• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.210-213
• /
• 2024

이 연구는 거대한 언어 모델 (Large Language Model, LLM)과 도구를 결합한 시스템의 보안 문제를 다룬다. 프롬프트 주입과 같은 보안 취약점을 분석하고 이를 극복하기 위한 프롬프트 권한 분리 기법을 제안한다. 이를 통해 LLM-도구 결합 시스템에서의 사용자 데이터의 기밀성과 무결성을 보장한다.

• International Journal of Computer Science & Network Security
• /
• v.21 no.3
• /
• pp.37-40
• /
• 2021

This paper presents a design of a speech to sign language converter for deaf and hard of hearing people. The device is low-cost, low-power consumption, and it can be able to work entirely offline. The speech recognition is implemented using an open-source API, Pocketsphinx library. In this work, we proposed a context-oriented language model, which measures the similarity between the recognized speech and the predefined speech to decide the output. The output speech is selected from the recommended speech stored in the database, which is the best match to the recognized speech. The proposed context-oriented language model can improve the speech recognition rate by 21% for working entirely offline. A decision module based on determining the similarity between the two texts using Levenshtein distance decides the output sign language. The output sign language corresponding to the recognized speech is generated as a set of sequential images. The speech to sign language converter is deployed on a Raspberry Pi Zero board for low-cost deaf assistive devices.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.391-395
• /
• 2021

텍스트 스타일 변환은 입력 스타일(source style)로 쓰여진 텍스트의 내용(content)을 유지하며 목적 스타일(target style)의 텍스트로 변환하는 문제이다. 텍스트 스타일 변환을 시퀀스 간 변환 문제(sequence-to-sequence)로 보고 기존 기계학습 모델을 이용해 해결할 수 있지만, 모델 학습에 필요한 각 스타일에 대응되는 병렬 말뭉치를 구하기 어려운 문제점이 있다. 따라서 최근에는 비병렬 말뭉치를 이용해 텍스트 스타일 변환을 수행하는 방법들이 연구되고 있다. 이 연구들은 주로 인코더-디코더 구조의 생성 모델을 사용하기 때문에 입력 문장이 가지고 있는 내용이 누락되거나 다른 내용의 문장이 생성될 수 있는 문제점이 있다. 본 논문에서는 마스크 언어 모델(masked language model)을 이용해 입력 텍스트의 내용을 유지하면서 원하는 스타일로 변경할 수 있는 텍스트 스타일 변환 방법을 제안하고 한국어 긍정-부정, 채팅체-문어체 변환에 적용한다.

• Journal of Advanced Navigation Technology
• /
• v.28 no.3
• /
• pp.369-374
• /
• 2024

This study aims to improve the accessibility and efficiency of sign language education for both hearing impaired and non-deaf people. To this end, we developed VR realistic sign language education content that integrates hand tracking technology and conversational AI. Through this content, users can learn sign language in real time and experience direct communication in a virtual environment. As a result of the study, it was confirmed that this integrated approach significantly improves immersion in sign language learning and contributes to lowering the barriers to sign language learning by providing learners with a deeper understanding. This presents a new paradigm for sign language education and shows how technology can change the accessibility and effectiveness of education.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.2
• /
• pp.467-472
• /
• 2013

Many people are showing interest on computing and computer programming ever as much as the smartphone become more popular. Computer programming languages, however, like Java or C++ being used to teach freshmen in computer science-related majors as the first programming language they will study are so difficult to understand. In this paper, we proposed a short-term curriculum for teaching computer programming using App Inventor for Android to freshmen students major in computer science as the first programming language they will study, which can encourage their interest in computer programming. According to survey from students participated in actual teaching, we found that the proposed curriculum can contribute to increase their interest on computer programming and even self-confidence on development of applications for smarphone.

• International Journal of Computer Science & Network Security
• /
• v.22 no.12
• /
• pp.132-138
• /
• 2022

Due to the impacts of the current pandemic COVID-19 and the continuation of studying online. There is an urgent need for an effective and efficient education platform to help with the continuity of studying online. Therefore, the question bank system (QB) is introduced. The QB system is designed as a website to create a single platform used by faculty members in universities to generate questions and store them in a bank of questions. In addition to allowing them to add two types of questions, to help the lecturer create exams and present the results of the students to them. For the implementation, two languages were combined which are PHP and Python to generate questions by using Artificial Intelligence (AI). These questions are stored in a single database, and then these questions could be viewed and included in exams smoothly and without complexity. This paper aims to help the faculty members to reduce time and efforts by using the Question Bank System by using AI and Natural Language Processing (NLP) to extract and generate questions from given text. In addition to the tools used to create this function such as NLTK and TextBlob.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.837-852
• /
• 2021

Recurrent neural network (RNN) architectures have been used for language modeling (LM) tasks that require learning long-range word or character sequences. However, the RNN architecture is still suffered from unstable gradients on long-range sequences. To address the issue of long-range sequences, an attention mechanism has been used, showing state-of-the-art (SOTA) performance in all LM tasks. A differentiable neural computer (DNC) is a deep learning architecture using an attention mechanism. The DNC architecture is a neural network augmented with a content-addressable external memory. However, in the write operation, some information unrelated to the input word remains in memory. Moreover, DNCs have been found to perform poorly with low numbers of weight parameters. Therefore, we propose a robust memory deallocation method using a limited retention vector. The limited retention vector determines whether the network increases or decreases its usage of information in external memory according to a threshold. We experimentally evaluate the robustness of a DNC implementing the proposed approach according to the size of the controller and external memory on the enwik8 LM task. When we decreased the number of weight parameters by 32.47%, the proposed DNC showed a low bits-per-character (BPC) degradation of 4.30%, demonstrating the effectiveness of our approach in language modeling tasks.

• Proceedings of the Korea Database Society Conference
• /
• 1997.10a
• /
• pp.24-48
• /
• 1997

. Supported byㆍDARPA′s image Understanding (IU) program under "Video Retrieval Based on Language and image Analysis" project.DARPA′s Computer Assisted Education and Training Initiative program (CAETI)ㆍObjective: Develop practical systems for automatic understanding and indexing of video sequences using both audio and video tracks(omitted)

• Proceedings of the Korean Society for Language and Information Conference
• /
• 1996.02a
• /
• pp.63-72
• /
• 1996

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06b
• /
• pp.189-192
• /
• 2011

클라우드 컴퓨팅에서 사용되는 웹 서비스들은 BPEL에 의해 여러 서비스들이 새로운 웹 서비스로 조합 되어지며, 서비스가 제대로 동작하는지를 검증하기 위해 Petri nets, Abstract State Machine(ASM), BPECalculus 등의 검증 방법을 사용한다. 이러한 검증 방법은 BPEL을 사용하여 새로 만들어진 웹 서비스들이 안정적으로 동작하는지를 검증하는 것으로, 웹 서비스 설계와 검증이 서로 분리되어 있다. 본 논문에서는 명세, 분석 및 검증의 전 과정에서 프로세스의 포함관계, 상태정보, Interaction, Mobility 등을 그래프로 표현하며, 한 단계의 그래프를 통하여 시스템 전체의 복잡도 및 시스템의 행위를 예측할 수 있는 Onion Visual Language(OVL)을 사용하여 BPEL로 설계 되는 클라우드 웹 서비스들을 OVL로 변환 후 이를 분석 및 검증한다. 추후 OVL은 서로 다른 클라우드 안에서의 웹 서비스 재사용을 위한 동일성 검증을 위한 방법으로 사용될 수 있다.