• KIISE Transactions on Computing Practices
• /
• v.23 no.11
• /
• pp.632-640
• /
• 2017

Although solar insolation is the weather factor with the greatest influence on power generation in photovoltaic systems, the Meterological Agency does not provide solar insolation data for future dates. Therefore, it is essential to research prediction methods for solar insolation to efficiently manage photovoltaic systems. In this study, we propose a Dynamic Piecewise Prediction Model that can be used to predict solar insolation values for future dates based on information from the weather forecast. To improve the predictive accuracy, we dynamically divide the entire data set based on the sun altitude and cloudiness at the time of prediction. The Dynamic Piecewise Prediction Model is developed by applying a polynomial linear regression algorithm on the divided data set. To verify the performance of our proposed model, we compared our model to previous approaches. The result of the comparison shows that the proposed model is superior to previous approaches in that it produces a lower prediction error.

• Computers and Concrete
• /
• v.14 no.5
• /
• pp.577-597
• /
• 2014

This paper aims to develop a prediction model for the hardened properties of waste LCD glass that is used in concrete by analyzing a series of laboratory test results, which were obtained in our previous study. We also summarized the testing results of the hardened properties of a variety of waste LCD glass concretes and discussed the effect of factors such as the water-binder ratio (w/b), waste glass content (G) and age (t) on the concrete compressive strength, flexural strength and ultrasonic pulse velocity. This study also applied a hyperbolic function, an exponential function and a power function in a non-linear regression analysis of multiple variables and established the prediction model that could consider the effect of the water-binder ratio (w/b), waste glass content (G) and age (t) on the concrete compressive strength, flexural strength and ultrasonic pulse velocity. Compared with the testing results, the statistical analysis shows that the coefficient of determination $R^2$ and the mean absolute percentage error (MAPE) were 0.93-0.96 and 5.4-8.4% for the compressive strength, 0.83-0.89 and 8.9-12.2% for the flexural strength and 0.87-0.89 and 1.8-2.2% for the ultrasonic pulse velocity, respectively. The proposed models are highly accurate in predicting the compressive strength, flexural strength and ultrasonic pulse velocity of waste LCD glass concrete. However, with other ranges of mixture parameters, the predicted models must be further studied.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.25 no.3
• /
• pp.191-214
• /
• 2014

The goal of this study is an exploratory statistical analysis of the university and college library statistics in the Academic Information Statistics System(rinfo.kr) governed of Korean Education and Research Information Service(KERIS) with performance measures based on sustainability. For the goal, we adopt a preprocessing method to develop change-rate variables by considering preceding predictive elements and succeeding performance elements, and to control external factors, such as size and socioeconomic factors. Then we execute a series of factor analyses and multiple linear regression analyses. 13 factors are extracted by the factor analyses and some sets of significant variables affecting the performance measures are identified through the regression analyses. Based on the results, we discuss the problem of out-lier and low correlation between variables. A suggestion for developing new variables is also discussed based on low effect sizes of the developed regression models. We hope that this study contributes to diffuse discussions on statistics system, evaluation, and further library management based on sustainability.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.1
• /
• pp.87-91
• /
• 2005

The cascade-correlation (CC) learning algorithm of Fahlman and Lebiere is one of the most influential constructive algorithm in a neural network. Cascading the hidden neurons results in a network that can represent very strong nonlinearities. Although this power is in principle useful, it can be a disadvantage if such strong nonlinearity is not required to solve the problem. 3 models are presented and compared empirically. All of them are based on valiants of the cascade architecture and output neurons weights training of the CC algorithm. Empirical results indicate the followings： (1) In the pattern classification, the model that train only new hidden neuron to output layer connection weights shows the best predictive ability; (2) In the function approximation, the model that removed input-output connection and used sigmoid-linear activation function is better predictability than CasCor algorithm.

• Journal of Korean society of Dental Hygiene
• /
• v.16 no.2
• /
• pp.313-319
• /
• 2016

Objectives: The objective of the study was to infection control by analyzing the influence of working environment on infection control activities in the dental hygienists. Methods: The cross-sectional study was based on a survey on a total of 377 dental hygienists working in dental settings. Multiple linear regression analysis was performed to examine the relationship of general characteristics and infection control activities. All statistical analyses were performed using the SPSS for Windows version 20.0, and p<0.05 was considered to be significant. Results: Predictive powers(=Adjusted $R^2$ of the investigated factors such as operation room, sterilization disinfection laundry, dental unit waterline, staff individual, infection control system, personal protective equipment, medical waste, hand hygiene, oral surgical procedures, clinical contact surfaces were adjusted $R^2=0.394$, 0.306, 0.277, 0.244, 0.241, 0.177, 0.165,, 0.154, 0.134, 0.124 respectively. Conclusions: In order to enhance infection control activities, the program development and implementation for the aseptic procedure is very important. The program should include periodic reinforcement of infection control education and regular monitoring of infection control activities.

• Journal of Korean Society for Atmospheric Environment
• /
• v.32 no.1
• /
• pp.9-20
• /
• 2016

Ozone has been a problem in big cities. That is secondary air pollutant produced by nitrogen oxide and VOCs in the atmosphere. In order to solve this, the first to be the analysis of the $NO_x$ and VOCs. The main source of nitrogen oxide is the road mobile. Industrial sources in Seoul are particularly low, and mobile traffics on roads are large, so 45% of total $NO_x$ are estimated that road mobile emissions in Seoul. Thus, it is necessary to clarify the relation with the activity of road mobile source and $NO_x$ concentration. In this study, we analyzed the 4 locations with roadside automatic monitoring systems in their center. The V.K.T. calculating areas are set in circles with 50 meter spacing, 50 meter to 500 meter from their center. We assumed the total V.K.T. in the set radius affect the $NO_x$ concentration in the center. We used the hourly $NO_x$ concentrations data for the 4 observation points in July for the interference of the other sources are minimized. We used the intersection traffic survey data of all direction for construction of the V.K.T. data, the mobile activities on the roads. ArcGIS application was used for calculating the length of roads in the set radius. The V.K.T. data are multiplied by segment traffic volume and length of roads. As a result, the $NO_x$ concentration can be expressed as linear function formula for V.K.T. with high predictive power. Moreover we separated background concentration and concentrations due to road mobile source. These results can be used for forecasting the effect of traffic demand management plan.

• The Journal of the Acoustical Society of Korea
• /
• v.11 no.3
• /
• pp.61-66
• /
• 1992

There have been many researches that uses neural network models for automatic speech recognition, but the main trend was finding the neural network models and learning rules appropriate to automatic speech recognition. However, the choice of the input speech parameter for the neural network as well as neural network model itself is a very important factor for the improvement of performance of the automatic speech recognition system using neural network. In this paper we select 6 speech parameters from surveys of the speech recognition papers which uses neural networks, and analyze the performance for the same data and the same neural network model. We use 8 sets of 9 Korean plosives and 18 sets of 8 Korean vowels. We use recurrent neural network and compare the performance of the 6 speech parameters while the number of nodes is constant. The delta cepstrum of linear predictive coefficients showed best result and the recognition rates are 95.1% for the vowels and 100.0% for plosives.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.7
• /
• pp.1370-1376
• /
• 2012

This paper proposes a gender recognition algorithm which classifies a male or female speaker. In this paper, characteristic vectors for the male and female speaker are analyzed, and recognition experiments for the proposed gender recognition by a neural network are performed using these characteristic vectors for the male and female. Input characteristic vectors of the proposed neural network are 10 LPC (Linear Predictive Coding) cepstrum coefficients, 12 LPC cepstrum coefficients, 12 FFT (Fast Fourier Transform) cepstrum coefficients and 1 RMS (Root Mean Square), and 12 LPC cepstrum coefficients and 8 FFT spectrum. The proposed neural network trained by 20-20-2 network are especially used in this experiment, using 12 LPC cepstrum coefficients and 8 FFT spectrum. From the experiment results, the average recognition rates obtained by the gender recognition algorithm is 99.8% for the male speaker and 96.5% for the female speaker.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.5
• /
• pp.410-418
• /
• 2009

Based on a linear potential theory, the boundary element method(BEM) is developed and applied to analyze the hydrodynamic forces and the motion responses of a floating body with a partially reflective sidewall. The hydrodynamic forces (added mass and damping coefficients) are dependent on not only the submergence of a floating body and the reflection of a sidewall, but also the gap between body and sidewall. In particular, the partial reflection of a sidewall plays an importance role in the motion responses of a floating body at resonant frequencies. It reduces the resonant peaks caused by resonance phenomenon due to the wave trapping in an enclosed fluid domain between body and sidewall. Developed predictive tools can be used to assess the motion performance of a floating body for various combinations of configuration of a floating body, wave heading, sidewall properties, and wave characteristics and applied to supply the basic informations for the harbour design considering the motion characteristics of a moored ship.