• Journal of Internet Computing and Services
• /
• v.20 no.1
• /
• pp.1-10
• /
• 2019

Since big-data text mining extracts many features and data, clustering and classification can result in high computational complexity and low reliability of the analysis results. In particular, a term document matrix obtained through text mining represents term-document features, but produces a sparse matrix. We designed an advanced genetic algorithm (GA) to extract features in text mining for detection model. Term frequency inverse document frequency (TF-IDF) is used to reflect the document-term relationships in feature extraction. Through a repetitive process, a predetermined number of features are selected. And, we used the sparsity score to improve the performance of detection model. If a spam mail data set has the high sparsity, detection model have low performance and is difficult to search the optimization detection model. In addition, we find a low sparsity model that have also high TF-IDF score by using s(F) where the numerator in fitness function. We also verified its performance by applying the proposed algorithm to text classification. As a result, we have found that our algorithm shows higher performance (speed and accuracy) in attack mail classification.

• Database Research
• /
• v.34 no.3
• /
• pp.3-13
• /
• 2018

Text classification is one of the text mining technologies that classifies a given textual document into its appropriate categories and is used in various fields such as spam email detection, news classification, question answering, emotional analysis, and chat bot. In general, the text classification system utilizes machine learning algorithms, and among a number of algorithms, naïve Bayes and support vector machine, which are suitable for text data, are known to have reasonable performance. Recently, with the development of deep learning technology, several researches on applying deep neural networks such as recurrent neural networks (RNN) and convolutional neural networks (CNN) have been introduced to improve the performance of text classification system. However, the current text classification techniques have not yet reached the perfect level of text classification. This paper focuses on the fact that the text data is expressed as a vector only with the word dimensions, which impairs the semantic information inherent in the text, and proposes a neural network architecture based upon the semantic tensor space model.

• Journal of Platform Technology
• /
• v.9 no.2
• /
• pp.26-37
• /
• 2021

E-mail is one of the important tools for communicating with people in everyday life. With COVID-19 (Coronavirus) increasing non-face-to-face activity, security incidents through e-mail such as spam, phishing, and ransomware are increasing. E-mail security incidents are increasing as social engineering attack using human psychology rather than arising from technological weaknesses that e-mails have. Security incidents using human psychology can be prevented and defended by improving security awareness. This study empirically studies the analysis of changes in response to malicious e-mail due to improved security awareness through malicious e-mail simulations on executives and employees of domestic and foreign company. In this study, the factors of security training, top-down security management, and security issue sharing are found to be effective in safely responding to malicious e-mail. This study presents a new study by conducting empirical analysis of theoretical research on security awareness in relation to malicious e-mail responses, and results obtained from simulations in a practical setting may help security work.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.2
• /
• pp.129-137
• /
• 2008

In this paper, we propose an improved Bayesian Filtering mechanism to reduce the False Positives that occurs in the existing Bayesian Filtering mechanism. In the existing Bayesian Filtering mechanism, the same Bayesian Filtering DB trained at the e-mail server is applied to each e-mail user. Also, the training method using receiving e-mails only could not provide the high quality of ham DB. Due to these problems, the existing Bayesian Filtering mechanism can produce the False Positives which misclassify the ham e-mails into the spam e-mails. In the proposed mechanism, the sending e-mails of the user are treated as the high quality of ham information, and are trained to the Bayesian ham DB automatically. In addition, by providing a different Bayesian DB to each e-mail user respectively, more efficient e-mail filtering service is possible. Our experiments show the improvement of filtering accuracy by 3.13%, compared to the existing Bayesian Filtering mechanism.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.6A
• /
• pp.107-114
• /
• 2008

The Bot is a kind of worm/virus that can be used to launch the distributed denial-of-service(DDoS) attacks or send massive amount of spam e-mails, etc. A lot of organizations make an effort to counter the Botnet's attacks. In Korea, we use DNS sinkhole system to protect from the Botnet's attack, while in Japan "so called" CCC(Cyber Clean Center) has been developed to protect from the Botnet's attacks. But in case of DNS sinkhole system, there is a problem since it cannot cure the Bot infected PCs themselves and in case of CCC there is a problem since only 30% of users with the Botnet-infected PCs can cooperate to cure themself. In this paper we propose a new method that prevent the Botnet's attacks and cure the Bot-infected PCs at the same time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.22 no.4
• /
• pp.761-770
• /
• 2012

CAPTCHAs(Completely Automated Public Turing tests to tell Computer and Human Apart) have been widely used for preventing the automated attacks such as spam mails, DDoS attacks, etc.. In the early stages, the text-based CAPTCHAs that were made by distorting random characters were mainly used for frustrating automated-bots. Many researches, however, showed that the text-based CAPTCHAs were breakable via AI or image processing techniques. Due to the reason, the image-based CAPTCHAs, which employ images instead of texts, have been considered and suggested. In many image-based CAPTCHAs, however, the huge number of source images are required to guarantee a fair level of security. In 2008, Kang et al. suggested a new image-based CAPTCHA that uses test images made by composing multiple source images, to reduce the number of source images while it guarantees the security level. In their paper, the authors showed the convenience of their CAPTCHA in use through the use study, but they did not verify its security level. In this paper, we verify the security of the image-based CAPTCHA suggested by Kang et al. by performing several attacks in various scenarios and consider other possible attacks that can happen in the real world.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.19 no.6
• /
• pp.105-112
• /
• 2009

Nowadays, completely automated public turing tests to tell computers and humans apart(CAPTCHAs) are widely used to prevent various attacks by automated software agents such as creating accounts, advertising, sending spam mails, and so on. In early CAPTCHAs, the characters were simply distorted, so that users could easily recognize the characters. From that reason, using various techniques such as image processing, artificial intelligence, etc., one could easily break many CAPTCHAs, either. As an alternative, By adding noise to CAPTCHAs and distorting the characters in CAPTCHAs, it made the attacks to CAPTCHA more difficult. Naturally, it also made users more difficult to read the characters in CAPTCHAs. To improve the readability of CAPTCHAs, some CAPTCHAs used different colors for the characters. However, the usage of the different colors gives advantages to the adversary who wants to break CAPTCHAs. In this paper, we suggest a method of increasing the recognition ratio of CAPTCHAs based on colors.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.19 no.4
• /
• pp.51-61
• /
• 2009

Recently, the cyber-attacks using botnets are being increased, Because these attacks pursue the money, the criminal aspect is also being increased, There are spreading of spam mail, DDoS(Distributed Denial of Service) attacks, propagations of malicious codes and malwares, phishings. leaks of sensitive informations as cyber-attacks that used botnets. There are many studies about detection and mitigation techniques against centralized botnets, namely IRC and HITP botnets. However, P2P botnets are still in an early stage of their studies. In this paper, we analyzed the traffics of the Peacomm bot that is one of P2P-based storm bot by using honeynet which is utilized in active analysis of network attacks. As a result, we could see that the Peacomm bot sends a large number of UDP packets to the zombies in wide network through P2P. Furthermore, we could know that the Peacomm bot makes the scale of botnet maintained and extended through these results. We expect that these results are used as a basis of detection and mitigation techniques against P2P botnets.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.231-238
• /
• 2022

In this paper, among short-range wireless communication technologies such as Beacon, Bluetooth, UWB (Ultra-wideband), ZigBee, NFC (Near Field Communication), Z-Wave, 6LoWPAN (IPv6 over Low power WPAN), D2D (Device to Device), etc., proposed an IoT service platform based on a beacon that can provide indoor positioning. And, a beacon-linked web server was designed by blocking indiscriminate beacon spam signals and applying REST web service technology with flexibility and scalability. Data accessibility between different devices was verified by testing the success rate of data transmission, the success rate of blocking beacon push, the success rate of IoT interlocking processing, the accuracy of location positioning, and the success rate of REST web service-based data processing. Through the designed IoT service platform, various proposals and research on short-distance-based business models and service platforms will be conducted in the future.

• Asia Marketing Journal
• /
• v.7 no.3
• /
• pp.121-141
• /
• 2005

Most of the people in marketing area know that database marketing has been one of the most powerful marketing tools and thus database marketing industry grows bigger and bigger. For both effective database marketing and database marketing industry, personal data are the very essential resources. Unfortunately, in Korea, both database marketing and database marketing industry stays far behind compared to other countries because it is practically very hard to legally trade personal data for database marketing purpose. Instead Korea has a illegal spam problem which might be a natural consequency of strong restriction on personal data in the situation of huge demand for personal data. KT SODiS can be called the frontier of Korea's database marketing industry since it is the first legal business in this area. In the first 5 months, SODiS obtained 2 millions of legal customer consents which can be the strong base to help database marketing activities of other companies. This case shows marketing strategies of KT SODiS to establish infrastructure for Korea's database marketing industry and suggests some future tasks to further develop the industry.