• Annual Conference of KIPS
• /
• 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.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.628-631
• /
• 2023

We present a new approach to evaluate the generated texts by Large Language Models (LLMs) for meme classification. Analyzing an image with embedded texts, i.e. meme, is challenging, even for existing state-of-the-art computer vision models. By leveraging large image-to-text models, we can extract image descriptions that can be used in other tasks, such as classification. In our methodology, we first generate image captions using BLIP-2 models. Using these captions, we use GPT-4 to evaluate the relationship between the caption and the meme text. The results show that OPT_6.7B provides a better rating than other LLMs, suggesting that the proposed method has a potential for meme classification.

• Journal of Information Technology Applications and Management
• /
• v.30 no.4
• /
• pp.11-28
• /
• 2023

This paper proposes a method of integrating ChatGPT with traditional chatbot systems to enhance conversational artificial intelligence(AI) and create more efficient conversational systems. Traditional chatbot systems are primarily based on classification models and are limited to intent classification and simple response generation. In contrast, ChatGPT is a state-of-the-art AI technology for natural language generation, which can generate more natural and fluent conversations. In this paper, we analyze the business service areas that can be integrated with ChatGPT and traditional chatbots, and present methods for conducting conversational scenarios through case studies of service types. Additionally, we suggest ways to integrate ChatGPT with traditional chatbot systems for intent recognition, conversation flow control, and response generation. We provide a practical implementation example of how to integrate ChatGPT with traditional chatbots, making it easier to understand and build integration methods and actively utilize ChatGPT with existing chatbots.

• Korean Security Journal
• /
• no.57
• /
• pp.231-252
• /
• 2018

we have long used foreign words in using the term for guard security despite the obvious existence of own language, the use of foreign terms is strong in foreign feeling in delivery. and also weakens the true meaning of security. there are no terms expressed independently in korean, and we(they) are negligent in finding them and use the terms of foreign language. as a result, we(they) brought about a lack of choice in terms of proper security for our langage. currenty, it is widely used as a security guard even though there is an appropraite word that corresponds to the meaning and meaning of security guards in our words, we still use enlish expressions. there is because the English language is used for convenience regardless of weather the term is appropraite or not, and as the power of the English language is great amid in the trend of globalization. lt is easy to use english without thinking in terms of the use of terminology. ultimitely, however, this is due to the lack of awareness of the korean language. with these reasons, we must find the term of security guards in pure korean language. until now, we have used the terms 'guard, security, protect' as the terms security and protection the term 'Jikim' refers to the korean language as a means to be vigilant and guarded. Jikim refers to the action of maintaining the current safe state. Like school jikimi, children safety jikimi and environment jikimi, Jikim is already being used in many places. Therefore, the term 'guard' should be changed to an appropriate Korean term, and the term 'Jikim' is considered to be the most appropriate term in various sections. so, 'Jikim' will be appropriate in korean, which corresponds to the meaning of security guards. the guardian here is called the Jikimi. Jikimi is a combination of the word Jikim and the korean pronounce 'I' which means people

• Journal of Digital Convergence
• /
• v.13 no.10
• /
• pp.437-444
• /
• 2015

In this age of digital technology, we often experience overwhelming and inescapable feelings of weakness, vulnerability, loneliness and fear that propel us into a state of gloom or frustration. Poetry can also serve as a therapeutic medium for overcoming them because it provides unique opportunities for self-discovery and is a catalyst for healing and self-integration. This study examines the healing properties in English poetry. First, Poetry has expressed innermost feelings, which can serve as a catharsis the emotional release. Therefore it can nurture and strengthen the human capacity. Poetic language heals our soul and mind. Rhythm which comes in many forms and poetic form also are important components of the therapeutic value of poetry. So we can say English poetry can be used as a therapeutic means for healing and personal growth.

• Archives of design research
• /
• v.13 no.3
• /
• pp.235-244
• /
• 2000

The form or interpretation of today's communication design lies in content development and storytelling to the designer. However, the understanding as well as exteriment of the mechanisms behind content creation remain an obscure concept. Particularly given the rapid development of digital technology, since the 1980s typography has been quickly transforming itself from its static state to the more dynamic'Type in Motion'. following this came the need for solid research in'Type in Motion'as the best means of presenting story content and form The previous passage has studied the effectiveness of'Type in Motion'in presenting this kind of content. Type in Motion has allowed vast developments in the methods of story presentation. It is Due that today's digital technology has given many benefits to the designer, but on the other hand, it has required a more diverse understanding of digital design. furthermore, even the designers in digital typography lack the understanding and research in'Type in Motion'not only as a mode of visual language but also of verbal language. Given this point· in hand, the entire interpretation and interrelation of story and'Type in Motion'can be considered essential, given their shaping of effective digital typography.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.173-179
• /
• 2004

Due to the insufficiency of training data in large vocabulary continuous speech recognition, similar context dependent phones can be clustered by decision trees to share the data. When the decision trees are built and used to predict unseen triphones, a phonetic question set is required. The phonetic question set, which contains categories of the phones with similar co-articulation effects, is usually generated by phonetic or linguistic experts. This knowledge-based approach for generating phonetic question set, however, may reduce the homogeneity of the clusters. Moreover, the experts must adjust the question sets whenever the language or the PLU (phone-like unit) of a recognition system is changed. Therefore, we propose a data-driven method to automatically generate phonetic question set. Since the proposed method generates the phone categories using speech data distribution, it is not dependent on the language or the PLU, and may enhance the homogeneity of the clusters. In large vocabulary speech recognition experiments, the proposed algorithm has been found to reduce the error rate by 14.3%.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.400-405
• /
• 2018

본 연구는 한국어 화장품 리뷰 코퍼스의 자질기반 감성 분석을 위하여, 이 도메인에서 실현되는 중요한 다단어 표현(MWE)의 유한상태 그래프 사전과 문법을 구축하는 방법론을 제시하고, 실제 구축된 사전과 문법의 성능을 평가하는 것을 목표로 한다. 본 연구에서는 자연어처리(NLP)에서 중요한 화두로 논의되어 온 MWE의 어휘-통사적 특징을 부분문법 그래프(LGG)로 형식화하였다. 화장품 리뷰 코퍼스에 DECO 한국어 전자사전을 적용하여 어휘 빈도 통계를 획득하고 이에 대한 언어학적 분석을 통해 극성 MWE(Polarity-MWE)와 화제 MWE(Topic MWE)의 전체 네 가지 하위 범주를 분류하였다. 또한 각 모듈간의 상호관계에 대한 어휘-통사적 속성을 반복적으로 적용하는 이중 증식(double-propagation)을 통해 자원을 확장하였다. 이 과정을 통해 구축된 대용량 MWE 유한그래프 사전 DECO-MWE의 성능을 테스트한 결과 각각 0.844(Pol-MWE), 0.742(Top-MWE)의 조화평균을 보였다. 이를 통해 본 연구에서 제안하는 MWE 언어자원 구축 방법론이 다양한 도메인에서 활용될 수 있고 향후 자질기반 감성 분석에 중요한 자원이 될 것임을 확인하였다.

• The Journal of Engineering Research
• /
• v.6 no.1
• /
• pp.111-124
• /
• 2004

The advanced semiconductor technology and electronic design automation(EDA) tools make it possible to implement the system on the programmable logic devices. The electronic design method is also changing from schematic capture to hardware description language. In this paper, I present the architecture of multi-dimensional digital filter which can be efficiently implemented on PLDs. This is based on the former research results which are called algorithm decomposition technique. Algorithm decomposition technique is used to obtain the computational primitive from the state space equations of the multi-dimensional digital filtering algorithm. The obtained computational primitive is designed with VHDL. This can be used to implement the filtering system as a component. The designed filtering system is implemented on the PLD. Therefore, the filter can be upgradable on system. It is greatly reduced the time-to-market time of the system that is based on the multi-dimensional filter.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.2
• /
• pp.192-199
• /
• 2003

Agents in multi-agent systems (MAS ) are required to achieve their own goals. An agent s goal, however, can conflict with others either when agents compete with each other to achieve a common goal or when they have to use a set of limited resources to accomplish agents divergent goals. In either case, agents need to be designed to reach a mutual acceptable state where they can avoid any goal conflicts through negotiation with others to achieve their goals. In this paper, we consider a BDI agent architecture where belief, desire, and intention are the three major components for agents mental attitudes and represent resource-bounded BDI agents in logic programming framework. We propose a negotiation algorithm for BDI agents solving their problems without goal conflicts in distributed cooperative environments. Finally, we describe a simple scenario to show the effectiveness of the negotiation algorithm implemented in a negotiation meta-language.