• 한국멀티미디어학회논문지
• /
• 제21권3호
• /
• pp.391-399
• /
• 2018

In this paper, we construct and compare various guitar chord classification systems using perceptron neural network and convolutional neural network without pre-processing other than Fourier transform to identify the optimal chord classification system. Conventional guitar chord classification schemes use, for better feature extraction, computationally demanding pre-processing techniques such as stochastic analysis employing a hidden markov model or an acoustic data filtering and hence are burdensome for real-time chord classifications. For this reason, we construct various perceptron neural networks and convolutional neural networks that use only Fourier tranform for data pre-processing and compare them with dataset obtained by playing an electric guitar. According to our comparison, convolutional neural networks provide optimal performance considering both chord classification acurracy and fast processing time. In particular, convolutional neural networks exhibit robust performance even when only small fraction of low frequency components of the data are used.

• 한국정보기술학회 영문논문지
• /
• 제10권2호
• /
• pp.139-151
• /
• 2020

Entity name recognition is a part of information extraction that extracts entity names from documents and classifies the types of extracted entity names. Entity name recognition technologies are widely used in natural language processing, such as information retrieval, machine translation, and query response systems. Various deep learning-based models exist to improve entity name recognition performance, but studies that compared and analyzed these models on Korean data are insufficient. In this paper, we compare and analyze the performance of CRF, LSTM-CRF, BiLSTM-CRF, and BERT, which are actively used to identify entity names using Korean data. Also, we compare and evaluate whether embedding models, which are variously used in recent natural language processing tasks, can affect the entity name recognition model's performance improvement. As a result of experiments on patent data and Korean corpus, it was confirmed that the BiLSTM-CRF using FastText method showed the highest performance.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -1
• /
• pp.329-332
• /
• 2000

Research interests on temporal data have been almost focused on data models. There has been relatively less research in the area of temporal data management. In this paper, we propose two data migration strategies based on time-segmented storage structure: the migration strategy by Time Granularity, the migration strategy by LST-GET. We describe the criterion for data migration and moving process. We simulated the performance of the migration strategy by Time Granularity in order to compare it with non-segmentation method. We compared and analyzed two data migration strategies for temporal data.

• 한국안전학회지
• /
• 제32권5호
• /
• pp.115-121
• /
• 2017

This paper is to introduce the main modeling assumptions and data structures associated with right-censored data to describe the successful methodological ideas for analyzing such a field-failure-data when components operating in different environments. The Kaplan - Meier method is the most popular method used for survival analysis. Together with the log-rank test, it may provide us with an opportunity to estimate survival probabilities and to compare survival between groups. An important advantage of the Kaplan - Meier curve is that the method can take into account some types of censored data, particularly right-censoring. The above non-parametric method was used to verify the equality of parts life used in different environments. After that, we performed the life distribution analysis using the parametric method. We simulated data from three distributions: exponential, normal, and Weibull. This allowed us to compare the results of the estimates to the known true values and to quantify the reliability indices. Here we used the Akaike information criterion to find a suitable life time distribution. If the Akaike information criterion is the smallest, the best model of failure data is presented. In this paper, no-nparametrics and parametrics methods are analyzed using R program which is a popular statistical program.

• Genomics & Informatics
• /
• 제5권3호
• /
• pp.129-132
• /
• 2007

arraylmpute is a software for exploratory analysis of missing data and imputation of missing values in microarray data. It also provides a comparative analysis of the imputed values obtained from various imputation methods. Thus, it allows the users to choose an appropriate imputation method for microarray data. It is built on R and provides a user-friendly graphical interface. Therefore, the users can easily use arraylmpute to explore, estimate missing data, and compare imputation methods for further analysis.

• Journal of the Korean Data and Information Science Society
• /
• 제19권4호
• /
• pp.1007-1018
• /
• 2008

The purpose of this case study is to demonstrate database-marketing management. First, we explore original variables for insurance customer's data, modify them if necessary, and go through variable selection process before analysis. Then, we develop churn prediction models using logistic regression, neural network and SVM analysis. We also compare these three data mining models in terms of misclassification rate.

• Wind and Structures
• /
• 제2권3호
• /
• pp.133-150
• /
• 1999

Significant costs to the public and private sectors due to recent extreme wind events have motivated the need for systematic post-hurricane damage data collection and analysis. Current post disaster data are collected by many different interested groups such as government agencies, voluntary disaster relief agencies, representatives of media companies, academicians and companies in the private sector. Each group has an interest in a particular type of data. However, members of each group collect data using different techniques. This disparity in data is not conducive to quantifying damage data and, therefore, inhibits the statistical and spatial description of damage and comparisons of damage among different extreme wind events. The data collection does not allow comparisons of data or results of analyses within a group and also prohibits comparison of damage data and information among different groups. Typically, analyses of data from a given event lead to different conclusion depending upon the definition of damage used by individual investigators and the type of data collected making it difficult for members of groups to compare the results of their analyses with a common language and basis. A formal method of data collection and analysis-within any single group-would allow comparisons to be made among different individuals, hazardous events and eventually among different groups, thus facilitating the management and reduction of damage due to future disaster. This research introduces a definition of damage to single family dwellings, and a common method of data collection and analysis suited for groups interested in regional characterization of damage. The current state-of-data is presented and a method for data collection is recommended based on these existing data collection methods. A fixed-scale damage index is proposed to consider the damage to a dwelling's feature. Finally, the damage index is applied to three dwellings damaged by Hurricane Iniki (1992). The damage index reflects the reduced functionality of a structure as a single family detached dwelling and provides a means to evaluate regional damage due to a single event or to compare damage due to events of different severity. Evaluation of the damage index and the data available support recommendation for future data collection efforts.

• 한국멀티미디어학회논문지
• /
• 제13권12호
• /
• pp.1814-1819
• /
• 2010

Since 1987, when statistical analyzing guide for thermal life test of Accelerated Life Test(ALT) was proposed as ANSI/IEEE Std 101, this guide has been used widely for many experiment data. Shim(2004) had done Monte Carlo simulation to compare life of two different systems or materials, based on statistic values obtained from ANSI/IEEE Std 101 data. In this study, a profile analysis is proposed for comparing life of two different systems or materials, and some examples using pre-existing data are given.

• Journal of the Korean Data and Information Science Society
• /
• 제18권4호
• /
• pp.1145-1155
• /
• 2007

In this paper, we consider the permutation tests for the multivariate data under the two-sample problem setting. We review some testing procedures, which are parametric and nonparametric and compare them with the permutation ones. Then we consider to try to apply the permutation tests to the multivariate data having the continuous and discrete components together by choosing some suitable combining function through the partial testing. Finally we discuss more aspects for the permutation tests as concluding remarks.

• 통상정보연구
• /
• 제12권4호
• /
• pp.405-427
• /
• 2010

This legal study is to compare the Electronic Transactions Act in Thailand(hereinafter 'ETA') with mainly other countries electronic transactions acts, such as UNCITRAL MLEC, UECIC, VETA, UCITA and Korea' Electronic Transactions Act The ETA is consisted of 6 chapters which included preamble and definitions. Each chapter's main point as follows. Preamble is related to the name, time of legal effect, scope and definitions. Chapter 1 is not only general principles of electronic transactions, required restriction in addition to specify the limit of application, documentation, evidential weight in reference to the data message, but also the conditions of offer and acceptance through data message, time and place of dispatch and receipt of data message, certification between origination and addressee. According to media-neutrality and the effectiveness security requirement of data message under the information system, legal certification is related to the exchange's declaration of intention, define about origination-addressee of data message. Chapter 2 is composed to provide expressly about the effectiveness security in electronic signature. Those contents are to compare the MLEC, UECIC and Electronic Transactions Act in Korea. Chapter 3 is related to legal definitions that present legal requirement about service relating electronic transaction which contents accept domestic law, the adequate requirement as eligibility, satisfied matter, self-reliance ratio of finance and other detail standard Chapter 4 is deal with the transaction which are public sector and those application requirements. And also this chapter are composed regulations about direct-indirect purpose of Thailand domestic electronic government.