• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.11
• /
• pp.74-81
• /
• 1995

Several non-manifold data structures have been compared, which are radial-edge data structure, partial-face data structure, vertex-based data structure, and Yamaguchi's data structrue. All the entities in the data structures are classified into common entities and special entities. The entities are also classified as model entities, primitive entities bounding entities, and coupling entities. The four data structures for nonmanifold solid modelers are compared in terms of accessing efficiency, storage requirements, and inclusion of circulation. The results of comparison will serve as the basis to develope a nonmanifold modeler.

• Journal of Digital Convergence
• /
• v.12 no.12
• /
• pp.351-356
• /
• 2014

Generally, although securing the condition and location of container freights or normal freights, which load a fresh goods, has been a very important issue in the cold-chain system implementations, it has not gotten out of the traditional methods in the related business world yet. To solve this problem, we propose the designing method and architecture which can be used to implement a smart cold-chain monitoring automation systems. The proposed system architecture is based on the oneM2M standards, and it has 3 layers and entities, which can be implemented to S/W and H/W, network services layer and entity, common services layer and entity, application layer and entity. Based on this architecture, we will not only expect an innovative retrenchment of distribution cost, but also automatically secure the freight condition and location.

• The KIPS Transactions:PartB
• /
• v.10B no.2
• /
• pp.229-236
• /
• 2003

Named entity recognition is the process indispensable to question answering and information extraction systems. This paper presents an HMM based named entity (m) recognition method using the construction principles of compound words. In Korean, many named entities can be decomposed into more than one word. Moreover, there are contextual relationships among nouns in an NE, and among an NE and its surrounding words. In this paper, we classify words into a word as an NE in itself, a word in an NE, and/or a word adjacent to an n, and train an HMM based on NE-related word types and parts of speech. Proposed named entity recognition (NER) system uses trigram model of HMM for considering variable length of NEs. However, the trigram model of HMM has a serious data sparseness problem. In order to solve the problem, we use multi-level back-offs. Experimental results show that our NER system can achieve an F-measure of 87.6% in the economic articles.

• The Journal of Korean Association of Computer Education
• /
• v.18 no.6
• /
• pp.81-88
• /
• 2015

Along with advent of Web 3.0 era and advanced technologies of IoT(Internet of Things), massive amounts of information are generated. Reflecting this trend, this paper developed a semi-automatic construction tool for named entity dictionary based on active learning. Our proposed method chose error candidates to verify among the preliminary results using initial trained model and re-trained the model for correctly labeled data by user. We adopt active learning approach for minimizing human effort utilized metadata features of Wikipedia. Based on experimental results using our tool, we show that 68.6% errors were automatically corrected.

• Journal of Broadcast Engineering
• /
• v.21 no.2
• /
• pp.192-199
• /
• 2016

Mixed and augmented reality technique deals with mixing content between real world and virtual world containing augmented reality and augmented virtuality excluding of pure real and pure virtual world. In mixed and augmented reality, if a live actor and entity moving in real world can be embedded more naturally in 3D virtual world, various advanced applications such 3D tele-presence, 3D virtual experience education and etc can be serviced. Therefore, in this paper, we propose a standard model which is supporting to embed the live actor and entity into 3D virtual space, and to interact with each other. And also the natural embedding and interaction of live actor and entity can be performed based on the proposed model.

• The Journal of Information Systems
• /
• v.18 no.4
• /
• pp.59-86
• /
• 2009

Data modeling can be regarded as a series of processes to abstract real-world business concerns. The conceptual modeling phase is often regarded as the most difficult stage in the entire modeling process, because quite different conceptual models may be produced even for similar business domains based on users' varying requirements and the data modelers' diverse perceptions of the requirements. This implies that an object considered as an entity in one domain may be considered as an attribute in another, and vice versa. However, many traditional knowledge-based automated database design systems unfortunately fail to construct appropriate Entity-Relationship Diagrams(ERDs) for a given set of requirements due to the rigid assumption that an object should be classified as an entity if it has been classified as an entity in previous applications. In this paper, we propose an alternative automation system which can generate ERDs from business descriptions using association rule mining technique. Our system can be differentiated from the traditional ones in that our system can perform data modeling only based on business description written by domain workers, rather than relying on any kind of knowledge base. Since the proposed system can produce various versions of ERDs from the same business descriptions simultaneously, users can have the opportunity to choose one of the ERDs as being the most appropriate, based on their business environment and requirements. We performed a case study for personnel management in a university to evaluate the practicability of the proposed system This paper summarizes the result of it in the experiment section.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.7
• /
• pp.479-486
• /
• 2013

The data model that can be said as skeleton of the information system constitutes important 2 axles in the information system together with the process model. There is entity, properties, relation as key factors of the data model, and entity is the most fundamental factor in the data model, and thus total data model becomes vague if not deriving entity definitely. This study dealt with entity deduction only. Deducing methods of existing entity depended on experiences, task knowledge of designers and clear procedures were not suggested, so there were many difficulties in approaching them from beginners or unskilled persons. For giving helps in solving the problem, this study proposes entity- deducing procedures based on tasks that can derive entity with a systematic process at previously derived target businesses through suggested methods from advancing researches. And the study enabled proposing procedures on imaginary tasks to be applied, objecting to undergraduates who had not experiences on the data modeling, and then verified suggesting process through a similarity checking between best answers with deduced entity by students after taking impossible points of comparing existing methods with suggesting process into consideration. By doing so, deducing entity closely to the best answer was confirmed accordingly. Therefore, a fact could be confirmed that beginners were able to deduce entity closely to the best answer even if letting beginners who had not experiences on the data modeling be applied to unfamiliar tasks. Regarding researches on properties and relation deduction besides entity, this study leaves them to next time.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1998.04a
• /
• pp.191-198
• /
• 1998

An entity-based integrated design model can be used to organize and represent information and activities involved in design. The model involves a number of product and process entities. Product entities describe design information, and process entities describe design activities. The relationships among entities Includes organizational, interaction, and sequence relationships. The paper focuses interaction relationships among design information. The interaction relationships can be represented as constraints. Types of constraints includes demand constraints, dependency constraints, and interaction constraints. The paper describes dependency and Interaction constraints. The concepts of representing and processing dependency and interaction constraints in an entity-based integrated design model are presented.

• International journal of advanced smart convergence
• /
• v.12 no.2
• /
• pp.34-46
• /
• 2023

A growing number of large scale knowledge graphs raises several issues how knowledge graph data can be organized, discovered, and integrated efficiently. We present a novel semantic-based mashup platform for contents convergence which consists of acquisition, RDF storage, ontology learning, and mashup subsystems. This platform servers a basis for developing other more sophisticated applications required in the area of knowledge big data. Moreover, this paper proposes an entity matching method using graph convolutional network techniques as a preliminary work for automatic classification and discovery on knowledge big data. Using real DBP15K and SRPRS datasets, the performance of our method is compared with some existing entity matching methods. The experimental results show that the proposed method outperforms existing methods due to its ability to increase accuracy and reduce training time.

• Journal of the Korea Convergence Society
• /
• v.8 no.12
• /
• pp.55-62
• /
• 2017

The Named Entity Recognition system is a system that recognizes words or phrases with object names such as personal name (PS), place name (LC), and group name (OG) in the document as corresponding object names. Traditional approaches to named entity recognition include statistical-based models that learn models based on hand-crafted features. Recently, it has been proposed to construct the qualities expressing the sentence using models such as deep-learning based Recurrent Neural Networks (RNN) and long-short term memory (LSTM) to solve the problem of sequence labeling. In this research, to improve the performance of the Korean named entity recognition system, we used a hand-crafted feature, part-of-speech tagging information, and pre-built lexicon information to augment features for representing sentence. Experimental results show that the proposed method improves the performance of Korean named entity recognition system. The results of this study are presented through github for future collaborative research with researchers studying Korean Natural Language Processing (NLP) and named entity recognition system.