• Journal of Korea Game Society
• /
• v.19 no.5
• /
• pp.103-114
• /
• 2019

We have been studying a deep-learning program that can communicate with other users in online communities since 2017. But there were problems with processing a Korean data set because of Korean characteristics. Also, low usage of GPUs of RNN models was a problem too. In this study, as Natural Language Processing models are improved, we aim to make better results using these improved models. To archive this, we use a Transformer model which includes Self-Attention mechanism. Also we use MeCab, korean morphological analyzer, to address a problem with processing korean words.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4625-4647
• /
• 2020

The Generative Adversarial Networks, as one of the most creative deep learning models in recent years, has achieved great success in computer vision and natural language processing. It uses the game theory to generate the best sample in generator and discriminator. Recently, many deep learning models have been applied to the security field. Along with the idea of "generative" and "adversarial", researchers are trying to apply Generative Adversarial Networks to the security field. This paper presents the development of Generative Adversarial Networks. We review traditional generation models and typical Generative Adversarial Networks models, analyze the application of their models in natural language processing and computer vision. To emphasize that Generative Adversarial Networks models are feasible to be used in security, we separately review the contributions that their defenses in information security, cyber security and artificial intelligence security. Finally, drawing on the reviewed literature, we provide a broader outlook of this research direction.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.308-310
• /
• 2022

With the rise of the Al wave, task-oriented dialogue systems have become one of the popular research directions in academia and industry. Currently, task-oriented dialogue systems mainly adopt pipelined form, which mainly includes natural language understanding, dialogue state decision making, dialogue state tracking and natural language generation. However, pipelining is prone to error propagation, so many task-oriented dialogue systems in the market are only for single-round dialogues. Usually single- domain dialogues have relatively accurate semantic understanding, while they tend to perform poorly on multi-domain, multi-round dialogue datasets. To solve these issues, we developed a paddle-based multi-domain task-oriented Chinese dialogue system. It is based on NEZHA-base pre-training model and CrossWOZ dataset, and uses intention recognition module, dichotomous slot recognition module and NER recognition module to do DST and generate replies based on rules. Experiments show that the dialogue system not only makes good use of the context, but also effectively addresses long-term dependencies. In our approach, the DST of dialogue tracking state is improved, and our DST can identify multiple slotted key-value pairs involved in the discourse, which eliminates the need for manual tagging and thus greatly saves manpower.

• International conference on construction engineering and project management
• /
• 2013.01a
• /
• pp.60-65
• /
• 2013

Introducing the concept of construction safety in the design/engineering phase can improve the efficiency and effectiveness of safety management on construction sites. In this sense, further improvements for safety can be made in the design/engineering phase through the development of (1) an automated hazard identification process that is little dependent on user knowledge, (2) an automated construction schedule generation to accommodate varying hazard information over time, and (3) a visual representation of the results that is easy to understand. In this paper, we formulate an automated hazard identification framework for construction safety by extracting hazard information from related regulations to eliminate human interventions, and by utilizing a visualization technique in order to enhance users' understanding on hazard information. First, the hazard information is automatically extracted from textual safety and health regulations (i.e., Occupational Safety Health Administration (OSHA) Standards) by using natural language processing (NLP) techniques without users' interpretations. Next, scheduling and sequencing of the construction activities are automatically generated with regard to the 3D building model. Then, the extracted hazard information is integrated into the geometry data of construction elements in the industry foundation class (IFC) building model using a conformity-checking algorithm within the open source 3D computer graphics software. Preliminary results demonstrate that this approach is advantageous in that it can be used in the design/engineering phases of construction without the manual interpretation of safety experts, facilitating the designers' and engineers' proactive consideration for improving safety management.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.3
• /
• pp.123-131
• /
• 1994

This paper describes on the analysis of morphemes and it's synthesis being constituted of Korean word phrases. To analyze morphemes, we propose the introduction of "morph" for morpheme features in lexicon and the usage of chart data structures. it controls over the generation of unnecessary morpheme, and extracts every possible morpheme unit in a word phrase which minimized lexicon investigation by using heuristic information. Moreover, to synthesize morphemes, it is composed of every possible analyzed morphemes in word phrases to take advantage of speech and union information which can be obtained for program. Therefore, the systhesis of analyzed morphemes were designed to aid a syntactic analysis next step of natural language processing. This system for analyzing and systhesizing morpheme was to generate a word phrase by unifying syntactic and semantic features of analyzed morphemes in lexicon, and then established by C language of the personal computer.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.109-114
• /
• 2022

질의 생성 모델은 스마트 스피커, 챗봇, QA 시스템, 기계 독해 등 다양한 서비스에 사용되고 있다. 모델을 다양한 서비스에 잘 적용하기 위해서는 사용자들의 실제 질의 특성을 반영한 자연스러운 질의를 만드는 것이 중요하다. 본 논문에서는 사용자 질의 특성을 반영한 간결하고 자연스러운 질의 자동 생성 모델을 소개한다. 제안 모델은 topic 키워드를 통해 모델에게 생성 자유도를 주었으며, 키워드형 질의→자연어 질의→응답으로 연결되는 chain-of-thought 형태의 다중 출력 구조를 통해 인과관계를 고려한 결과를 만들도록 했다. 최종적으로 MRC 필터링과 일관성 필터링을 통해 고품질 질의를 선별했다. 베이스라인 모델과 비교해 제안 모델은 질의의 유효성을 크게 높일 수 있었다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.715-718
• /
• 2023

본 연구는 ChatGPT 모델의 특성과 장점을 활용하여 프롬프트 엔지니어링 기법을 연구하고자 하였다. 프롬프트는 엔지니어가 원하는 결과를 잘 얻을 수 있도록 하는 것이 목표이기 때문에 ChatGPT와 프롬프트 엔지니어링의 상호작용과 효과적인 프롬프트 엔지니어링 기법을 개발할 필요가 있다. 연구 방법으로는 ChatGPT에 대한 학습자 사전 설문조사에서 학습자를 분석하였고, 이를 반영하여 프로그래밍 문제를 제시하고 해결하는 과정을 거치면서 다양한 ChatGPT 사용에 대한 분석과 학습자 분석이 이루어졌다. 그 결과 비전공자가 듣고 있는 프로그래밍 수업에서 ChatGPT를 활용하여 얻은 통찰력으로 프롬프트에 필요한 가이드 라인을 마련하였다. 본 연구를 기반으로 향후 비전공자를 위한 파이썬 프로그래밍 수업에서 ChatGPT를 활용한 수업모델을 제시하고 학습자의 피드백 또는 적응형 학습에 활용할 수 있는 방법을 모색할 것이다.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.63-69
• /
• 2021

When a person sees a sentence and understands the sentence, the person understands the sentence by reminiscent of the main word in the sentence as an image. Text-to-image is what allows computers to do this associative process. The previous deep learning-based text-to-image model extracts text features using Convolutional Neural Network (CNN)-Long Short Term Memory (LSTM) and bi-directional LSTM, and generates an image by inputting it to the GAN. The previous text-to-image model uses basic embedding in text feature extraction, and it takes a long time to train because images are generated using several modules. Therefore, in this research, we propose a method of extracting features by using the attention mechanism, which has improved performance in the natural language processing field, for sentence embedding, and generating an image by inputting the extracted features into the GAN. As a result of the experiment, the inception score was higher than that of the model used in the previous study, and when judged with the naked eye, an image that expresses the features well in the input sentence was created. In addition, even when a long sentence is input, an image that expresses the sentence well was created.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1263-1271
• /
• 2010

Ontologies, the important implementation tools for semantic web, are widely used in various areas such as search, reasoning, and knowledge representation. Developing well-defined ontologies, however, requires a lot of resources in terms of time and materials. There have been efforts to construct ontologies automatically to overcome these problems. In this paper, ontologies are automatically constructed from the natural languages sentences directly. To do this, the analysis of morphemes and a sentence structure is performed at first. then, the program finds predicates inside the sentence and the predicates are transformed to the corresponding ontology predicates. For matching the corresponding ontology predicate from a predicate in the sentence, we develop the "predicate ontology". An experimental comparison between human ontology engineer and the program shows that the proposed system outperforms the human engineer in an accuracy.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.503-504
• /
• 2022

최근 자연어 처리 분야에서 언어 모델을 활용하여 문장 생성에 관한 연구가 이루어지고 있다. 기존 언어 모델을 활용하여 생성된 시나리오는 텍스트를 학습하여 활용하는 것 외에는 작가의 의도를 반영하는 것에 한계가 존재했고 문맥에 일관성 없는 모습을 보여주었다. 시나리오를 작성하는 것은 작가가 흐름을 주도하며 작업해야 하는 내용이다. 본 논문에서는 GPT-3 기반 언어 모델을 기반으로 다양한 시나리오 문장을 생성하여 작가가 선택하거나 원하는 문장을 직접 입력하는 등 작가의 의도에 부합하는 시나리오를 생성하는 보조 시스템을 제안한다. 본 연구를 통해 시나리오 생성을 포함한 문장 생성 분야의 보조 도구로 활용하여 작가의 의도를 반영하는 결과물을 생성하는 것을 목표로 한다.