• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.1
• /
• pp.89-98
• /
• 2016

In this paper, we proposed an automated areal feature matching method based on geometric similarity without user intervention and is applied into areal features of many-to-many relation, for confusion of spatial data-sets of different scale and updating cycle. Firstly, areal feature(node) that a value of inclusion function is more than 0.4 was connected as an edge in adjacency matrix and candidate corresponding areal features included many-to-many relation was identified by multiplication of adjacency matrix. For geometrical matching, these multiple candidates corresponding areal features were transformed into an aggregated polygon as a convex hull generated by a curve-fitting algorithm. Secondly, we defined matching criteria to measure geometrical quality, and these criteria were changed into normalized values, similarity, by similarity function. Next, shape similarity is defined as a weighted linear combination of these similarities and weights which are calculated by Criteria Importance Through Intercriteria Correlation(CRITIC) method. Finally, in training data, we identified Equal Error Rate(EER) which is trade-off value in a plot of precision versus recall for all threshold values(PR curve) as a threshold and decided if these candidate pairs are corresponding pairs or not. To the result of applying the proposed method in a digital topographic map and a base map of address system(KAIS), we confirmed that some many-to-many areal features were mis-detected in visual evaluation and precision, recall and F-Measure was highly 0.951, 0.906, 0.928, respectively in statistical evaluation. These means that accuracy of the automated matching between different spatial data-sets by the proposed method is highly. However, we should do a research on an inclusion function and a detail matching criterion to exactly quantify many-to-many areal features in future.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.1
• /
• pp.37-44
• /
• 2015

In traditional social network analysis, the betweenness centrality measure has been heavily used to identify the relative importance of nodes. Since the time complexity to calculate the betweenness centrality is very high, however, it is difficult to get it of each node in large-scale social network where there are so many nodes and edges. In our past study, we defined a new type of network, called the expanded ego network, which is built only with each node's local information, i.e., neighbor information of the node's neighbor nodes, and also defined a new measure, called the expanded ego betweenness centrality. In this paper, We propose algorithm that quickly computes expanded ego betweenness centrality by exploiting structural properties of expanded ego network. Through the experiment with virtual network used Barab$\acute{a}$si-Albert network model to represent the generic social network and facebook network to represent actual social network, We show that the node's importance rank based on the expanded ego betweenness centrality has high similarity with that the node's importance rank based on the existing betweenness centrality. We also show that the proposed algorithm computes the expanded ego betweenness centrality quickly than existing algorithm.

• Communications for Statistical Applications and Methods
• /
• v.19 no.4
• /
• pp.547-558
• /
• 2012

Social network analysis is a graphical technique to search the relationships and characteristics of nodes (people, companies, and organizations) and an important node for positioning a visualized social network figure; however, it is difficult to characterize nodes in a social network figure. Therefore, their relationships and characteristics could be presented through an application of correspondence analysis to an affiliation matrix that is a type of similarity matrix between nodes. In this study, we provide the relationships and characteristics around authors and keywords in the JKSS(Journal of the Korean Statistical Society) of the Korean Statistical Society through the use of social network analysis and correspondence analysis.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.48
• /
• pp.123-132
• /
• 1998

The group formation problem of the machine and part is a very important issue in the planning stage of cellular manufacturing systems. This paper investigates Self-Organizing Map(SOM) neural networks approach to machine-part grouping problem. We present a two-phase algorithm based on SOM for grouping parts and machines. SOM can learn from complex, multi-dimensional data and transform them into visually decipherable clusters. Output layer in SOM network is one-dimensional structure and the number of output node has been increased sufficiently to spread out the input vectors in the order of similarity. The proposed algorithm performs remarkably well in comparison with many other algorithms for the well-known problems shown in previous papers.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.11
• /
• pp.1536-1544
• /
• 1995

We propose a design scheme of a binary decision tree and apply it to the tire tread pattern recognition problem. In this scheme, a binary decision tree is constructed by using fuzzy C-means( FCM ) algorithm. All the available features are used while clustering. At each node, the best feature or feature subset among these available features is selected based on proposed similarity measure. The decision tree can be used for the classification of unknown patterns. The proposed design scheme is applied to the tire tread pattern recognition problem. The design procedure including feature extraction is described. Experimental results are given to show the usefulness of this scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.8
• /
• pp.3950-3964
• /
• 2017

With the increasing popularity of 3D technology such as 3D printing, 3D modeling, etc., there is a growing need to search for similar models on the internet. Matching non-rigid shapes has become an active research field in computer graphics. In this paper, we present an efficient and effective non-rigid model retrieval method based on topological structure and local volume. The integral geodesic distances are first calculated for each vertex on a mesh to construct the topological structure. Next, each node on the topological structure is assigned a local volume that is calculated using the shape diameter function (SDF). Finally, we utilize the Hungarian algorithm to measure similarity between two non-rigid models. Experimental results on the latest benchmark (SHREC' 15 Non-rigid 3D Shape Retrieval) demonstrate that our method works well compared to the state-of-the-art.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.441-452
• /
• 2021

Federated learning provides an efficient integrated model for distributed data, allowing the local training of different data. Meanwhile, the goal of multi-task learning is to simultaneously establish models for multiple related tasks, and to obtain the underlying main structure. However, traditional federated multi-task learning models not only have strict requirements for the data distribution, but also demand large amounts of calculation and have slow convergence, which hindered their promotion in many fields. In our work, we apply the rank constraint on weight vectors of the multi-task learning model to adaptively adjust the task's similarity learning, according to the distribution of federal node data. The proposed model has a general framework for solving optimal solutions, which can be used to deal with various data types. Experiments show that our model has achieved the best results in different dataset. Notably, our model can still obtain stable results in datasets with large distribution differences. In addition, compared with traditional federated multi-task learning models, our algorithm is able to converge on a local optimal solution within limited training iterations.

• Journal of the Korean Society for information Management
• /
• v.40 no.1
• /
• pp.121-148
• /
• 2023

The purpose of this study is to propose a scholarly paper recommendation system based on metadata attribute similarity with excellent performance. This study suggests a scholarly paper recommendation method that combines techniques from two sub-fields of Library and Information Science, namely metadata use in Information Organization and co-citation analysis, author bibliographic coupling, co-occurrence frequency, and cosine similarity in Bibliometrics. To conduct experiments, a total of 9,643 paper metadata related to "inequality" and "divide" were collected and refined to derive relative coordinate values between author, keyword, and title attributes using cosine similarity. The study then conducted experiments to select weight conditions and dimension numbers that resulted in a good performance. The results were presented and evaluated by users, and based on this, the study conducted discussions centered on the research questions through reference node and recommendation combination characteristic analysis, conjoint analysis, and results from comparative analysis. Overall, the study showed that the performance was excellent when author-related attributes were used alone or in combination with title-related attributes. If the technique proposed in this study is utilized and a wide range of samples are secured, it could help improve the performance of recommendation techniques not only in the field of literature recommendation in information services but also in various other fields in society.

• Design Convergence Study
• /
• v.14 no.4
• /
• pp.133-150
• /
• 2015

In this study, we proposed Interactive Visualization methods that can be analyzed relations between nations in various viewpoints such as period, issue using total 5211 of the UN General Assembly voting data.For this research, we devised a similarity matrix between nations and developed two visualization method based similarity matrix. The first one is Network Graph Visualization that can be showed relations between nations which participated in the vote of the UN General Assembly like Social Network Graph by year. and the second one is Proximity based Circular Visualization that can be analyzed relations between nations focus on one nation or Changes in voting patterns between nations according to time. This study have a great signification. that's because we proposed Proximity based Circular Visualization methods which merged Line and Circle Graph for network analysis that never tried from other cases of studies that utilize conventional voting data and made it. We also derived co-operatives of each visualization through conducting a comparative experiment for the two visualization. As a research result, we found that Proximity based Circular Visualization can be better analysis each node and Network Graph Visualization can be better analysis patterns for the nations.

• The KIPS Transactions:PartD
• /
• v.12D no.2 s.98
• /
• pp.233-246
• /
• 2005

To enhance the user response time of content-based retrieval service for multimedia information, several multi-dimensional index schemes have been proposed. M-tree, a well-known multidimensional index scheme is of metric space access method, and is based on the distance between objects in the metric space. However, since the overlap between index spaces of nodes might enlarge the number of nodes of M-tree accessed for query processing, the user response time for content-based multimedia information retrieval grows longer. In this paper, we propose a node split algorithm which is able to reduce the sire of overlap between index spaces of nodes in M-tree. In the proposed scheme, we choose a virtual center point as the routing object and entry redistribution as the postprocessing after node split in order to reduce the radius of index space of a node, and finally in order to reduce the overlap between the index spaces of routing nodes. From the experimental results, we can see the proposed split algorithm reduce the overlap between index space of nodes and finally enhance the user response time for similarity-based query processing.