• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.5
• /
• pp.277-286
• /
• 2019

In this paper, we propose an efficient hyperplane generation technique to classify human activity from combination of events and sequence information obtained from multiple-event sensors. By generating hyperplane efficiently, our machine learning algorithm classify with less memory and run time than the LSVM (Linear Support Vector Machine) for embedded system. Because the fact that light weight and high speed algorithm is one of the most critical issue in the IoT, the study can be applied to smart home to predict human activity and provide related services. Our approach is based on reducing numbers of hyperplanes and utilizing robust string comparing algorithm. The proposed method results in reduction of memory consumption compared to the conventional ML (Machine Learning) algorithms; 252 times to LSVM and 34,033 times to LSTM (Long Short-Term Memory), although accuracy is decreased slightly. Thus our method showed outstanding performance on accuracy per hyperplane; 240 times to LSVM and 30,520 times to LSTM. The binarized image is then divided into groups, where each groups are converted to binary number, in order to reduce the number of comparison done in runtime process. The binary numbers are then converted to string. The test data is evaluated by converting to string and measuring similarity between hyperplanes using Levenshtein algorithm, which is a robust dynamic string comparing algorithm. This technique reduces runtime and enables the proposed algorithm to become 27% faster than LSVM, and 90% faster than LSTM.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.12
• /
• pp.896-902
• /
• 2018

This study examined the flow and noise inside a sirocco fan for ventilation as a commercial program. To confirm only the location and power of the noise source, flow analysis was performed with steady state flow analysis. Through flow analysis, the flow was observed in the sirocco fan and the velocity vector. The pressure distribution inside was observed with contours. From the results of steady analysis, the position and size of the noise source could be seen using the 'Curle surface acoustic power' and 'Proudman acoustic power'. The Curle surface acoustic power can be used to observe the noise from the surface. The Proudman acoustic power can be used to detect noise generated in the flow region because the position and size of the noise source generated inside the sirocco fan can be seen only in the steady state. Therefore it is necessary to further analyze the unsteady state to check the frequency of the noise generated. This study provides basic data for improving the performance of the Sirocco fan and reducing the noise.

• Journal of Broadcast Engineering
• /
• v.24 no.1
• /
• pp.77-86
• /
• 2019

This paper proposes using machine-learning technology to analyze and classify historical collected documents based on them. Data is collected based on keywords associated with a specific domain and the non-conceptuals such as special characters are removed. Then, tag each word of the document collected using a Korean-language morpheme analyzer with its nouns, verbs, and sentences. Embedded documents using Doc2Vec model that converts documents into vectors. Measure the similarity between documents through the embedded model and learn the document classifier using the machine running algorithm. The highest performance support vector machine measured 0.83 of F1-score as a result of comparing the classification model learned.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1296-1301
• /
• 2021

With the recent increase in diabetes incidence worldwide, research has been conducted to predict diabetes through various machine learning and deep learning technologies. In this work, we present a model for predicting diabetes using machine learning techniques with German Frankfurt Hospital data. We apply outlier handling using Interquartile Range (IQR) techniques and Pearson correlation and compare model-specific diabetes prediction performance with Decision Tree, Random Forest, Knn (k-nearest neighbor), SVM (support vector machine), Bayesian Network, ensemble techniques XGBoost, Voting, and Stacking. As a result of the study, the XGBoost technique showed the best performance with 97% accuracy on top of the various scenarios. Therefore, this study is meaningful in that the model can be used to accurately predict and prevent diabetes prevalent in modern society.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.603-610
• /
• 2021

In this paper, we extract the user's power consumption patterns, and model the optimal consumption patterns by applying the user's environment and emotion. Based on the comparative analysis of these two patterns, we present an efficient power consumption method through changes in the user's power consumption behavior. To extract significant consumption patterns, vector standardization and binary data transformation methods are used, and learning about the ensemble's ensemble with k-means clustering is applied, and applying the support factor according to the value of k. The optimal power consumption pattern model is generated by applying forced and emotion-based control based on the learning results for ensemble aggregates with relatively low average consumption. Through experiments, we validate that it can be applied to a variety of windows through the number or size adjustment of clusters to enable forced and emotion-based control according to the user's intentions by identifying the correlation between the number of clusters and the consistency ratios.

• International Area Studies Review
• /
• v.16 no.1
• /
• pp.3-27
• /
• 2012

This paper adopts recently developed panel unit root test that is cross-sectionally robust. Cointegration test is also used to find whether regional house prices are in line with gross regional domestic production (GRDP) in the long run in Korea during 1989-2009. Based on the panel VECM and the panel ARDL models, we examine causal relationships among the variables and estimate the long-run elasticity. We find evidence of cointegration and bidirectional causal relationships between regional house prices and GRDP. The results of long-run estimates, using both fixed effect and ARDL models, show that house prices positively and significantly influence on the GRDP and vice versa. Together with these results, the findings of ARDL-ECM imply that there exists a long-run equilibrium relationship between house prices and regional economic variables even if there is a possibility of short-run deviation from its long-run path.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.17 no.5
• /
• pp.39-53
• /
• 2007

The purpose of this paper is the development of a compensation algorithm by which the interoperability of fingerprint recognition can be improved among various different fingerprint sensor. In order to compensate for the different characteristics of fingerprint sensor, an initial evaluation of the sensors using both the ink-stamped method and the flat artificial finger pattern method was undertaken. This paper proposes Common resolution method and Relative resolution method for compensating different resolution of fingerprint images captured by disparate sensors. Both methods can be applied to image-level and minutia-level. In order to compensate the direction of minutiae in minutia-level, Unit vector method is proposed. The EER of the proposed method was improved by average 64.8% better than before compensation. This paper will make a significant contribution to interoperability in the system integration using different sensors.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.786-791
• /
• 2021

In this paper, in order to minimize limitations such as signal interference and positioning errors in existing indoor positioning systems, a smart LED-based positioning system for excellent line-of-sight radio environments and precise location tracking is proposed to improve accuracy. An IEEE 802.4 Zigbee module is mounted on the SMPS board of a smart LED; RSSI and LQI signals are received from a moving tag, and the system is configured to transmit the measured data to the positioning server through a gateway. For the experiment, the necessary hardware, such as the gateway and the smart LED module, were separately designed, and the experiment was conducted after configuring the system in an external field office. The positioning error was within 70cm as a result of performing complex calculations in the positioning server after transmitting a vector value of the moving object obtained from the direction sensor, together with a signal from the moving object received by the smart LED. The result is a significantly improved positioning error, compared to an existing short-range wireless communications-based system, and shows the level at which commercial products can be implemented.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.1
• /
• pp.21-30
• /
• 2022

The restart of scramjet engine is almost impossible in case its unstart happens during engine operation. Therefore, it is essential to prognosticate the scramjet engine unstart phenomena. A numerical simulation of a scramjet engine is conducted to investigate the unstart process of the scramjet engine by adjusting the backpressure at the isolator outlet to the engine analysis environment. The start and unstart of the engine are identified by applying a support vector machine (SVM) through the data measured by wall pressure so that the locations of the pressure sensors most suitable for the unstart detection are selected. And the operation conditions in which the engine is avoid to be unstarted are quantified to perceive the safety margin.