• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.23-28
• /
• 2020

A comment system is essential for communication on the Internet. However, there are also malicious comments such as inappropriate expression of others by exploiting anonymity online. In order to protect users from malicious comments, classification of malicious / normal comments is necessary, and this can be implemented as text classification. Text classification is one of the important topics in natural language processing, and studies using pre-trained models such as BERT and graph structures such as GCN and GAT have been actively conducted. In this study, we implemented a comment classification system using BERT, GCN, and GAT for actual published comments and compared the performance. In this study, the system using the graph-based model showed higher performance than the BERT.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1114-1117
• /
• 1993

The paper presents an effective method for finding the propagation structure of the real origin of a system malfunction. It uses a combined system model consisting of Structural Model (SM) in the form of Fuzzy Directed Graph and Behavior Model (BM) as a set of Fuzzy Relational Equations $A\;{\circ}\;R\;=\;B$. Here a specially proposed fuzzy inference technique is checked and investigated. Finally a test example for fault diagnosis of an industrial system is given and analyzed.

• The Journal of the Korea Contents Association
• /
• v.22 no.8
• /
• pp.33-43
• /
• 2022

Labeled graphs are used to represent entities, their relationships, and their structures in real data such as knowledge graphs and protein interactions. With the rapid development of IT and the explosive increase in data, there has been a need for a subgraph matching technology to provide information that the user is interested in. In this paper, we propose an approximate Top-k labeled subgraph matching scheme that considers the semantic similarity of labels and the difference in graph structure. The proposed scheme utilizes a learning model using FastText in order to consider the semantic similarity of a label. In addition, the label similarity graph(LSG) is used for approximate subgraph matching by calculating similarity values between labels in advance. Through the LSG, we can resolve the limitations of the existing schemes that subgraph expansion is possible only if the labels match exactly. It supports structural similarity for a query graph by performing searches up to 2-hop. Based on the similarity value, we provide k subgraph matching results. We conduct various performance evaluations in order to show the superiority of the proposed scheme.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.785-791
• /
• 2005

Reconfigurability of machine tools is one of the critical factors to realize the responsive manufacturing systems to satisfy the mass-customization production. This paper presents the methods to model and simulate the machine tools on Internet in response to change in the machining requirements. Specifically, a set of module combination rules and a modeling method of the structure of machine tools using connectivity graph are developed. In response to the user requirements, kinematic relations and structures of machine tools can be derived using the module combination rules and connectivity graph relationships. Internet-based simulator of machine tools is implemented and presented. The developed machine tool simulator can be used to verify the structure of machine tools derived from the user requirements.

• Journal of Korean Institute of Industrial Engineers
• /
• v.32 no.2
• /
• pp.82-90
• /
• 2006

Reconfigurability of machine tools is one of the critical factors to realize the responsive manufacturing systems to satisfy the mass customization production. This paper presents the methods to model the reconfigurable machine tools (RMTs) on Internet in response to change in the machining requirements. Specifically, the structure and motion model of machine tools using module combination rules and connectivity graph are developed. And we developed Internet-based simulator for designing of RMTs (ISD-RMT). In response to the user requirements, various structures of RMTs can be derived using the module combination rules and connectivity graph relationships. In addition, the user can verify generating structures through the control and simulation procedures.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.11 no.1
• /
• pp.154-164
• /
• 2019

A rotor dynamic analysis is mandatory for stability and design optimization of submerged propellers and turbines. An accurate simulation requires a proper consideration of fluid-induced reaction forces. This paper presents a bi-directional coupling of a bond graph method solver and an unsteady vortex lattice method solver where the former is used to model the rotor dynamics of the power train and the latter is used to predict transient hydrodynamic forces. Due to solver coupling, determination of hydrodynamic coefficients is obsolete and added mass effects are considered automatically. Additionally, power grid and structural faults like grid fluctuations, eccentricity or failure could be investigated using the same model. In this research work a fast, time resolved dynamic simulation of the complete power train is conducted. As an example, the rotor dynamics of a tidal stream turbine is investigated under two inflow conditions: I - shear flow, II - shear flow + water waves.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.5
• /
• pp.1464-1473
• /
• 2000

In this paper, we analyze the cycle property of the (n,k)-star graph that has an attention as an alternative interconnection network topology in recent years. Based on the graph-theoretic properties in (n,k)-star graphs, we show the pancyclic property of the graph and also present the corresponding algorithm. Based on the recursive structure of the graph, we present such top-down approach that the resulting cycle can be constructed by applying series of "dimension expansion" operations to a kind of cycles consisting of sub-graphs. This processing naturally leads to such property that the resulting cycles tend to be integrated compactly within some minimal subset of sub-graphs, and also means its applicability of another classes of the disjoint-style cycle problems. This result means not only the graph-theoretic contribution of analyzing the pancyclic property in the underlying graph model but also the parallel processing applications such a as message routing or resource allocation and scheduling in the multi-computer system with the corresponding interconnection network.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.22 no.2
• /
• pp.399-407
• /
• 1998

This paper deals with a method for finding the physical structure of dynamic systems which shows reasonable response to a given specifiation. The method uses only four basic models of bond graph prototypes which have been originally proposed by the authors as a general model for dynamic systems, and then makes its procedure highly physical in the sense that it can synthesize a dynamic system through the structural transformation directly on bond graph models without any mathematical manipulation. Also, it is shown that this method has further advantages in optimizing parameters for an existing system rather than developing design concepts for a new device, the latter being more suitable using the so-called analytical synthesis method introduced by Park and Redfield. One example serves to trace the outlines of the direct synthesis method proposed in this paper for dynamic systems in terms of bond graph prototypes.

• Journal of KIISE:Databases
• /
• v.29 no.4
• /
• pp.274-284
• /
• 2002

Queries on XML are based on paths in the data graph, which is represented as an edge labeled graph model. All proposed query languages for XML express queries using regular expressions to traverse arbitrary paths in the data graph. A meaningful query usually has several regular path expressions in it, but much of recent research is more concerned with optimizing a single path expression. In this paper, we present an efficient technique to process multiple path expressions in a query. We developed a data structure named as the path identifier(PID) to identify whether two given nodes lie on the fame path in the data graph or not, and utilized the PID for efficient processing of multiple path expressions. We implement our technique and present preliminary performance results.

• Journal of Information Science Theory and Practice
• /
• v.11 no.3
• /
• pp.43-57
• /
• 2023

Government codes are created and utilized to streamline and standardize government administrative procedures. They are generally employed in government information systems. Because they are included in open datasets of public data, users must be able to understand them. However, information that can be used to comprehend administrative code is lost during the process of releasing data in the government system, making it difficult for data consumers to grasp the code and limiting the connection or convergence of different datasets that use the same code.This study proposes a way to employ the administrative code produced by the Korean government as a standard in a public data environment on a regular basis. Because consumers of public data are barred from accessing government systems, a means of universal access to administrative code is required. An ontology model is used to represent the administrative code's data structure and meaning, and the full administrative code is built as a knowledge graph. The knowledge graph thus created is used to assess the accuracy and connection of administrative codes in public data. The method proposed in this study has the potential to increase the quality of coded information in public data as well as data connectivity.