• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.4
• /
• pp.777-785
• /
• 2016

IM(Instant Messenger) chatting service can carry user's various information including life style, geographical position, and psychology & crime history and thus forensic analysis on the IM service is desirable. But, forensic analysis for KakaoTalk's chatting service is not well studied yet. For this reason, we study KakaoTalk's forensic analysis focusing on chatting service. This paper first details a general method of IM forensics investigating the previous articles about IM forensics although there are not many articles. Second, we discuss methodologies for IM forensics wherein we present analysis of table structure and method for reconstruction of chatting message. These result in the basic element of forensic tools of KakaoTalk chatting message. Last, we compare artifacts of KakaoTalk with that of WhatsApp. We conclude that these applications are, at least, different in that table structures and the ways to reconstruct chatting messages are not same and therefore digital evidences or artifacts are not same and somewhat distinct.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.27 no.1
• /
• pp.7-14
• /
• 2017

Theoretically, it is known that common beliefs and/or common knowledge cannot be attained in asynchronously distributed multiagent environments, however, it show that some propositions with deadlines can be attained as common beliefs among a set of fully trusted agents even when they communicate to each other asynchronously. Generally, in the multiagent environment, the attainment of common beliefs is approached as a problem of communication, and for the common beliefs paradox that the common beliefs is not attained on a system without communication time restriction is applied to loose coarser granularity and it prove that forming common beliefs is possible by relaxing necessary requirements through the KakaoTalk chatting model. I also experimented with the reasoning function that confirms the common beliefs by inquiring about the common belief generated by implementing the inference function in each agent of the KakaoTalk chatting model. Through utilizing metalogic programming, a formalization of the presentation and reasoning of common beliefs has been achieved, and the group chatting model of KakaoTalk was adopted in experiments to show that common beliefs can be formed among distributed agents using asynchronous communication.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.362-367
• /
• 2019

본 논문에서는 Self-Attention을 활용한 딥러닝 기반 문맥의존 철자오류 교정 모델을 제안한다. 문맥의존 철자오류 교정은 최근 철자오류 교정 분야에서 활발히 연구되고 있는 문제 중 하나이다. 기존에는 규칙 기반, 확률 기반, 임베딩을 활용한 철자오류 교정이 연구되었으나, 아직 양질의 교정을 수행해내기에는 많은 문제점이 있다. 따라서 본 논문에서는 기존 교정 모델들의 단점을 보완하기 위해 Self-Attention을 활용한 문맥의존 철자오류 교정 모델을 제안한다. 제안 모델은 Self-Attention을 활용하여 기존의 임베딩 정보에 문맥 의존적 정보가 반영된 더 나은 임베딩을 생성하는 역할을 한다. 전체 문장의 정보가 반영된 새로운 임베딩을 활용하여 동적으로 타겟 단어와의 관련 단어들을 찾아 문맥의존 철자 오류교정을 시행한다. 본 논문에서는 성능평가를 위해 세종 말뭉치를 평가 데이터로 이용하여 제안 모델을 실험하였고, 비정형화된 구어체(Kakao Talk) 말뭉치로도 평가 데이터를 구축해 실험한 결과 비교 모델보다 높은 정확율과 재현율의 성능향상을 보였다.