• Annual Conference on Human and Language Technology
• /
• 2011.10a
• /
• pp.143-147
• /
• 2011

본 논문에서는 한국어로 입력된 문장을 분석한 결과로부터 그에 해당하는 영어 대역문을 생성하는 과정에서, 어떻게 한국어의 보조용언을 영어 대역문에 반영 할 것인가를 다룬다. 특히 대화체 분야를 다루는 한영 자동번역 시스템에서는 한국어의 보조용언 생성이 대역문의 품질을 향상시키는데 중요한 위치를 차지하기 때문에, 한영 자동 번역에서의 자연스러운 영어 보조용언 생성을 위한 방법론을 제안한다. 첫째, 기존 패턴 기반 한영 자동 번역 엔진과 한국어 말뭉치를 형태소 분석한 결과를 살펴보고, 자연스러운 보조 용언 대역어 생성의 어려움을 살펴본다. 둘째, 자연스러운 보조용언 생성에 필요한 양상을 규칙화 한 지식을 기반으로 자연스러운 단일 보조용언 생성을 위한 방법을 제시한다. 셋째, 두 개 이상의 보조용언이 연속해서 나타나는 다중 보조용언의 생성 방법을 제시한다. 마지막으로, 실험과 결론을 통하여 본 논문이 제안하는 방법론을 사용했을 때, 자동 번역 엔진의 성능 평가 지표 중 하나인 BLEU와 NIST점수의 변화를 나타내봄으로 그 성능을 보인다.

• Journal of KIISE
• /
• v.42 no.4
• /
• pp.451-458
• /
• 2015

Previous multimodal learning methods focus on problem-solving aspects, such as image and video search and tagging, rather than on knowledge acquisition via content modeling. In this paper, we propose the Multimodal Concept Hierarchy (MuCH), which is a content modeling method that uses a cartoon video dataset and a character-based subtitle generation method from the learned model. The MuCH model has a multimodal hypernetwork layer, in which the patterns of the words and image patches are represented, and a concept layer, in which each concept variable is represented by a probability distribution of the words and the image patches. The model can learn the characteristics of the characters as concepts from the video subtitles and scene images by using a Bayesian learning method and can also generate character-based subtitles from the learned model if text queries are provided. As an experiment, the MuCH model learned concepts from 'Pororo' cartoon videos with a total of 268 minutes in length and generated character-based subtitles. Finally, we compare the results with those of other multimodal learning models. The Experimental results indicate that given the same text query, our model generates more accurate and more character-specific subtitles than other models.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.618-621
• /
• 2023

최근 거대언어모델(LLM)이 기계 번역 및 기계 독해를 포함한 다양한 문제들에서 높은 성능을 보이고 있다. 특히 프롬프트 기반의 대규모 언어 모델은 사고사슬 방식으로 적절한 프롬프팅을 통해 원하는 형식의 답변을 생성할 수 있으며 자연어 추론 단계에서도 높은 정확도를 보여주고 있다. 그러나 근본적으로 LLM의 매개변수에 질문에 관련된 지식이 없거나 최신 정보로 업데이트 되지 않은 경우 추론이 어렵다. 이를 해결하기 위해, 본 연구는 검색문서와 생성모델의 상호작용을 통해 답변하는 한국어 질의응답 모델을 제안한다. 검색이 어려운 경우 생성형 모델을 통해 질문과 관련된 문장을 생성하며, 이는 다시 검색모델과 추론 과정에서 활용된다. 추가로 "판단불가"라는 프롬프팅을 통해 모델이 답변할 수 없는 경우를 스스로 판단하게 한다. 본 연구결과에서 GPT3를 활용한 사고사슬 모델이 63.4의 F1 점수를 보여주며 생성형 모델과 검색모델의 융합이 적절한 프롬프팅을 통해 오픈-도메인 질의응답에서 성능의 향상을 보여준다.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.91-102
• /
• 2012

Computer aided language learning has become popular. But the level of automation of constructing a Korean learning assistant system is not so high because a practical language learning system needs large scale knowledge resources, which is very hard to acquire. In this paper, we propose a Korean learning assistant system that utilizes easily obtainable knowledge resources like a corpus, web documents and a lexicon. Our system has three modules - problem solving, pronunciation marker and writing assistant. Automatic problem generator uses a corpus and a lexicon to make problems with one correct answer and three distracters, then verifies their suitability by utilizing frequency information from web documents. We analyze pronunciation rules for a pronunciation marker and recommend appropriate words and sentences in real-time by using data extracted from a corpus. In experiment, we evaluate 400 automatically generated problems, which show 89.9% problem suitability and 64.9% example suitability.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.13-21
• /
• 2022

Recent deep learning-based research shows excellent performance in most natural language processing (NLP) fields with pre-trained language models. In particular, the auto-encoder-based language model proves its excellent performance and usefulness in various fields of Korean language understanding. However, the decoder-based Korean generative model even suffers from generating simple sentences. Also, there is few detailed research and data for the field of conversation where generative models are most commonly utilized. Therefore, this paper constructs multi-turn dialogue data for a Korean generative model. In addition, we compare and analyze the performance by improving the dialogue ability of the generative model through transfer learning. In addition, we propose a method of supplementing the insufficient dialogue generation ability of the model by extracting recommended response candidates from external knowledge information through a retrival model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.12
• /
• pp.2406-2412
• /
• 2007

Recently, there are many researches about natural language processing programs using artificial intelligence methods. But the researches mostly concentrate on Korean morphological analyses and there are few researches about application of the results of Korean morphological analyses. In this paper, we implemented a natural language conversation program that NPC in games can talk with users by natural language sentences using the results of morphological analyses and a rule-based inference method. We proposed representation and implementation methods of rules suitable for the processing of natural language conversation using NEO, a rule-based inference engine. In the experiment using rules and facts about knowledge of conversation for diet counselor NPC, we could verify that natural conversation results were produced.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.137-142
• /
• 2017

원격 지도 학습은 자동으로 매우 큰 코퍼스와 지식베이스 간의 주석 데이터를 생성하여 기계 학습에 필요한 학습 데이터를 사람의 손을 빌리지 않고 저렴한 비용으로 만들 수 있어, 많은 연구들이 관계 추출 문제를 해결하기 위해 원격 지도 학습 방법을 적용하고 있다. 그러나 기존 연구들에서는 모델 학습의 입력으로 사용되는 단어 임베딩에서 단어의 동형이의어 성질을 반영하지 못한다는 단점이 있다. 때문에 서로 다른 의미를 가진 동형이의어가 하나의 임베딩 값을 가지다 보니, 단어의 의미를 정확히 파악하지 못한채 관계 추출 모델을 학습한다고 볼 수 있다. 본 논문에서는 원격 지도 학습 기반 관계 추출 모델에 다중-어의 단어 임베딩을 적용한 모델을 제안한다. 다중-어의 단어 임베딩 학습을 위해 어의 중의성 해소 모듈을 활용하였으며, 관계 추출 모델은 문장 내 주요 특징을 효율적으로 파악하는 모델인 CNN과 PCNN을 활용하였다. 본 논문에서 제안하는 다중-어의 단어 임베딩 적용 관계추출 모델의 성능을 평가하기 위해 추가적으로 2가지 방식의 단어 임베딩을 학습하여 비교 평가를 수행하였고, 그 결과 어의 중의성 해소 모듈을 활용한 단어 임베딩을 활용하였을 때 관계추출 모델의 성능이 향상된 결과를 보였다.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.137-142
• /
• 2017

원격 지도 학습은 자동으로 매우 큰 코퍼스와 지식베이스 간의 주석 데이터를 생성하여 기계 학습에 필요한 학습 데이터를 사람의 손을 빌리지 않고 저렴한 비용으로 만들 수 있어, 많은 연구들이 관계 추출 문제를 해결하기 위해 원격 지도 학습 방법을 적용하고 있다. 그러나 기존 연구들에서는 모델 학습의 입력으로 사용되는 단어 임베딩에서 단어의 동형이의어 성질을 반영하지 못한다는 단점이 있다. 때문에 서로 다른 의미를 가진 동형이의어가 하나의 임베딩 값을 가지다 보니, 단어의 의미를 정확히 파악하지 못한 채 관계 추출 모델을 학습한다고 볼 수 있다. 본 논문에서는 원격 지도 학습 기반 관계 추출 모델에 다중-어의 단어 임베딩을 적용한 모델을 제안한다. 다중-어의 단어 임베딩 학습을 위해 어의 중의성 해소 모듈을 활용하였으며, 관계 추출 모델은 문장 내 주요 특징을 효율적으로 파악하는 모델인 CNN과 PCNN을 활용하였다. 본 논문에서 제안하는 다중-어의 단어 임베딩 적용 관계추출 모델의 성능을 평가하기 위해 추가적으로 2가지 방식의 단어 임베딩을 학습하여 비교 평가를 수행하였고, 그 결과 어의 중의성 해소 모듈을 활용한 단어 임베딩을 활용하였을 때 관계추출 모델의 성능이 향상된 결과를 보였다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.2
• /
• pp.207-216
• /
• 2024

As big data was built due to the 4th Industrial Revolution, personalized services increased rapidly. As a result, the amount of personal information collected from online services has increased, and concerns about users' personal information leakage and privacy infringement have increased. Online service providers provide privacy policies to address concerns about privacy infringement of users, but privacy policies are often misused due to the long and complex problem that it is difficult for users to directly identify risk items. Therefore, there is a need for a method that can automatically check whether the privacy policy is safe. However, the safety verification technique of the conventional blacklist and machine learning-based privacy policy has a problem that is difficult to expand or has low accessibility. In this paper, to solve the problem, we propose a safety verification technique for the privacy policy using the GPT-3.5 API, which is a generative artificial intelligence. Classification work can be performed evenin a new environment, and it shows the possibility that the general public without expertise can easily inspect the privacy policy. In the experiment, how accurately the blacklist-based privacy policy and the GPT-based privacy policy classify safe and unsafe sentences and the time spent on classification was measured. According to the experimental results, the proposed technique showed 10.34% higher accuracy on average than the conventional blacklist-based sentence safety verification technique.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.161-177
• /
• 2019

In this paper, we study the performance improvement of the answer extraction in Question-Answering system by using sentence dependency parsing result. The Question-Answering (QA) system consists of query analysis, which is a method of analyzing the user's query, and answer extraction, which is a method to extract appropriate answers in the document. And various studies have been conducted on two methods. In order to improve the performance of answer extraction, it is necessary to accurately reflect the grammatical information of sentences. In Korean, because word order structure is free and omission of sentence components is frequent, dependency parsing is a good way to analyze Korean syntax. Therefore, in this study, we improved the performance of the answer extraction by adding the features generated by dependency parsing analysis to the inputs of the answer extraction model (Bidirectional LSTM-CRF). The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. In this study, we compared the performance of the answer extraction model when inputting basic word features generated without the dependency parsing and the performance of the model when inputting the addition of the Eojeol tag feature and dependency graph embedding feature. Since dependency parsing is performed on a basic unit of an Eojeol, which is a component of sentences separated by a space, the tag information of the Eojeol can be obtained as a result of the dependency parsing. The Eojeol tag feature means the tag information of the Eojeol. The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. From the dependency parsing result, a graph is generated from the Eojeol to the node, the dependency between the Eojeol to the edge, and the Eojeol tag to the node label. In this process, an undirected graph is generated or a directed graph is generated according to whether or not the dependency relation direction is considered. To obtain the embedding of the graph, we used Graph2Vec, which is a method of finding the embedding of the graph by the subgraphs constituting a graph. We can specify the maximum path length between nodes in the process of finding subgraphs of a graph. If the maximum path length between nodes is 1, graph embedding is generated only by direct dependency between Eojeol, and graph embedding is generated including indirect dependencies as the maximum path length between nodes becomes larger. In the experiment, the maximum path length between nodes is adjusted differently from 1 to 3 depending on whether direction of dependency is considered or not, and the performance of answer extraction is measured. Experimental results show that both Eojeol tag feature and dependency graph embedding feature improve the performance of answer extraction. In particular, considering the direction of the dependency relation and extracting the dependency graph generated with the maximum path length of 1 in the subgraph extraction process in Graph2Vec as the input of the model, the highest answer extraction performance was shown. As a result of these experiments, we concluded that it is better to take into account the direction of dependence and to consider only the direct connection rather than the indirect dependence between the words. The significance of this study is as follows. First, we improved the performance of answer extraction by adding features using dependency parsing results, taking into account the characteristics of Korean, which is free of word order structure and omission of sentence components. Second, we generated feature of dependency parsing result by learning - based graph embedding method without defining the pattern of dependency between Eojeol. Future research directions are as follows. In this study, the features generated as a result of the dependency parsing are applied only to the answer extraction model in order to grasp the meaning. However, in the future, if the performance is confirmed by applying the features to various natural language processing models such as sentiment analysis or name entity recognition, the validity of the features can be verified more accurately.