• Journal of the Korea Society of Computer and Information
• /
• v.27 no.10
• /
• pp.59-66
• /
• 2022

This research conducted a study classifying gender of speakers by analyzing feature vectors extracted from the voice data. The study provides convenience in automatically recognizing gender of customers without manual classification process when they request any service via voice such as phone call. Furthermore, it is significant that this study can analyze frequently requested services for each gender after gender classification using a learning model and offer customized recommendation services according to the analysis. Based on the voice data of males and females excluding blank spaces, the study extracts feature vectors from each data using MFCC(Mel Frequency Cepstral Coefficient) and utilizes SVM(Support Vector Machine) models to conduct machine learning. As a result of gender classification of voice data using a learning model, the gender recognition rate was 94%.

• Proceedings of the KIEE Conference
• /
• 1996.07a
• /
• pp.591-593
• /
• 1996

In order to implement the direct vector control type sensorless vector control, the rotor flux and the angular speed of the rotor can only be estimated through the measurement of the stationary voltage and current states. To estimate the rotor flux, the use of the rotor flux observer(RFO) has been proposed. It is known that the RFO is relatively insensitive to parameter variations. Using the rotor flux value obtained from the RFO, the rotor flux vector can be estimated. The angular speed of the rotor is estimated by the difference between the synchronous angular speed and the slip angular speed, both of which are derived from the rotor flux vector. However unwanted high order frequency waves become incorporated into the synchronous angular speed during calculations. Thus we propose the use of digital filters that will eliminate these high frequency waves. We have demonstrated through computer simulations that the use of filters results in stable system activity over a wide speed range and good response to load variations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.1
• /
• pp.323-343
• /
• 2020

With the rapid development of computer and communication techniques, copyright protection of vector geographic data has attracted considerable research attention because of the high cost of such data. A novel adaptive watermark detection algorithm is proposed for vector geographic data that can be used to qualitatively analyze the robustness of watermarks against data addition attacks. First, a watermark was embedded into the vertex coordinates based on coordinate mapping and quantization. Second, the adaptive watermark detection model, which is capable of calculating the detection threshold, false positive error (FPE) and false negative error (FNE), was established, and the characteristics of the adaptive watermark detection algorithm were analyzed. Finally, experiments were conducted on several real-world vector maps to show the usability and robustness of the proposed algorithm.

• Journal of Korea Multimedia Society
• /
• v.18 no.3
• /
• pp.298-311
• /
• 2015

Vector data represents a map by its coordinate and Raster data represents a map by its pixel. Since these data types have very large data size, data compression procedure is a compulsory process. This paper compare the results from three different methodologies; GIS (Geographic Information System) vector map data compression using DP(Douglas-Peucker) Simplification algorithm, vector data compression based on Bin classification and the combination between two previous methods. The results shows that the combination between the two methods have the best performance among the three tested methods. The proposed method can achieve 4-9% compression ratio while the other methods show a lower performance.

• Journal of Power Electronics
• /
• v.17 no.5
• /
• pp.1203-1210
• /
• 2017

This paper presents an improved switching selection method for the direct torque control (DTC) of five-phase induction motors (IMs). The proposed method is conducted using optimal switching selection. A five-phase inverter has 32 voltage vectors which are divided into 30 nonzero voltage vectors and two zero voltage vectors. The magnitudes of the voltage vectors consist of large, medium, and small voltage vectors. In addition, these vectors are related to the torque response and torque ripple. When a large voltage vector is selected in a drive system, the torque response time decreases with an increased torque ripple. On the other hand, when a small voltage vector is selected, the torque response time and torque ripple increase. As a result, this paper proposes an optimal voltage vector selection method for improved DTC of a five-phase induction machine depending on the situation. Simulation and experimental results verify the effectiveness of the proposed control algorithm.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.1
• /
• pp.49-57
• /
• 2008

In this paper, we propose a novel hierarchical distributed intrusion detection system, named HNHDIDS(Home Network Hierarchical Distributed Intrusion Detection System), which is not only based on the structure of distributed intrusion detection system, but also fully consider the environment of secure home networks service. The proposed system is hierarchically composed of the one-class support vector machine(support vector data description) and local agents, in which it is designed for optimizing for the environment of secure home networks service. We support our findings with computer experiments and analysis.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.4
• /
• pp.749-755
• /
• 2009

Support vector classification (SVC) provides more complete description of the lin-ear and nonlinear relationships between input vectors and classifiers. In this paper. we propose the sparse kernel classifier to solve the optimization problem of classification with a modified hinge loss function and absolute loss function, which provides the efficient computation and the sparsity. We also introduce the generalized cross validation function to select the hyper-parameters which affects the classification performance of the proposed method. Experimental results are then presented which illustrate the performance of the proposed procedure for classification.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.3 no.2
• /
• pp.99-106
• /
• 1998

This paper proposed a speed control system for induction motors robust to variations in torque and parameters by feedforward compensating the current portion of load torque, adding a load torque observer to the conventional PI controller in the indirect vector controlled induction motor system. Computer simulations and exeperimental works using the proposed control confirm that the transient response for the variation of the reference speed and load torque becomes improved, compared with the conventional PI controled method.

• Bulletin of the Korean Mathematical Society
• /
• v.50 no.6
• /
• pp.1915-1922
• /
• 2013

We prove the admissibility of the space $L_0(\mathcal{A},X)$ of vector-valued measurable functions determined by real-valued finitely additive set functions defined on algebras of sets.

• 한국데이터정보과학회:학술대회논문집
• /
• 2005.10a
• /
• pp.161-169
• /
• 2005

In this paper we propose a fuzzy c-regression model based on weighted least squares support vector machine(LS-SVM), which can be used to detect outliers in the switching regression model while preserving simultaneous yielding the estimates of outputs together with a fuzzy c-partitions of data. It can be applied to the nonlinear regression which does not have an explicit form of the regression function. We illustrate the new algorithm with examples which indicate how it can be used to detect outliers and fit the mixed data to the nonlinear regression models.