• Journal of Korean Society for Quality Management
• /
• v.30 no.2
• /
• pp.72-85
• /
• 2002

Advances in computer and sensor technology have made it possible to obtain superlarge manufacturing process data in real time, letting us extract meaningful information from these superlarge data sets. We propose a systematic data analysis procedure which field engineers can apply easily to manufacture quality products. The procedure consists of data cleaning and data analysis stages. Data cleaning stage is to construct a database suitable for statistical analysis from the original superlarge manufacturing process data. In the data analysis stage, we suggest a graphical easy-to-implement approach to extract practical information from the cleaned database. This study will help manufacturing companies to achieve six sigma quality.

• ETRI Journal
• /
• v.26 no.2
• /
• pp.149-160
• /
• 2004

In this paper, we present an interactive data broadcasting system compliant with the Advanced Television Systems Committee (ATSC) standards. The proposed system provides users not only with various data broadcasting services but also remote interactive services. For various data broadcasting services, we have adopted a synchronized data injector that calculates the transmission time of synchronized data accurately and multiplexes synchronized data with the data of an MPEG-2 audio-visual program according to the calculated transmission time. To support remote interactive services, we designed and implemented a return channel server connected on a bi-directional interaction channel. Test results show that the proposed system provides both an asynchronous and synchronized data broadcasting service and remote interactive service appropriately.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.2
• /
• pp.112-121
• /
• 2016

This paper focuses on competencies of data scientists and behavioral intention that affect big data analysis performance. This experiment examined nine core factors required by data scientists. In order to investigate this, we conducted a survey to gather data from 103 data scientists who participated in big data competition at Kaggle platform and used factor analysis and PLS-SEM for the analysis methods. The results show that some key competency factors have influential effect on the big data analysis performance. This study is to provide a new theoretical basis needed for relevant research by analyzing the structural relationship between the individual competencies and performance, and practically to identify the priorities of the core competencies that data scientists must have.

• Journal of Korea Multimedia Society
• /
• v.22 no.7
• /
• pp.749-758
• /
• 2019

This paper proposes a data augmentation algorithm to improve the performance of DNN(Deep Neural Network) based speech enhancement. Many deep learning models are exploring algorithms to maximize the performance in limited amount of data. The most commonly used algorithm is the data augmentation which is the technique artificially increases the amount of data. For the effective data augmentation algorithm, we used a formant enhancement method that assign the different weights to the formant frequencies. The DNN model which is trained using the proposed data augmentation algorithm was evaluated in various noise environments. The speech enhancement performance of the DNN model with the proposed data augmentation algorithm was compared with the algorithms which are the DNN model with the conventional data augmentation and without the data augmentation. As a result, the proposed data augmentation algorithm showed the higher speech enhancement performance than the other algorithms.

• International journal of advanced smart convergence
• /
• v.7 no.4
• /
• pp.10-18
• /
• 2018

For the information services supporting requests of data items from a great number of mobile clients, wireless data broadcasting is an effective way because it can accommodate any number of clients. In the wireless data broadcasting, various air indexing schemes and data scheduling schemes have been developed in order to enable the clients to access their desired data items efficiently. The broadcasting system needs a method to simulate newly designed air indexing and scheduling schemes of the system, and to evaluate the performance parameters of the schemes. In this paper, we design an expandable and efficient simulation framework for the wireless data broadcasting based on the partition of data ID space. The framework can adopt regular and irregular space partition and evaluate various performance parameters of the broadcasting system. We implement a testbed of the broadcasting system using the framework, that adopts IIP, GDI and EXP as its air indexing schemes. We simulate the system using the testbed and evaluate the performance parameters of the system. Thus, we show the efficiency and expandability of the designed and implemented framework.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.7
• /
• pp.39-44
• /
• 2019

Data integration technology is one of the key elements in building data warehouses or big data, and is used to combine data from multiple sources and provide an integrated view to users. Traditionally, the performance of data integration uses a handed-code approach or a tool-based approach that utilizes data integration tools such as ETL. There is a debate about which methods are efficient. This study is conducted to give practitioners preparing for a data integration project an insight into how to perform data integration. This paper examines the views of experts on the controversy over the adoption of ETL tools that have been on the agenda of the data integration area for over a decade.

• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.1-16
• /
• 2023

Synthetic data generation is generally used in performance evaluation and function tests in data-intensive applications, as well as in various areas of data analytics, such as privacy-preserving data publishing (PPDP) and statistical disclosure limit/control. A significant amount of research has been conducted on tools and languages for data generation. However, existing tools and languages have been developed for specific purposes and are unsuitable for other domains. In this article, we propose a regular expression-based data generation language (DGL) for flexible big data generation. To achieve a general-purpose and powerful DGL, we enhanced the standard regular expressions to support the data domain, type/format inference, sequence and random generation, probability distributions, and resource reference. To efficiently implement the proposed language, we propose caching techniques for both the intermediate and database queries. We evaluated the proposed improvement experimentally.

• Journal of information and communication convergence engineering
• /
• v.21 no.3
• /
• pp.216-224
• /
• 2023

Smart lighting systems have become increasingly popular in several public sectors because of trends toward urbanization and intelligent technologies. In this study, we designed and implemented a web application platform to explore and monitor data acquired from lighting devices at Thammasat University (Rangsit Campus, Thailand). The platform provides a convenient interface for administrative and operative staff to monitor, control, and collect data from sensors installed on campus in real time for creating geographically specific big data. Platform development focuses on both back- and front-end applications to allow a seamless process for recording and displaying data from interconnected devices. Responsible persons can interact with devices and acquire data effortlessly, minimizing workforce and human error. The collected data were analyzed using an exploratory data analysis process. Missing data behavior caused by system outages was also investigated.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.2
• /
• pp.240-247
• /
• 2023

In this paper, effective data management in industrial sites such as intelligent factories using time series data was studied. For effective management of time series data, variables considering the significance of the data were used, and hyper parameters calculated through LSTM AE were applied. We propose an optimized modeling considering the importance of each data section, and through this, outlier data of time series data can be efficiently processed. In the case of applying data significance and applying hyper parameters to which the research in this paper was applied, it was confirmed that the error rate was measured at 5.4%/4.8%/3.3%, and the significance of each data section and the significance of applying hyper parameters to optimize modeling were confirmed.

• Journal of Information Technology Services
• /
• v.23 no.2
• /
• pp.49-62
• /
• 2024

The purpose of this study is empirically to identify what factors affect the utilization of public data on the perspective of users. This study proposes four crucial factors including understandability, processing-convenience, linkage, and timeliness from the previous studies. As a result, understandability and linkage factors have significant impact on the utilization of public data and no different impact depending on the industry classification. The implication of this study is that it is important to provide sufficient information so that open data users can understand easily what kind of data it is, and to facilitate the linkage of open data with other data in order to activate the utilization of public data.