• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.480-482
• /
• 2021

최근 다수의 문서를 고려해야하는 다중홉(multi-hop) 추론과 같은 복잡한 문제를 해결하기 위해 계층적 그래프 신경망기반 질의응답 시스템이 제안되었다. 계층적 그래프 신경망 기반 질의응답 시스템은 사람의 정확도를 뛰어넘었으나 제한된 문서를 통해 추론을 진행하기 때문에 문서에 충분한 정보가 없을 경우 추론에 실패할 가능성이 존재한다. 따라서 본 논문에서는 위 문제를 해결하기 위해 정보를 재탐색하고 기존의 그래프 정보와 병합하여 기존의 정보와 새로운 정보를 고려하여 재추론 할 수 있는 그래프 병합 기법을 제안한다. 제안하는 그래프 병합 기법은 사전에 정의된 규칙에 의해 수행되며 노드의 병합 및 연결을 통해 새로운 그래프를 도출한다. 새로운 그래프는 그래프 신경망을 통해 추론을 진행하여 기존 정보와 새로운 정보를 고려한 정답을 도출할 수 있다.

• Education of Primary School Mathematics
• /
• v.22 no.3
• /
• pp.149-166
• /
• 2019

The purpose of this study is to analyze the statistical literacy in elementary school students when they beginning inference. Picto-graphs provide statistical information and often data-related arguments they certainly qualify as objects for interpretation, for critical evaluation, and for discussion or communication of the conclusions presented. For research, the inference from pictograph task was designed and statistical literacy standards for evaluating the student's level was presented based on prior studies. Evaluating student's statistical literacy is meaningful in that it can check their current level. To know the student's current level can help them achieve a higher level of performance. The outcomes of this research indicate that pictograph can provide a basis for rich tasks displaying not only student's counting skills but also their appreciation of variation and uncertainty in prediction. Raising statistical thinking by students is an important goal in statistical education, and the experience of informal statistical reasoning can help with formal statistical reasoning that will be learned later. Therefore, the task about the inference from a pictograph, discussions on statistical learning of elementary school children are expected to present meaningful implications for statistical education.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.563-566
• /
• 2020

지식 그래프 기반의 질문 응답 문제는 자연어 질문에 대한 이해뿐만 아니라, 기반이 되는 지식 그래프상에서 올바른 답변을 찾기 위한 효과적인 추론 능력을 요구한다. 본 논문에서는 다중 홉 추론을 요구하는 복잡한 자연어 질문에 대해 연관 지식 그래프 위에서 답변 추론을 효과적으로 수행할 수 있는 심층 신경망 모델을 제안한다. 제안 모델에서는 지식 그래프상의 추론 과정에서 추른 경로를 명확히 하기 위한 노드의 양방향 특정 전파와 이웃 노드들 간의 맥락 정보까지 각 노드의 특정값에 반영할 수 있는, 표현력이 풍부한 쌍 선형 그래프 신경망 (BGNN)을 이용한다. 본 논문에서는 오픈 도메인의 지식 베이스 Freebase와 자연어 질문 응답 데이터 집합 WebQuestionsSP를 이용한 실험들을 통해, 제안 모델의 효과와 우수성을 확인하였다.

• Journal of KIISE
• /
• v.42 no.8
• /
• pp.998-1009
• /
• 2015

In recent years, there has been a growing interest in RDFS Inference to build a rich knowledge base. However, it is difficult to improve the inference performance with large data by using a single machine. Therefore, researchers are investigating the development of a RDFS inference engine for a distributed computing environment. However, the existing inference engines cannot process data in real-time, are difficult to implement, and are vulnerable to repetitive tasks. In order to overcome these problems, we propose a method to construct an in-memory distributed inference engine that uses a parallel graph structure. In general, the ontology based on a triple structure possesses a graph structure. Thus, it is intuitive to design a graph structure-based inference engine. Moreover, the RDFS inference rule can be implemented by utilizing the operator of the graph structure, and we can thus design the inference engine according to the graph structure, and not the structure of the data table. In this study, we evaluate the proposed inference engine by using the LUBM1000 and LUBM3000 data to test the speed of the inference. The results of our experiment indicate that the proposed in-memory distributed inference engine achieved a performance of about 10 times faster than an in-storage inference engine.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1561-1567
• /
• 2015

Abstract Semi-supervised learning is an area in machine learning that employs both labeled and unlabeled data in order to train a model and has the potential to improve prediction performance compared to supervised learning. Graph-based semi-supervised learning has recently come into focus with two phases: graph construction, which converts the input data into a graph, and label inference, which predicts the appropriate labels for unlabeled data using the constructed graph. The inference is based on the smoothness assumption feature of semi-supervised learning. In this study, we propose an enhanced label inference algorithm by incorporating the importance of each vertex. In addition, we prove the convergence of the suggested algorithm and verify its excellence.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.211-214
• /
• 2020

Abstract Meaning Representation(AMR)은 문장의 의미를 그래프 구조로 인코딩하여 표현하는 의미 형식표현으로 문장의 각 노드는 사건이나 개체를 취급하는 개념으로 취급하며 간선들은 이러한 개념들의 관계를 표현한다. AMR 파싱은 주어진 문장으로부터 AMR 그래프를 생성하는 자연어 처리 태스크이다. AMR 그래프의 각 개념은 추상 표현으로 문장 내의 토큰과 명시적으로 정렬되지 않는 어려움이 존재한다. 이러한 문제를 해결하기 위해 별도의 사전 학습된 정렬기를 이용하여 해결하거나 별도의 정렬기 없이 Sequence-to-Sequence 계열의 모델로 입력 문장으로부터 그래프의 노드를 생성하는 방식으로 연구되어 왔다. 본 논문에서는 문장의 입력 시퀀스와 부분 생성 그래프 사이에서 반복 추론을 통해 새로운 노드와 기존 노드와의 관계를 구성하여 점진적으로 그래프를 구성하는 모델을 한국어 AMR 데이터 셋에 적용하여 Smatch 점수 39.8%의 실험 결과를 얻었다.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.8
• /
• pp.243-250
• /
• 2020

Knowledge graph-based question answering not only requires deep understanding of the given natural language questions, but it also needs effective reasoning to find the correct answers on a large knowledge graph. In this paper, we propose a deep neural network model for effective reasoning on a knowledge graph, which can find correct answers to complex questions requiring multi-hop inference. The proposed model makes use of highly expressive bilinear graph neural network (BGNN), which can utilize context information between a pair of neighboring nodes, as well as allows bidirectional feature propagation between each entity node and one of its neighboring nodes on a knowledge graph. Performing experiments with an open-domain knowledge base (Freebase) and two natural-language question answering benchmark datasets(WebQuestionsSP and MetaQA), we demonstrate the effectiveness and performance of the proposed model.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.984-987
• /
• 2020

최근 오픈 도메인 자연어 질문 응답 분야에서는 폭넓은 다중 문서들을 토대로 다중 홉 추론과 동시에 서로 다른 수준의 여러 문제들을 한꺼번에 해결해야 하는 다중 작업 질문 응답에 관한 관심이 높다. 본 논문에서는 이러한 다중 홉 추론과 다중 작업을 요구하는 복잡 질문들에 효과적으로 응답하기 위해, 계층적 그래프 기반의 새로운 심층 신경망 모델을 제안한다. 제안 모델에서는 계층적 그래프와 그래프 신경망을 이용해 다중 문서들로부터 서로 다른 수준의 맥락 정보를 얻어낸 후, 이들을 활용하여 뒷받침 문장들, 답변 영역, 응답 유형 등을 동시에 구해야 하는 다중 작업 문제에 관한 답들을 예측해낸다. 본 논문에서는 오픈 도메인 자연어 질문 응답 데이터 집합인 HotpotQA를 이용한 실험들을 통해, 제안 모델의 긍정적 효과를 입증한다.

• Journal of Educational Research in Mathematics
• /
• v.27 no.1
• /
• pp.23-49
• /
• 2017

This study examined two current learning models for covariational reasoning(Carlson et al.(2002), Thompson, & Carlson(2017)), applied the models to teaching two $9^{th}$ grade students, and analyzed the results according to the types of graphs(a quantitative graph or qualitative graph). Results showed that the model of Thompson and Carlson(2017) was more useful than that of Carlson et al.(2002) in figuring out the students' levels in their quantitative graphing activities. Applying Carlson et al.(2002)'s model made it possible to classify levels of the students in their qualitative graphs. The results of this study suggest that not only quantitative understanding but also qualitative understanding is important in investigating students' covariational reasoning levels. The model of Thompson and Carlson(2017) reveals more various aspects in exploring students' levels of quantitative understanding, and the model of Carlson et al.(2002) revealing more of qualitative understanding.

• School Mathematics
• /
• v.18 no.3
• /
• pp.457-478
• /
• 2016

The purpose of this study is to investigate middle school students' understanding and development of function graphs. We collected the data from the teaching experiment with two middle school students who had not yet received instruction on linear function in school. The students participated in a 15-day teaching experiment(Steffe, & Thompson, 2000). Each teaching episode lasted one or two hours. The students initially focused on numerical values rather than the overall relationship between the variables in functional situations. This study described meaning, role of and students' responses for the given tasks, which revealed the students' understanding and development of function graphs. Especially we analyzed students' responses based on their methods to solve the tasks, reasoning that derived from those methods, and their solutions. The results indicate that their continuous reasoning played a significant role in their understanding of function graphs.