을 구축하여 실험을 진행하고, 이를 통해 F1-Score 약 0.88 의 결과를 보인다.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.113-120
• /
• 2009

Currently the verbal abuse in text message over on-line game is so serious. However we do not have any effective policy or technical tools yet. Till now in order to cope with this problem, the online game service providers have accumulated a set of forbidden words and applied this list on the textual word used in on-line game, which is called 'Swear filter'. But young on-line game players easily avoid this filtering method by coining another words which is not kept in the list. Especially Korean is very easy to make new variations of a vulgar word. In this paper, we propose one smart filtering algorithm to identify newly coined profanities. Important features of our method include the canonical form transformation of coined profanities, semi-global alignment between in the level of consonant and vowel units. For experiment, we have collected more than 1000 newly coined vulgar words in on-line gaming sites and tested these word against our methods. where our system have successfully filtered more than 90% of those newly coined vulgar words.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.998-1001
• /
• 2019

최근에 게임이나 채팅 프로그램 내에서의 비속어 필터링은 금칙어 기반으로 운영되고 있다. 하지만 금칙어 기반의 프로그램은 여러 한계점을 보이며, 따라서, 본 논문에서는 'Text-CNN'을 활용한 딥러닝 기법에 기반하여 비속어 필터링 프로그램을 제안한다. 데이터의 자질을 '자모' 단위로 전처리하여 학습시키고 어느 부분이 비속어인지 검출하여 마스킹 처리하는 'LIME 알고리즘'을 사용하여 우리의 프로그램을 이용하는 사용자들에게 바른 언어습관을 지향하며 더 나아가 올바른 인터넷 문화를 조성할 수 있도록 필터링 채팅 프로그램을 제안한다.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.1
• /
• pp.83-87
• /
• 2021

By the increase of internet usage, communicating online became an everyday thing. Thereby various people have experienced profanity by anonymous users. Nowadays lots of studies tried to solve this problem using artificial intelligence, but most of the solutions were for non-real time situations. In this paper, we propose a Telegram plugin that detects swear words using word2vec, and an algorithm to find the target of the sentence. We vectorized the input sentence to find connections with other similar words, then inputted the value to the pre-trained CNN (Convolutional Neural Network) model to detect any swears. For target recognition we proposed a sequential algorithm based on KoNLPY.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.47-52
• /
• 2012

Today the Internet is being used in all areas. However, part of everyday life, the lack of ethics on the Internet, the Internet lies in state. Especially the youth, should exploit the anonymity. As a result, has caused many problems. Fps online game from young people enjoy the use of profanity and slang, and increasing the damage of cyber violence has also increased. The damage type of cyber violence, abusive and slang, and the age, the lower the investigation was to increase the damage. Lot of publicity and public service advertising is in progress in order to prevent problems, but its expected effects is lacking. In addition, using abusive and slang words, This is a fundamental and provide a service to handle deleted or blinds, but if you can not see the solution. In this paper suggests the language convert that can be applied to Fps online games. Language the conversion process, if applicable to the game, the abusive and slang language, helps to purify. Proposed language the conversion process is expected to reduce cyber violence used abusive and slang.