• Communications for Statistical Applications and Methods
• /
• v.9 no.1
• /
• pp.271-280
• /
• 2002

Recently, many companies have applied data mining techniques to promote competitive power in the field of their business market. In this study, we address how data mining, that is a technique to enable to discover knowledge from a deluge of data, Is used in an executed project in order to support decision making of an enterprise. Also, we develope scoring model on customer attrition probability for automobile-insurance company using data mining techniques. The development of scoring model in domestic insurance is given as an example concretely.

• Journal of Information Processing Systems
• /
• v.6 no.4
• /
• pp.521-536
• /
• 2010

Spatio-temporal patterns extracted from historical trajectories of moving objects reveal important knowledge about movement behavior for high quality LBS services. Existing approaches transform trajectories into sequences of location symbols and derive frequent subsequences by applying conventional sequential pattern mining algorithms. However, spatio-temporal correlations may be lost due to the inappropriate approximations of spatial and temporal properties. In this paper, we address the problem of mining spatio-temporal patterns from trajectory data. The inefficient description of temporal information decreases the mining efficiency and the interpretability of the patterns. We provide a formal statement of efficient representation of spatio-temporal movements and propose a new approach to discover spatio-temporal patterns in trajectory data. The proposed method first finds meaningful spatio-temporal regions and extracts frequent spatio-temporal patterns based on a prefix-projection approach from the sequences of these regions. We experimentally analyze that the proposed method improves mining performance and derives more intuitive patterns.

• Journal of information and communication convergence engineering
• /
• v.3 no.1
• /
• pp.49-55
• /
• 2005

This study is to suggest a visualization technique to display the relations of associated data in an optimal way when trying to display the whole data on a limited space by dealing with a large amount of data with linked information. For example, if you track an IP address through several steps and display the data on a screen, or if you visualize the human gene information on a 3-dimensional space, then it becomes even easier to understand the data flow in such cases. In order to simulate the technique given in this study, the given algorithm was applied to a large number of nodes made in a random fashion to optimize the data and we visually observed the result. According to the result, the technique given in this study is more efficient than any previous method in terms of visualization and utilizing space and allows to more easily understand the whole structure of a node because it consists of sub-groups.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.231-234
• /
• 2020

In the past few years, the industrial internet of things (IIoT) has received great attention in various industrial sectors which have potentially increased a high level of integrity, availability, and scalability. The increasing of IIoT is expected to create new smart industrial enterprises and build the next generation smart system. However existing IIoT systems rely on centralized servers that are vulnerable to a single point of failure and malicious attack, which exposes the data to security risks and storage. To address the above issues, blockchain is widely considered as a promising solution, which can build a secure and efficient environment for data storing, processing and sharing in IIoT. In this paper, we propose a decentralized, peer-to-peer platform for secure data storing in industrial IoT base on the ethereum blockchain. We exploit ethereum to ensure data security and reliability when smart devices store the data.

• The Korean Journal of Applied Statistics
• /
• v.37 no.1
• /
• pp.49-59
• /
• 2024

The popularity of data science, influenced by the trends from the United States around 2010, has significantly impacted the education of various statistics departments at domestic universities. However, it is challenging to find research papers in domestic academic journals that address the efficient teaching of data science topics in relation to computing environment. This article will discuss and propose the establishment of a suitable computing infrastructure for the education and research in statistics and data science departments in domestic universities.

• Journal of information and communication convergence engineering
• /
• v.22 no.3
• /
• pp.249-255
• /
• 2024

Advancements in deep learning have enhanced vision-based aggregate analysis. However, further development and studies have encountered challenges, particularly in acquiring large-scale datasets. Data collection is costly and time-consuming, posing a significant challenge in acquiring large datasets required for training neural networks. To address this issue, this study introduces a simulation that efficiently generates the necessary data and labels for training neural networks. We utilized a genetic algorithm (GA) to create optimized lists of aggregates based on the specified values of weight and particle size distribution for the aggregate sample. This enabled sample data collection without conducting sieving tests. Our evaluation of the proposed simulation and GA methodology revealed errors of 1.3% and 2.7 g for aggregate size distribution and weight, respectively. Furthermore, we assessed a segmentation model trained with data from the simulation, achieving a promising preliminary F1 score of 78.18 on the actual aggregate image.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.3
• /
• pp.54-68
• /
• 2011

As the prediction accuracy of conditional branch instruction is increased highly, the importance of prediction accuracy for unconditional branch instruction is also increased accordingly. Except the case of RAS(Return Address Stack) overflow, the prediction accuracy of function return address should be 100% theoretically. However, there exist some possibilities of miss-predictions for RAS return addresses, when miss-speculative execution paths are invalidated, in case of modern speculative microprocessor environments. In this paper, we propose the RAS rename technique to prevent RAS pollution, results in the reduction of RAS miss-prediction. We divide a RAS stack into a soft-stack and a hard-stack and we handle the instructions for speculative execution in the soft-stack. When some overwrites happen in the soft-stack, we move the soft-stack data into the hard-stack. In addition, we propose an enhanced version of RAS rename scheme. In simulation results, our solution provide 1/90 reduction of miss-prediction of function return address, results in up to 6.85% IPC improvement, compared to normal RAS method. Furthermore, it reduce miss-prediction ratio as 1/9, compared to previous technique.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.5
• /
• pp.1077-1088
• /
• 2013

High-dimensional data analysis arises from almost all scientific areas, evolving with development of computing skills, and has encouraged penalized estimations that play important roles in statistical learning. For the past years, various penalized estimations have been developed, and the least absolute shrinkage and selection operator (LASSO) proposed by Tibshirani (1996) has shown outstanding ability, earning the first place on the development of penalized estimation. In this paper, we first introduce a number of recent advances in high-dimensional data analysis using the LASSO. The topics include various statistical problems such as variable selection and grouped or structured variable selection under sparse high-dimensional linear regression models. Several unsupervised learning methods including inverse covariance matrix estimation are presented. In addition, we address further studies on new applications which may establish a guideline on how to use the LASSO for statistical challenges of high-dimensional data analysis.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.4
• /
• pp.33-41
• /
• 2020

Nowadays, distributed mobility management (DMM) approaches have been widely adopted to address the limitations of centralized architectural methods to support seamless data transmission schemes in wireless sensor networks. This paper deals with the end-to-end (E2E) integration of Network Mobility (NEMO) basic support protocol in distributed wireless sensor network systems in structural health and environmental monitoring of social overhead capital (SOC) public infrastructures such as bridges, national highways, tunnels, and railroads. The proposed scheme takes advantage of the features of both the NEMO basic support protocol and partially distributed network-based DMM framework in providing seamless data transmission and robust mobility support. The E2E seamless data transmission scheme allows mobile users to roam from fixed-point network access locations and mobile platforms (i.e., vehicles such as cars, buses, and trains) without disconnecting its current sessions (i.e., seamless handover).

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.5
• /
• pp.327-336
• /
• 2016

Prediction of customer failure rates plays an important role for establishing appropriate management policies and improving the profitability for industries. For these reasons, many LCD (Liquid crystal display) manufacturing industries have attempted to construct prediction models for customer failure rates. However, most traditional models are based on the parametric approaches requiring the assumption that the data follow a certain probability distribution. To address the limitation posed by the distributional assumption underpinning traditional models, we propose using parameter-free data mining models for predicting customer failure rates. In addition, we use various information associated with product attributes and field return for more comprehensive analysis. The effectiveness and applicability of the proposed method were demonstrated with a real dataset from one of the leading LCD companies in South Korea.