• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.1013-1015
• /
• 2022

본 논문은 보이스피싱에 취약한 VoIP와 일반 유선전화 상의 보안을 위해 유선전화의 대화내용을 Google STT API 및 텍스트 자연어 처리를 통해 실시간으로 보이스피싱 위험도를 알 수 있는 모델을 제안했다. 보이스피싱 데이터를 Data Augmentation와 BERT 모델을 활용해 보이스피싱을 예방하는 솔루션을 구상했다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1045-1046
• /
• 2023

본 논문은 디지털 소외계층과 사회적 약자를 고려한 보이스피싱 예방 솔루션을 제안한다. 통화 내용을 AWS Transcribe를 활용한 STT와 NLP 알고리즘을 사용해 실시간으로 보이스피싱 위험도를 파악하고 결과를 사용자에게 전달하도록 한다. NLP 알고리즘은 KoBIGBIRD와 DeBERTa 모델 각각을 커스터마이즈하여 보이스피싱 탐지에 적절하게 파인튜닝 했다. 이후, 성능과 인퍼런스를 비교하여 더 좋은 성능을 보인 KoBIGBIRD 모델로 보이스피싱 탐지를 수행한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.794-795
• /
• 2023

금융 소외 계층은 여전히 보이스피싱과 같은 금융 범죄의 위험에 직면하고 있다. 본 논문에서는 이러한 위험을 완화하기 위해 모바일 애플리케이션을 통한 보이스피싱 예방을 강화하고자 한다. 이를 위해 AI 기술을 적용하여 데이터를 추출, 학습 및 분석하는 혁신적인 보안 기술을 개발하고자 한다. 이러한 기술의 적용을 통해 금융 소외 계층을 더 효과적으로 보호하고 금융범죄로부터 보안을 강화하는데 기여할 것으로 기대한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.214-215
• /
• 2024

삼성 스마트폰 계정 기반 서비스인 다른 기기에서 전화/문자하기(CMC) 기능이 보이스피싱의 새로운 기술로 동원되고 있다. 기존의 심박스와 같은 불법 중계기보다 발신 번호 변작에 쉽게 활용할 수 있어 CMC 기능을 악용한 보이스피싱 범죄가 증가하고 있으나, 이에 대한 연구가 미비한 현실이다. 본 논문에서는 삼성 기기에서의 CMC 활성화 및 기능 사용 여부에 따른 안드로이드 시스템 로그에서의 차이를 분석하고, 이를 바탕으로 보이스피싱 수사에 활용할 수 있는 포렌식 아티팩트 분석 방법을 제안한다.

• KSCI Review
• /
• v.14 no.2
• /
• pp.255-262
• /
• 2006

A Malicious code is used to SMiShing disguised as finance mobile Vishing, using Phishing, Pharming mail, VoIP service etc. to capture of personal information. A Malicious code deletes in Anti-Virus Spyware removal programs. or to cure use. By the way, the Malicious cord which is parasitic as use a DLL Injection technique, and operate are Isass.exe, winlogon.exe. csrss.exe of the window operating system. Be connected to the process that you shall be certainly performed of an exe back, and a treatment does not work. A user forces voluntarily a process, and rebooting occurs, or a blue screen occurs, and Compulsory end, operating system everyone does. Propose a treatment way like a bird curing a bad voice code to use a DLL Injection technique to occur in these fatal results. Click KILL DLL since insert voluntarily an end function to Thread for a new treatment, and Injection did again the Thread which finish an action of DLL, and an end function has as control Thread, and delete. The cornerstone that the treatment way that experimented on at these papers and a plan to solve will become a researcher or the revolutionary dimension that faced of a computer virus, and strengthen economic financial company meeting Ubiquitous Security will become.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.5 s.43
• /
• pp.251-258
• /
• 2006

A Malicious code is used to SMiShing disguised as finance mobile Vishing, using Phishing, Pharming mail, VoIP service etc. to capture of personal information. A Malicious code deletes in Anti-Virus Spyware removal programs, or to cure use. By the way, the Malicious cord which is parasitic as use a DLL Injection technique, and operate are Isass.exe, winlogon.exe, csrss.exe of the window operating system. Be connected to the process that you shall be certainly performed of an exe back, and a treatment does not work. A user forces voluntarily a process, and rebooting occurs, or a blue screen occurs, and Compulsory end, operating system everyone does. Propose a treatment way like a bird curing a bad voice code to use a DLL Injection technique to occur in these fatal results. Click KILL DLL since insert voluntarily an end function to Thread for a new treatment, and Injection did again the Thread which finish an action of DLL, and an end function has as control Thread, and delete. The cornerstone that the treatment wav that experimented on at these papers and a plan to solve will become a researcher of the revolutionary dimension that faced of a computer virus, and strengthen economic financial company meeting Ubiquitous Security will become.

• The Journal of the Korea Contents Association
• /
• v.13 no.9
• /
• pp.49-57
• /
• 2013

Recently, in spite of fast development of smartphone, recording technology has saved, listened and searched in server based on PSTN and wired network. We introduce a mobile transcript recording system for mobile office environment. All calls using smartphone are automatically saved on memory in mobile device with this system. Thereafter, recording files are encrypted and uploaded to recording server by user authority. Beside recording data in database on web server is saved and managed efficiently and economically through web application. This system is able to be used in legal corroborative facts and the prevention of crime such as voice phishing.

• Journal of Information Processing Systems
• /
• v.17 no.5
• /
• pp.867-878
• /
• 2021

In context awareness and user intention tasks, dataset construction is expensive because specific domain data are required. Although pretraining with a large corpus can effectively resolve the issue of lack of data, it ignores domain knowledge. Herein, we concentrate on data domain knowledge while addressing data scarcity and accordingly propose a multi-channel long short-term memory (LSTM). Because multi-channel LSTM integrates pretrained vectors such as task and general knowledge, it effectively prevents catastrophic forgetting between vectors of task and general knowledge to represent the context as a set of features. To evaluate the proposed model with reference to the baseline model, which is a single-channel LSTM, we performed two tasks: voice phishing with context awareness and movie review sentiment classification. The results verified that multi-channel LSTM outperforms single-channel LSTM in both tasks. We further experimented on different multi-channel LSTMs depending on the domain and data size of general knowledge in the model and confirmed that the effect of multi-channel LSTM integrating the two types of knowledge from downstream task data and raw data to overcome the lack of data.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.354-356
• /
• 2022

Profiles are the most basic means of representing who you are, and you represent yourself. In SNS and chatting apps that use profiles the most, profiles are used to express their experiences, moods, and thoughts. However, in voice phishing, impersonation crimes, etc., they steal and impersonate a photo such as a profile, and then deceive the victim into committing a crime. In this paper, to prevent the crime of impersonation, the authenticity of the original is determined using NFT that can distinguish the original. Minting the original profile picture with NFT, registering it on the blockchain network, and storing the original file in off-chain IPFS. This is expected to be effective in preventing impersonation crimes because it is possible to easily determine the authenticity of the profile picture.

• Journal of Digital Convergence
• /
• v.16 no.7
• /
• pp.9-17
• /
• 2018

Recently, much attention in electronic financial fraud has been dramatically increased. In particular, the electronic financial fraud has been transforming to social engineering. Despite the growing interest in electronic financial fraud, few guidelines exist how to effectively avoid the serious damage from electronic financial fraud. Moreover, it is rarely investigated cases of victims from financial fraud. Therefore, the purpose of this study is to investigate why financial fraud crime victims occurs. To enhance mundane realism, we conducted Focus Group Interview(FGI) with actual victims from financial fraud crime. Drawing analysis of FGI with actual victims, we found that there are certain damage patterns. Further, we found that the reason why financial fraud crime victims occurs is optimistic biases of humans rooted in behavioral economics. Therefore, this study provides the valuable guidelines and directions to prevent electronic financial fraud based on risk and crisis management perspective. Ultimately, this study is able to help the establishment and implementation of a comprehensive electronic financial fraud prevention policy.