• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.10
• /
• pp.437-446
• /
• 2022

Text classification task from Natural Language Processing (NLP) combined with state-of-the-art (SOTA) Machine Learning (ML) and Deep Learning (DL) algorithms as the core engine is widely used to detect and classify voice phishing call transcripts. While numerous studies on the classification of voice phishing call transcripts are being conducted and demonstrated good performances, with the increase of non-face-to-face financial transactions, there is still the need for improvement using the latest NLP technologies. This paper conducts a benchmarking of Korean voice phishing detection performances of the pre-trained Korean language model KoBERT, against multiple other SOTA algorithms based on the classification of related transcripts from the labeled Korean voice phishing dataset called KorCCVi. The results of the experiments reveal that the classification accuracy on a test set of the KoBERT model outperforms the performances of all other models with an accuracy score of 99.60%.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.3
• /
• pp.138-142
• /
• 2009

We propose an effective voice-phishing detection algorithm based on discriminative weight training. The detection of voice phishing is performed based on a Gaussian mixture model (GMM) incorporaiting minimum classification error (MCE) technique. Actually, the MCE technique is based on log-likelihood from the decoding parameter of the SMV(Selectable Mode Vocoder) directly extracted from the decoding process in the mobile phone. According to the experimental result, the proposed approach is found to be effective for the voice phishing detection.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1267-1277
• /
• 2021

The phone number used for advertising messages and voices used as bait in the voice phishing crime access stage is being used to send out a large amount of illegal loan spam, so we want to quickly block it. In this paper, our system is designed to block the usage of the phone number by rapidly restricting the use of the voice spam phone number that conducts illegal loan spam and voice phishing, and at the same time sends continuous calls to the phone number to prevent smooth phone call connection. The proposed system is a representative collaboration model between an illegal spam reporting agency and an investigation agency. As a result of developing the system and applying it in practice, the number of reports of illegal loaned voice spam and text spam decreased by 1/3, respectively. We can prove the effectiveness of this system by confirming that.

• Convergence Security Journal
• /
• v.14 no.5
• /
• pp.65-74
• /
• 2014

Phishing scans through wired telephones have been evolving into smissing and pharming. While we use wire or wireless telephones, text messages, e-mails, and online-banking conveniently, the ways of hacking and phishing attacks are getting developed and various. This paper investigates the various aspects of attacks depending on the kinds of phishing and suggests general prevention measures. In addition, the user-oriented practical preventive measures and government-driven long term measures are proposed in this paper. Technological developments, short or long term preventive measures proposed by the government, and continuous public relations could be solutions since in a short time, it could be difficult to eradicate phishing scams evolving continuously. Besides, the internet media as well as SNS are great helps in promoting the preventives against phishing and smissing. Finally this paper asserts that the newly developed service technology should be made carefully without security problems.

• Journal of the Korea Convergence Society
• /
• v.2 no.4
• /
• pp.9-14
• /
• 2011

Phishing is an attack to theft an user's identity by masquerading the user or the device. The number of phishing victims are sharply increased due to wide spread use of smart phones and messenger programs. Smart phones can operate various wi-fi based apps besides typical voice call and SMS functions. Generally, the messenger program such as Kakao Talk or Nate On is consisted of client and server functions. Thus, the authentication between the client and the server is essential to communicate securely. In this paper, we propose the messenger authentication protocol safe against smart phone phishing. To protect communications among clients, the proposed method provides message encryption and authentication functions.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.156-159
• /
• 2019

The damage from voice phishing reaches one trillion won in the past 5 years following report of Business Korea on August 28, 2018. Voice phishing and mobile phone scams are recognized as a top concern not only in Korea but also in over the world in recent years. In this paper, we propose an efficient system to identify the caller and alert or prevent of dangerous to users. Our system includes a mobile application and web server using client and server architecture. The main purpose of this system is to automatically display the information of unidentified callers when a user receives a call or message. A mobile application installs on a mobile phone to automatically get the caller phone number and send it to the server through web services to verify. The web server applies a machine learning to a global phone book with Blacklist and Whitelist to verify the phone number getting from the mobile application and returns the result.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.647-648
• /
• 2009

'보이스피싱(Voice Phishing)'은 전화를 통해 자신을 신뢰할 수 있는 대상으로 위장하여 개인 정보를 훔치는 행위이다. 최근 들어 보이스피싱 피해 사례가 급증하고 있으며 아직 딱히 대안이 없는 상태이다. 또한, IP 환경에서의 전화통화는 보이스피싱을 더욱 용이하게 하고 있다. 본 논문에서는 VoIP 환경에서 공개키 암호화 기법을 이용하여 발신자의 신원을 정확히 밝히는 인증을 통해 사용자가 악성 발신자에게 정보를 제공하거나 금전적 피해를 피할 수 있는 방법론을 제시한다. 이를 통해 사용자는 발신자의 정확한 신원 정보를 제공 받아 보이스피싱 당할 가능성을 줄일 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.297-299
• /
• 2021

Text classification is one of the popular tasks in Natural Language Processing (NLP) used to classify text or document applications such as sentiment analysis and email filtering. Nowadays, state-of-the-art (SOTA) Machine Learning (ML) and Deep Learning (DL) algorithms are the core engine used to perform these classification tasks with high accuracy, and they show satisfying results. This paper conducts a benchmarking performance's analysis of multiple SOTA algorithms on the first known labeled Korean voice phishing dataset called KorCCVi. Experimental results reveal performed on a test set of 366 samples reveal which algorithm performs the best considering the training time and metrics such as accuracy and F1 score.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.947-948
• /
• 2023

본 논문은 보이스피싱 취약 계층을 위해 통화 내용을 신속하게 처리하여 실시간으로 범죄 여부를 판별하는 VoIP 에 특화된 시스템을 제안하였다. 실제 보이스 피싱 통화 유형을 학습한 탐지 모델을 개발하여 API 로 배포하였다. 또한 보이스피싱 위험도가 일정 수준에 도달할 경우 사용자에게 보이스피싱 가능성을 경고하는 장치를 제작하였다. 본 연구는 보이스피싱을 사전에 탐지함으로써 개인정보의 유출 및 금융 피해를 예방하고 정보 보안을 실천하는 데 기여할 것으로 기대된다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.95-96
• /
• 2024

본 논문은 금융 범죄 중 하나인 보이스피싱을 실시간으로 예방하기 위한 탐지 기법을 제안한다. 제안된 모델은 수화기에 출력되는 음성을 녹음하고 네이버 CSR(Cloud Speech Recognition)을 통해 텍스트 파일로 변환한 후 딥러닝 기반의 KoBERT를 바탕으로 다양한 보이스피싱 패턴을 학습하여 실시간 환경에서의 신속하고 정확한 탐지를 위해 실제 통화 데이터를 적절하게 처리하여, 이를 통해 효과적인 보이스피싱 예방에 도움을 줄 것으로 예상된다.