• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.646-647
• /
• 2024

사이버 범죄가 증가함에 따라 익명성을 보장하는 암시장인 다크웹 내 불법적인 활동에 대한 모니터링의 중요성이 커졌다. 최근 다양한 분야에서 ChatGPT 의 쓰임이 주목받고 있듯이 다크웹에서도 전용 GPT 가 등장하였으며, 다크웹 생태계를 분석하고 정보를 수집하는데 이러한 다크웹 전용 생성형 인공지능 모델을 활용할 수 있다. 본 연구에서는 다크웹 GPT 에서 불법 행위와 관련된 질의를 통해 정보를 수집하고 해당 정보가 표면웹과 다크웹 상에서 다르게 쓰이고 있음을 확인함으로써 수사를 위한 다크웹 전용 GPT 활용 가능성 및 프롬프트 엔지니어링의 필요성을 탐구한다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.51-65
• /
• 2024

Accurate electrical load forecasting is important to the effective operation of power systems in smart grids. With the recent development in machine learning, artificial intelligence-based models for predicting power demand are being actively researched. However, since existing models get input variables as numerical features, the accuracy of the forecasting model may decrease because they do not reflect the semantic relationship between these features. In this paper, we propose a scheme for short-term load forecasting by using features extracted through the large language models for input data. We firstly convert input variables into a sentence-like prompt format. Then, we use the large language model with frozen weights to derive the embedding vectors that represent the features of the prompt. These vectors are used to train the forecasting model. Experimental results show that the proposed scheme outperformed models based on numerical data, and by visualizing the attention weights in the large language models on the prompts, we identified the information that significantly influences predictions.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1522-1534
• /
• 2015

This work presents a new method that can predict, without the use of training data, whether an input essay is written on a given topic. ConceptNet is a common-sense knowledge base that is generated automatically from sentences that are extracted from a variety of document types. An essay prompt is the topic that an essay should be written about. The method that is proposed in this paper uses ConceptNet and an essay prompt to decide whether or not an input essay is off-topic. We introduce a way to find the shortest path between two nodes on ConceptNet, as well as a way to calculate the semantic similarity between two nodes. Not only an essay prompt but also a student's essay can be represented by concept nodes in ConceptNet. The semantic similarity between the concepts that represent an essay prompt and the other concepts that represent a student's essay can be used for a calculation to rank "on-topicness" ; if a low ranking is derived, an essay is regarded as off-topic. We used eight different essay prompts and a student-essay collection for the performance evaluation, whereby our proposed method shows a performance that is better than those of the previous studies. As ConceptNet enables the conduction of a simple text inference, our new method looks very promising with respect to the design of an essay prompt for which a simple inference is required.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.539-545
• /
• 2023

초거대 언어모델은 심화된 언어적 이해를 요구하는 여러 분야에 높은 영향력을 미치고 있으나, 그에 수반되는 편향성과 윤리성에 대한 우려 또한 함께 증대되었다. 특히 편향된 언어모델은 인종, 성적 지향 등과 같은 다양한 속성을 가진 개인들에 대한 편견을 강화시킬 수 있다. 그러나 이러한 편향성에 관한 연구는 대부분 영어 문화권에 한정적이며 한국어에 관한 연구 또한 한국에서 발생하는 지역 갈등, 젠더 갈등 등의 사회적 문제를 반영하지 못한다. 이에 본 연구에서는 ChatGPT의 내재된 편향성을 도출하기 위해 의도적으로 다양한 페르소나를 부여하고 한국의 사회적 쟁점들을 기반으로 프롬프트 집합을 구성하여 생성된 문장의 독성을 분석하였다. 실험 결과, 특정 페르소나 또는 프롬프트에 관해서는 지속적으로 유해한 문장을 생성하는 경향성이 나타났다. 또한 각 페르소나-쟁점에 대해 사회가 갖는 편향된 시각이 모델에 그대로 반영되어, 각 조합에 따라 생성된 문장의 독성 분포에 유의미한 차이를 보이는 것을 확인했다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.499-504
• /
• 2023

본 연구에서는 민감한 개인정보의 유출과 남용 위험이 높아지고 있는 상황에서 정확한 개인정보 탐지 및 비식별화의 효율을 높이기 위해 개인정보 항목에 특화된 개체명 체계를 개발하였다. 개인정보 태그셋이 주석된 대화 데이터 4,981세트를 구축하고, 생성 AI 모델을 활용하여 개인정보 개체명 탐지 실험을 수행하였다. 실험을 위해 최적의 프롬프트를 설계하여 퓨샷러닝(few-shot learning)을 통해 탐지 결과를 평가하였다. 구축한 데이터셋과 영어 기반의 개인정보 주석 데이터셋을 비교 분석한 결과 고유식별번호 항목에 대해 본 연구에서 구축한 데이터셋에서 더 높은 탐지 성능이 나타났으며, 이를 통해 데이터셋의 필요성과 우수성을 입증하였다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.664-671
• /
• 2023

최근 다양한 거대 언어모델(Large Language Model)들이 개발되면서 프롬프트 엔지니어링의 대한 다양한 연구가 진행되고 있다. 본 논문에서는 퓨삿 학습 환경에서 개체명 인식의 성능을 높이기 위해서 제안된 템플릿이 필요 없는 프롬프트 튜닝(Template-free Prompt Tuning) 방법을 이용하고, 이 방법에서 사용된 라벨 단어 집합 생성 방법에 Maximal Marginal Relevance 알고리즘을 적용하여 해당 개체명에 대해 보다 다양하고 구체적인 라벨 단어 집합을 생성하도록 개선하였다. 실험 결과, 'LOC' 타입을 제외한 나머지 개체명 타입에서 'PER' 타입은 0.60%p, 'ORG' 타입은 4.98%p, 'MISC' 타입은 1.38%p 성능이 향상되었고, 전체 개체명 인식 성능은 1.26%p 향상되었다. 이를 통해 본 논문에서 제안한 라벨 단어 집합 생성 기법이 개체명 인식 성능 향상에 도움이 됨을 보였다.

• Journal of Fashion Business
• /
• v.28 no.3
• /
• pp.122-139
• /
• 2024

Image-generated AI is rapidly emerging as a powerful tool to augment human creativity and transform the art and design process through deep learning capabilities. The purpose of this study was to propose and demonstrate the feasibility of a new design development method that combined traditional design methods and technology by constructing image-generated AI prompts based on artists' formative elements. The study methodology consisted of analyzing Kazmir Malevich's theoretical considerations and applying them to AI prompts for design, print pattern development, and 3D digital design. This study found that the suprematist works of Kazmir Malevich were suitable as design and print pattern prompts due to their clear geometric shapes, colors, and spatial arrangement. The AI-prompted designs and print patterns produced diverse results quickly and enabled an efficient design process compared to traditional methods, although additional refinement was required to perfect the details. The AI-generated designs were successfully produced as 3D garments, thereby demonstrating that AI technology could significantly contribute to fashion design through its integration with artistic principles. This study has academic significance in that it proposes a prompt composition method applicable to fashion design by combining AI and artistic elements. It also has industrial significance in that it contributes to design innovation and the implementation of creative ideas by presenting an AI-based design process that can be practically applied.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.11
• /
• pp.481-492
• /
• 2023

This study explores how to build a Korean dataset to extract information from text using generative large language models. In modern society, mixed information circulates rapidly, and effectively categorizing and extracting it is crucial to the decision-making process. However, there is still a lack of Korean datasets for training. To overcome this, this study attempts to extract information using text-based zero-shot learning using a generative large language model to build a purposeful Korean dataset. In this study, the language model is instructed to output the desired result through prompt engineering in the form of "system"-"instruction"-"source input"-"output format", and the dataset is built by utilizing the in-context learning characteristics of the language model through input sentences. We validate our approach by comparing the generated dataset with the existing benchmark dataset, and achieve 25.47% higher performance compared to the KLUE-RoBERTa-large model for the relation information extraction task. The results of this study are expected to contribute to AI research by showing the feasibility of extracting knowledge elements from Korean text. Furthermore, this methodology can be utilized for various fields and purposes, and has potential for building various Korean datasets.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.493-498
• /
• 2024

With the rapid advancement of generative artificial intelligence technology, there is growing interest in how to utilize it in practical applications. Additionally, the importance of prompt engineering to generate results that meet user demands is being newly highlighted. Exploring the new possibilities of generative AI can hold significant value. This study aims to utilize ChatGPT 4.0, a leading generative AI, to propose an effective method for evaluating user experience through the analysis of online customer review data. The user experience evaluation method was based on the six-layer elements of user experience: 'functionality', 'reliability', 'usability', 'convenience', 'emotion', and 'significance'. For this study, a literature review was conducted to enhance the understanding of prompt engineering and to grasp the clear concept of the user experience hierarchy. Based on this, prompts were crafted, and experiments for the user experience evaluation method were carried out using the analysis of collected online customer review data. In this study, we reveal that when provided with accurate definitions and descriptions of the classification processes for user experience factors, ChatGPT demonstrated excellent performance in evaluating user experience. However, it was also found that due to time constraints, there were limitations in analyzing large volumes of data. By introducing and proposing a method to utilize ChatGPT 4.0 for user experience evaluation, we expect to contribute to the advancement of the UX field.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.77-90
• /
• 2024

This research aims to investigate methods of leveraging generative artificial intelligence in service sectors where consumer sentiment and experience are paramount, focusing on minimizing hallucination phenomena during usage and developing strategic services tailored to consumer sentiment and experiences. To this end, the study examined both mechanical approaches and user-generated prompts, experimenting with factors such as business item definition, provision of persona characteristics, examples and context-specific imperative verbs, and the specification of output formats and tone concepts. The research explores how generative AI can contribute to enhancing the accuracy of personalized content and user satisfaction. Moreover, these approaches play a crucial role in addressing issues related to hallucination phenomena that may arise when applying generative AI in real services, contributing to consumer service innovation through generative AI. The findings demonstrate the significant role generative AI can play in richly interpreting consumer sentiment and experiences, broadening the potential for application across various industry sectors and suggesting new directions for consumer sentiment and experience strategies beyond technological advancements. However, as this research is based on the relatively novel field of generative AI technology, there are many areas where it falls short. Future studies need to explore the generalizability of research factors and the conditional effects in more diverse industrial settings. Additionally, with the rapid advancement of AI technology, continuous research into new forms of hallucination symptoms and the development of new strategies to address them will be necessary.