• Knowledge Management Research
• /
• v.16 no.1
• /
• pp.225-242
• /
• 2015

Cyberbullying, i.e., posting malicious comments online, has been identified as a critical issue in the online and social media context. It has become prevalent on a global scale, which happens across all ages. As a way to reduce and prevent cyberbullying, it is important to promote online prosocial behavior. In line with the concept of online prosocial behavior, we suggest posting benevolent comments against posting malicious comments as a new type of online prosocial behavior, which can combat cyberbullying and facilitate positive online culture. This study thus aims to analyze what motivates people to post benevolent comments in the online context. Based on interview methods, we extracted seven driving factors (self-presentation, pleasure, social contribution, emotional support, reputation, monetary reward, and reciprocity) and two inhibiting factors (social anxiety and effort) of posting benevolent comments online. This study has its theoretical contribution in exploring the motivation factors leading to the posting of benevolent comments by extending the concept of online prosocial behavior. It also has its practical implications by providing guidance for promoting prosocial behavior in the online context.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.35-45
• /
• 2019

Recently, the damage with social cost of malicious comments is increasing. In addition to the news of talent committing suicide through the effects of malicious comments. The damage to malicious comments including abusive language and slang is increasing and spreading in various type and forms throughout society. In this paper, we propose a technique for detecting abusive language using a bi-directional long short-term memory neural network model. We collected comments on the web through the web crawler and processed the stopwords on unused words such as English Alphabet or special characters. For the stopwords processed comments, the bidirectional long short-term memory neural network model considering the front word and back word of sentences was used to determine and detect abusive language. In order to use the bi-directional long short-term memory neural network, the detected comments were subjected to morphological analysis and vectorization, and each word was labeled with abusive language. Experimental results showed a performance of 88.79% for a total of 9,288 comments screened and collected.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.641-651
• /
• 2023

Malicious hate speech and gender bias comments are common in online communities, causing social problems in our society. Gender bias and hate speech detection has been investigated. However, it is difficult because there are diverse ways to express them in words. To solve this problem, we attempted to detect malicious comments in a Korean hate speech dataset constructed in 2020. We explored bidirectional encoder representations from transformers (BERT)-based deep learning models utilizing hyperparameter tuning, data sampling, and logits ensembles with a label distribution. We evaluated our model in Kaggle competitions for gender bias, general bias, and hate speech detection. For gender bias detection, an F1-score of 0.7711 was achieved using an ensemble of the Soongsil-BERT and KcELECTRA models. The general bias task included the gender bias task, and the ensemble model achieved the best F1-score of 0.7166.

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.71-79
• /
• 2021

One of the social aspects that changes as the use of the Internet becomes widespread is communication in online space. In the past, only one-on-one conversations were possible remotely, except when they were physically in the same space, but nowadays, technology has been developed to enable communication with a large number of people remotely through bulletin boards, communities, and social network services. Due to the development of such information and communication networks, life becomes more convenient, and at the same time, the damage caused by rapid information exchange is also constantly increasing. Recently, cyber crimes such as sending sexual messages or personal attacks to certain people with recognition on the Internet, such as not only entertainers but also influencers, have occurred, and some of those exposed to these cybercrime have committed suicide. In this paper, in order to reduce the damage caused by malicious comments, research a method for improving the performance of discriminate malicious comments through feature extraction based on parts-of-speech.

• Journal of Convergence for Information Technology
• /
• v.9 no.10
• /
• pp.81-86
• /
• 2019

Problems related to information overload and harmful information have already expanded to national social problems as well as personal problems. This study explores the causes of Internet addiction, exposure to harmful information, malicious comments, fake information/information manipulation, and new alternatives that have recently been felt as social problems. Assuming that existing technologies/policies were not applied effectively, psychological cause analysis was performed for the fundamental problem approach. As a result, internal problems such as obsession with knowledge/understanding of wrong information/black and white stereotypes and prejudice were analyzed as main causes. Each proposed solution aims to help improve national technology/policy regarding internet addiction and blocking harmful information.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.3
• /
• pp.103-110
• /
• 2023

With the spread of the Internet, anonymous communities emerged along with the activation of communities for communication between people, and many users are doing harm to others, such as posting aggressive posts and leaving comments using anonymity. In the past, administrators directly checked posts and comments, then deleted and blocked them, but as the number of community users increased, they reached a level that managers could not continue to monitor. Initially, word filtering techniques were used to prevent malicious writing from being posted in a form that could not post or comment if a specific word was included, but they avoided filtering in a bypassed form, such as using similar words. As a way to solve this problem, deep learning was used to monitor posts posted by users in real-time, but recently, the community uses words that can only be understood by the community or from a human perspective, not from a general Korean word. There are various types and forms of characters, making it difficult to learn everything in the artificial intelligence model. Therefore, in this paper, we proposes a preprocessing technique in which each character of a sentence is imaged using a CNN model that learns the consonants, vowel and spacing images of Korean word and converts characters that can only be understood from a human perspective into characters predicted by the CNN model. As a result of the experiment, it was confirmed that the performance of the LSTM, BiLSTM and CNN-BiLSTM models increased by 3.2%, 3.3%, and 4.88%, respectively, through the proposed preprocessing technique.

• The Journal of the Korea Contents Association
• /
• v.15 no.4
• /
• pp.391-406
• /
• 2015

As the individual participation in online news sites proliferates, the importance of online news comments has been increasing. Social comment services which help people leave comments on news articles using their own SNS (social networking site) accounts have gained popularity recently. Using data gathered from an online news site, this study, therefore, (1) identifies factors differentiating social comments from general comments, (2) examines how social comments are significantly different from general comments in terms of each factor, (3) and further validates how the social comments' characteristics vary among different type of SNS. Then, we investigated this study by applying t-test, ANOVA, and Duncan test of SPSS Statistics. Our results provide insights on the significant differences in all the factors between general and social comments. We also found that there is a significant difference between Facebook and Twitter groups among three types of SNS. The findings of this study would help assess the actual benefit of social comment services as they may provide us with several valuable leads to solve the malicious comments issue. Moreover, they would suggest the need to apply this service to other areas, such as online environments in private and public sectors.

• 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.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.6
• /
• pp.1491-1498
• /
• 2017

As the use of smartphones and the launch of various apps have increased rapidly, the number of malicious apps has also increased, and the damage is continuing. The Google Market where Android apps are registered is inevitably present at the same time as normal apps and malicious apps even though there are regulations for app registration. Especially, as social networks are activated, users are connected with social networks, and the ratings, downloads and awareness information are reflected in the number of downloaded apps. As a result, when users choose their apps by simply reflecting ratings, popularity, popular comments, and highly-categorized apps, malicious app downloads can sometimes cause significant harm. Therefore, this study first analyzed the tendency of malicious apps by directly crawling and analyzing long-term social information in the currently active Android market.

• ETRI Journal
• /
• v.35 no.1
• /
• pp.170-172
• /
• 2013

With the rapid progress of RFID security technologies, the international standard group ISO/IEC JTC 1/SC 31 is developing a few security technologies for RFID systems. One of the initial proposals is ISO/IEC working draft (WD) 29167-6. Recently, Song and others stated that Protocol 1 of ISO/IEC WD 29167-6 is vulnerable to a malicious adversary. However, their analysis comes from a misunderstanding regarding a communication parameter called Handle. In this letter, we point out that an adversary cannot obtain any sensitive information from intervening in Protocol 1.