• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.621-624
• /
• 2018

최근 지식 베이스의 발전과 함께 지식 베이스 기반의 질의 응답에 관한 연구가 많은 관심을 받고 있다. 특히 지식 베이스상의 여러 개의 사실이 필요한 복합 질의에 대한 처리의 중요성이 높아지고 있다. 그러나 기존 연구에서는 일반적인 지식을 묻는 질의 처리에만 집중하여, 그 외의 다른 유혀을 갖는 복합 질의에 대한 처리의 연구는 시작 단계에 머물러 있다. 이에 본 논문은 질의 유형 분류기를 활용한 지식 베이스 기반의 복합 질의 응답 시스템을 제안한다. 복합 질의 응답 시스템은 단순 질의를 포함하여 다양한 유형(일반형, 판정형, 비교형)을 갖는 복합 질의를 처리한다. 우리는 실험을 통해서 질의 유형 분류기가 복합 질의 응답 시스템의 정답률을 높임을 보였다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.604-607
• /
• 2022

본 논문은 사용자의 질의에 대한 답변을 제공하는 질의 응답 시스템에서, 제공하는 답변이 사용자의 질의에 대하여 문서에 근거하여 올바르게 대답하였는지 검증하는 QDR validator에 대해 기술한 논문이다. 본 논문의 과제는 문서에 대한 주장을 판별하는 자연어 추론(Natural Language inference, NLI)와 유사한 과제이지만, 문서(D)와 주장(R)을 포함하여 질의(Q)까지 총 3가지 종류의 입력을 받아 NLI 과제보다 난도가 높다. QDR validation 과제를 수행하기 위하여, 약 16,000 건 데이터를 생성하였으며, 다양한 입력 형식 실험 및 NLI 과제 데이터 추가 학습, 임계 값 조절 실험을 통해 최종 83.05% 우수한 성능을 기록하였다

• Journal of Digital Contents Society
• /
• v.18 no.2
• /
• pp.339-346
• /
• 2017

There is increasing interest in text analysis based on unstructured data such as articles and comments, questions and answers. This is because they can be used to identify, evaluate, predict, and recommend features from unstructured text data, which is the opinion of people. The same holds true for TEL, where the MOOC service has evolved to automate debating, questioning and answering services based on the teaching-learning support system in order to generate question topics and to automatically classify the topics relevant to new questions based on question and answer data accumulated in the system. Therefore, in this study, we propose topic modeling using LDA to automatically classify new query topics. The proposed method enables the generation of a dictionary of question topics and the automatic classification of topics relevant to new questions. Experimentation showed high automatic classification of over 0.7 in some queries. The more new queries were included in the various topics, the better the automatic classification results.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.31-39
• /
• 2023

The evolution of Natural Language Processing (NLP) and the rise of large language models (LLM) like ChatGPT have paved the way for specialized question-answering (QA) systems tailored to specific domains. This study outlines a system harnessing the power of LLM in conjunction with document search algorithms to interpret and address user inquiries using documents from the Korea Atomic Energy Research Institute (KAERI). Initially, the system refines multiple documents for optimized search and analysis, breaking the content into managable paragraphs suitable for the language model's processing. Each paragraph's content is converted into a vector via an embedding model and archived in a database. Upon receiving a user query, the system matches the extracted vectors from the question with the stored vectors, pinpointing the most pertinent content. The chosen paragraphs, combined with the user's query, are then processed by the language generation model to formulate a response. Tests encompassing a spectrum of questions verified the system's proficiency in discerning question intent, understanding diverse documents, and delivering rapid and precise answers.

• The Journal of the Korea Contents Association
• /
• v.7 no.1
• /
• pp.10-20
• /
• 2007

This paper describes a concept-based question analysis to analyze concept which is more important than keyword for the accurate answer extraction. Our idea is that we can extract correct answers from various paragraphs with different structures when we use well-defined concepts because concepts occurred in questions of same answer type are similar. That is, we will analyze the syntactic and semantic role of each word or phrase in a question in order to extract more relevant documents and more accurate answer in them. For each answer type, we define a concept frame which is composed of concepts commonly occurred in that type of questions and analyze user's question by filling a concept frame with a word or phrase. Empirical results show that our concept-based question analysis can extract more accurate answer than any other conventional approach. Also, concept-based approach has additional merits that it is language universal model, and can be combined with arbitrary conventional approaches.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.351-360
• /
• 2020

If various unreasonable safety standards are preemptively and effectively readjusted, the risk of accidents can be reduced. In this paper, we proposed a machine reading comprehension-based safety standard Q&A system to secure supporting technology for effective search and analysis of safety standards for integrated and systematic management of safety standards. The proposed model finds documents related to safety standard questions in the various laws and regulations, and then divides these documents into provisions. Only those provisions that are likely to contain the answer to the question are selected, and then the BERT-based machine reading comprehension model is used to find answers to questions related to safety standards. When the proposed safety standard Q&A system is applied to KorQuAD dataset, the performance of EM 40.42% and F1 55.34% are shown.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.429-437
• /
• 2022

Recently, the demand for a QA (Question Answering) system for human-machine communication has increased. Among the QA systems, a closed domain QA system that can handle spatial-related questions is called GeoQA. In this study, a new type of graph database, LPG (Labeled Property Graph) was used to overcome the limitations of the RDF (Resource Description Framework) based database, which was mainly used in the GeoQA field. In addition, GraphQL (Graph Query Language), an API-type query language, is introduced to address the fact that the LPG query language is not standardized and the GeoQA system may depend on specific products. In this study, database was built so that answers could be retrieved when spatial-related questions were entered. Each data was obtained from the national spatial information portal and local data open service. The spatial relationships between each spatial objects were calculated in advance and stored in edge form. The user's questions were first converted to GraphQL through FOL (First Order Logic) format and delivered to the database through the GraphQL server. The LPG used in the experiment is Neo4j, the graph database that currently has the highest market share, and some of the built-in functions and QGIS were used for spatial calculations. As a result of building the system, it was confirmed that the user's question could be transformed, processed through the Apollo GraphQL server, and an appropriate answer could be obtained from the database.

• The Journal of Korean Association of Computer Education
• /
• v.20 no.5
• /
• pp.79-88
• /
• 2017

To upgrade the subjective question grading, we need the syntactic-semantic analysis to analyze syntatic-semantic relation between words in answering. However, since the syntactic-semantic tree has structural and semantic relation between words, we can not apply the method calculating the similarity between vectors. This paper suggests the comparing system with syntactic-semantic tree which has structural and semantic relation between words. In this thesis, we suggest similarity calculation principles for comparing the trees and verify the principles through experiments. This system will help the subjective question grading by comparing the trees and be utilized in distinguishing similar documents.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.12
• /
• pp.35-48
• /
• 2017

Extracting temporal information from documents is becoming more important, because it can be used to various applications such as Question-Answering (QA) systems, Recommendation systems, or Information Retrieval (IR) systems. Most previous studies only focus on English documents, and they are not applicable to the other languages due to the inherent characteristics of languages. In this paper, we propose a new system, named ExoTime, designed to extract temporal information from Korean documents. The ExoTime adopts an external Knowledge Base (KB) in order to achieve better prediction performance, and it also applies a bagging method to the temporal relation prediction. We show that the effectiveness of the proposed approaches by empirical results using Korean TimeBank. The ExoTime system works as a part of ExoBrain that is an artificial intelligent QA system.

• Journal of Information Processing Systems
• /
• v.15 no.4
• /
• pp.931-956
• /
• 2019

Documents contain information that can be used for various applications, such as question answering (QA) system, information retrieval (IR) system, and recommendation system. To use the information, it is necessary to develop a method of extracting such information from the documents written in a form of natural language. There are several kinds of the information (e.g., temporal information, spatial information, semantic role information), where different kinds of information will be extracted with different methods. In this paper, the existing studies about the methods of extracting the temporal information are reported and several related issues are discussed. The issues are about the task boundary of the temporal information extraction, the history of the annotation languages and shared tasks, the research issues, the applications using the temporal information, and evaluation metrics. Although the history of the tasks of temporal information extraction is not long, there have been many studies that tried various methods. This paper gives which approach is known to be the better way of extracting a particular part of the temporal information, and also provides a future research direction.