• 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.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1991.10a
• /
• pp.285-292
• /
• 1991

객체-관계성(ER:entity-relationship)모델을 이용한 지식표현모델은 실세계를 객체(entity)들 또는 객체들의 집합들 사이의 서로간에 관계성(relationship)으로 나타낸다. 그러나 고정된 측면에서 표현되기 때문에 하나의 객체를 여러가지 측면에서 관찰할 수 없다. 반면 객체-측면(EA:entity-aspect)모델은 객체노드와 측면노드의 두가지 형을 갖는 노드들로 구성되어 측면에 따라 서로 다른 지식을 표현 할 수 있으므로 하나의 객체를 여러가지 측면에서 관찰할 수 있고, 그 세부적 계층구조를 나타낼 수 있는 장점이 있으나 너무 계층성을 강조하며, 객체간의 관계성을 나타낼 수가 없어 계층구조 속에 포함되지 않은 객체는 지식으로 표현 할 수 없어 실세계의 다양한 지식을 표현하는데 부자연스럽다. 따라서 본 논문에서는 객체-관계성(ER)모델의 관계성과 객체-측면(EA)모델의 측면성을 통합하여 객체 측면-관계성(EAR)모델을 제시하고, 이 모델에서 객체간의 관계성을 하나의 객체로 간주함으로 IS-A측면에 의하여 하위레벨로 계승할 수 있음을 보였다.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2780-2785
• /
• 1997

Although KBANN(knowledge-based artificial neural network) has been shown to be more effective than other machine learning algorithms, KBANN doesn't have the theory refinement capability because the topology of the network can't be altered dynamically. Although TopGen algorithm was proposed to extend the ability of KABNN in this respect, it also had some defects due to the connection of hidden nodes from all input nodes and the use of beam search. An algorithm, which could solve this TopGen's defects by adding the hidden nodes connected from only related input nodes and using hill-climbing search with backtracking, is proposed.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.7
• /
• pp.1773-1780
• /
• 1996

KBANN (knowledge-based artificial neural network) combining the symbolic approach and the numerical approach has been shown to be more effective than other machine learning models. However KBANN doesn't have the theory refinement ability because the topology of network can't be altered dynamically. Although TopGen was proposed to extend the ability of KABNN in this respect, it also had some defects due to the link-ing of hidden nodes to input nodes and the use of beam search. The algorithm which could solve this TopGen's defects, by adding the hidden nodes linked to next layer nodes and using hill-climbing search with backtracking, is designed.

• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.67-80
• /
• 2021

The node embedding technique for learning graph representation plays an important role in obtaining good quality results in graph mining. Until now, representative node embedding techniques have been studied for homogeneous graphs, and thus it is difficult to learn knowledge graphs with unique meanings for each edge. To resolve this problem, the conventional Triple2Vec technique builds an embedding model by learning a triple graph having a node pair and an edge of the knowledge graph as one node. However, the Triple2 Vec embedding model has limitations in improving performance because it calculates the relationship between triple nodes as a simple measure. Therefore, this paper proposes a feature extraction technique based on a graph convolutional neural network to improve the Triple2Vec embedding model. The proposed method extracts the neighborliness vector of the triple graph and learns the relationship between neighboring nodes for each node in the triple graph. We proves that the embedding model applying the proposed method is superior to the existing Triple2Vec model through category classification experiments using DBLP, DBpedia, and IMDB datasets.

• Journal of Intelligence and Information Systems
• /
• v.13 no.1
• /
• pp.59-79
• /
• 2007

P2P (Peer to Peer) techniques have been well applied to file sharing due to its cost-effectiveness and convenience. Dynamic network evolution is another good thing for P2P according to addition and deletion of nodes and change of files a node has. Our research proposes a P2P-based KMS (Knowledge Management System). Knowledge of enterprises spreads all over sub-organizations like oversea factories and sales departments and is changed in dynamic manner. P2P techniques are, therefore well matched with knowledge management domain. In order to increase search efficiency, we introduce social network theory into P2P-based KMS. Social network technique makes the most similar nodes (in KMS domain, nodes which has the most similar knowledge) its own neighbors, which makes eventually search efficiency increase. We developed our prototype system P2P-SN-KMS and evaluated by simulation.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.170-172
• /
• 2019

개체명 연결이란 주어진 문장에 출현한 단어를 위키피디아와 같은 지식 기반 상의 하나의 개체에 연결하는 것을 의미한다. 문장에 나타나는 개체들은 주로 동일한 주제를 가지게 되는데 본 논문에서는 이러한 특징을 활용하기 위해서 개체들을 그래프상의 노드로 표현하고, 그래프 신경망을 이용하여 주변 노드의 정보를 통해 노드 표상을 업데이트한다. 한국어 위키피디아 링크 데이터를 사용하여 실험을 진행한 결과 개발 셋에서 82.09%, 평가 셋에서 81.87%의 성능을 보였다.

• Information Systems Review
• /
• v.2 no.1
• /
• pp.1-11
• /
• 2000

• Journal of KIISE:Software and Applications
• /
• v.29 no.5
• /
• pp.289-294
• /
• 2002

It is known that restructuring feed-forward neural network affects generalization capability and efficiency of the network. In this paper, we introduce a new approach to restructure a neural network using abstraction of the hidden knowledge that the network has teamed. This method involves extracting local rules from non-input nodes and aggregation of the rules into global rule base. The extracted local rules are used for pruning unnecessary connections of local nodes and the aggregation eliminates any possible redundancies arid inconsistencies among local rule-based structures. Final network is generated by the global rule-based structure. Complexity of the final network is much reduced, compared to a fully-connected neural network and generalization capability is improved. Empirical results are also shown.

• Journal of the Korean Society for information Management
• /
• v.29 no.4
• /
• pp.123-142
• /
• 2012

To develop an intelligent search engine to help users retrieve information effectively, various methods, such as Semantic Web, have been used, An effective retrieval method of such methods uses ontology technology. In this paper, we built National R&D ontology after analyzing National R&D Information in NTIS and then implemented National R&D Knowledge Map to represent and retrieve information of the relationship between object and subject (project, human information, organization, research result) in R&D Ontology. In the National R&D Knowledge Map, center-node is the object selected by users, node is subject, subject's sub-node is user's favorite query in National R&D ontology after analyzing the relationship between object and subject. When a user selects sub-node, the system displays the results from inference engine after making query by SPARQL in National R&D ontology.