• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.271-276
• /
• 2002

The test statistic in ANOVA tests has a single or doubly noncentral F distribution and the noncentral F distribution is applied to the calculation of the power functions of tests of general linear hypotheses. Although various approximations of noncentral F distribution are suggested, they are troublesome to compute. In this paper, the calculation of noncentral F distribution is applied to the neural network theory, to solve the computation problem. The neural network consists of the multi-layer perceptron structure and learning process has the algorithm of the backpropagation. Using fables and figs, comparisons are made between the results obtained by neural network theory and the Patnaik's values. Regarding of accuracy and calculation, the results by neural network are efficient than the Patnaik's values.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.2
• /
• pp.237-242
• /
• 2011

This paper proposes a detection algorithm for each section which detects the voiced section, unvoiced section, and the silence section at each frame using a multi-layer perceptron neural network. First, a power spectrum and FFT (fast Fourier transform) coefficients obtained by FFT are used as the input to the neural network for each frame, then the neural network is trained using these power spectrum and FFT coefficients. In this experiment, the performance of the proposed algorithm for detection of the voiced section, unvoiced section, and silence section was evaluated based on the detection rates using various speeches, which are degraded by white noise and used as the input data of the neural network. In this experiment, the detection rates were 92% or more for such speech and white noise when training data and evaluation data were the different.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.1
• /
• pp.167-176
• /
• 2010

Managing the growing number of data generated through various processes requires the aid of Database Management System (DBMS) to efficiently handle the huge amount of data. These data can be inserted into database m real time or in batch, that come from multiple sources, including those that are coming from inside and outside of a network. The insertion of large amount of data is commonly done through specific bulk loading or insertion function supplied by each individual DBMS. In this paper, we analyze and evaluate on handling data bulk loading for heterogeneous systems that is organised as multi-tiered architecture and compare the result of DBMS bulk loader against program insertion from a software development perspective. We propose a hybrid solution using staging database that can be easily deployed for enhancing bulk loading performance compared to insertion by application.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.6
• /
• pp.56-68
• /
• 2003

Difficult problems In implementing digital neural network hardware are the extension of synapses and the programmability for relocating neurons. In this paper, the structure of a new hardware is proposed for solving these problems. Our structure based on traditional SIMD can be dynamically and easily reconfigured connections of network without synthesizing and mapping original design for each use. Using additional modular processing unit the numbers of neurons find synapses increase. To show the extensibility of our structure, various models of neural networks : multi-layer perceptrons and Kohonen network are formed and tested. The performance comparison with software simulation shows its superiority in the aspects of performance and flexibility.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.12
• /
• pp.1865-1870
• /
• 2010

We develop an IPMC actuator with self-sensing behavior based on an accurate neural network model (NNM). The supplied voltage and voltage signals measured at two determined points on both sides of the IPMC sheet are used as inputs to the NNM. A CCD laser displacement sensor is installed in the rig for accurate measurement of the IPMC tip displacement that is used as the training output of the proposed NNM. Consequently, the NNM model is used to estimate the IPMC tip displacement; the NNM parameters are optimized by the collected input/output training data. The effectiveness of the model for the IPMC actuator is then verified by modeling results.

• Journal of Intelligence and Information Systems
• /
• v.10 no.2
• /
• pp.53-63
• /
• 2004

Recurrent-Cascade-Correlation(RCC) is a supervised teaming algorithm that automatically determines the size and topology of the network. RCC adds new hidden neurons one by one and creates a multi-layer structure in which each hidden layer has only one neuron. By second order RCC, new hidden neurons are added to only one hidden layer. These created neurons are not connected to each other. We present a generalization of the RCC Architecture by combining the standard RCC Architecture and the second order RCC Architecture. Whenever a hidden neuron has to be added, the new RCC teaming algorithm automatically determines whether the network topology grows vertically or horizontally. This new algorithm using sigmoid, tanh and new activation functions was tested with the morse-benchmark-problem. Therefore we recognized that the number of hidden neurons was decreased by the experiments of the RCC network generalization which used the activation functions.

• The Journal of the Korea Contents Association
• /
• v.16 no.1
• /
• pp.190-203
• /
• 2016

Based on the premise that a systematic career process was one of the essential elements of successful task performance both for individuals and the organization in the field of social welfare, this study set out to empirically analyze factors influencing the career commitment of social workers at a multidimensional level and provide practical implications for the directionality of career management on the basis of data with theoretical and statistical accuracy. For those purposes, the study collected individual and organizational characteristics data from 787 social workers at 46 agencies through a structured questionnaire and analyzed influential factors through the multilevel analysis technique by taking organizational effects into account. The analysis results show that explanations by the organization characteristics recorded significant 15% in the total variance of career commitment and that its influential factors included such significant variables as the protean career attitude, desire for growth, human network, and self-efficacy at the individual level and also the qualification compensation system at the organizational level. The study then proposed and discussed integrated practice strategies between individuals and agencies as the measures to promote career success through the activation of individual factors based on the consideration of organizational effects such as the application of an employee assistant program, provision of incentives to professional career development, and shift to a learning organization.

• Industry Promotion Research
• /
• v.2 no.2
• /
• pp.1-6
• /
• 2017

SCADA (Supervisory Control and Data Acquisition) System was first introduced to the Seoul Electricity Authority, which manages the transportation part among the three business fields that produce, transport and supply electric power in Korea. Has been using the data link with 1200 bps and 9600 bps in 5 protocols such as HARRIS 6000, BSC, HDLC, L & N and Toshiba by configuring 3 layers of EMS, SCADA and RTU computer equipments in 1: N radial form. This paper presents the OSI standard network packet flow, analyzing DataLink and Network Layer, and presents a KEPCO SCADA network model composed of X.25 high - speed communication network using 3 layers of network. We proposed a future SCADA communication structure that improved the current SCADA communication structure, defined the SCADA DB structure, introduced the concept of the remote SCADA gateway to the SCADA functional structure, applied the standard communication protocol, Multiplexing of surveillance and control in other local facilities and ensuring communication openness.

• Korean Journal of School Psychology
• /
• v.18 no.2
• /
• pp.175-195
• /
• 2021

Intimate friendships and a sense of belonging have positive effects on adolescent's psychological adaptation. Belongingness orientation is the motivation to belong. It is divided into growth orientation and deficit-reduction orientation, both of which have different effects on psychological adaptation and interpersonal characteristics. This study was conducted to determine how adolescents' belongingness orientation affected their psychological adaptation and friend networks. Students in their second year of middle school were surveyed both at the beginning and end of the spring semester. Friend networks were measured through network centrality analysis. Multilevel regression analysis produced three major results. The first major result was that the correlations between growth orientation and loneliness and between growth orientation and stress at the beginning of the spring semester was statistically significant even when friend network centrality was included in the analysis model, but the correlation between deficit-reduction orientation and loneliness and between deficit-reduction orientation and stress were not statistically significant. The second major result was that growth orientation significantly predicted friend network centrality at the end of the spring semester. This effect was significant even when friend network centrality at the beginning of the semester and psychological adaptation level at the end of the spring semester were added to the analysis model. The third major result was that the correlation between friend network centrality at the end of the semester and psychological adaptation level was statistically significant even when psychological adaptation levels at the beginning and the end of the semester were included in the analysis model. This study is meaningful in that it had a short-term longitudinal design and empirically demonstrated the relationship between belongingness orientation and psychological adaptation level of adolescents and between belongingness orientation and the development of friend networks. Lastly, we discussed limitations of this study and provided suggestions for future research.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.6
• /
• pp.35-47
• /
• 2020

Box-office prediction is important to movie stakeholders. It is necessary to accurately predict box-office and select important variables. In this paper, we propose a multivariate time series classification and important variable selection method to improve accuracy of predicting the box-office. As a research method, we collected daily data from KOBIS and NAVER for South Korean movies, selected important variables using Random Forest and predicted multivariate time series using Deep Learning. Based on the Korean screen quota system, Deep Learning was used to compare the accuracy of box-office predictions on the 73^rd day from movie release with the important variables and entire variables, and the results was tested whether they are statistically significant. As a Deep Learning model, Multi-Layer Perceptron, Fully Convolutional Neural Networks, and Residual Network were used. Among the Deep Learning models, the model using important variables and Residual Network had the highest prediction accuracy at 93%.