• Journal of Intelligence and Information Systems
• /
• v.1 no.2
• /
• pp.1-16
• /
• 1995

The field of human judgment and decision making provides useful methodologies for examining the human decision making process and substantive results. One of the methodologies is a lens model analysis which can examine valid nonlinearity in the human decision making process. Using the method, valid nonlinearity in human decision behavior can be successfully detected. Two linear(statistical) models of human experts and two nonlinear models of human experts are compared in terms of predictive accuracy (predictive validity). The results indicate that nonlinear models can capture factors(valid nonlinearity) that contribute to the expert's predictive accuracy, but not factors (inconsistency) that detract from their predictive accuracy. Then, it is argued that nonlinear models cab be more accurate than linear models, or as accurate as human experts, especially when human experts employ valid nonlinear strategies in decision making.

• ETRI Journal
• /
• v.41 no.5
• /
• pp.684-695
• /
• 2019

In a cloud environment, performance degradation, or even downtime, of virtual machines (VMs) usually appears gradually along with anomalous states of VMs. To better characterize the state of a VM, all possible performance metrics are collected. For such high-dimensional datasets, this article proposes a feature extraction algorithm based on unsupervised fuzzy linear discriminant analysis with kernel (UFKLDA). By introducing the kernel method, UFKLDA can not only effectively deal with non-Gaussian datasets but also implement nonlinear feature extraction. Two sets of experiments were undertaken. In discriminability experiments, this article introduces quantitative criteria to measure discriminability among all classes of samples. The results show that UFKLDA improves discriminability compared with other popular feature extraction algorithms. In detection accuracy experiments, this article computes accuracy measures of an anomaly detection algorithm (i.e., C-SVM) on the original performance metrics and extracted features. The results show that anomaly detection with features extracted by UFKLDA improves the accuracy of detection in terms of sensitivity and specificity.

• Journal of information and communication convergence engineering
• /
• v.22 no.2
• /
• pp.139-144
• /
• 2024

In this study, four types of fall detection systems - designed with YOLOPose, principal component analysis (PCA), convolutional neural network (CNN), and long short-term memory (LSTM) architectures - were developed and compared in the detection of everyday falls. The experimental dataset encompassed seven types of activities: walking, lying, jumping, jumping in activities of daily living, falling backward, falling forward, and falling sideways. Keypoints extracted from YOLOPose were entered into the following architectures: RAW-LSTM, PCA-LSTM, RAW-PCA-LSTM, and PCA-CNN-LSTM. For the PCA architectures, the reduced input size stemming from a dimensionality reduction enhanced the operational efficiency in terms of computational time and memory at the cost of decreased accuracy. In contrast, the addition of a CNN resulted in higher complexity and lower accuracy. The RAW-LSTM architecture, which did not include either PCA or CNN, had the least number of parameters, which resulted in the best computational time and memory while also achieving the highest accuracy.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.940-943
• /
• 2022

This study aims to analyze the Korean dataset by performing Korean sentence Emotion Analysis in the radio listeners' text messages collected personally. Currently, in Korea, research on the Emotion Analysis of Korean sentences is variously continuing. However, it is difficult to expect high accuracy of Emotion Analysis due to the linguistic characteristics of Korean. In addition, a lot of research has been done on Binary Sentiment Analysis that allows positive/negative classification only, but Multi-class Emotion Analysis that is classified into three or more emotions requires more research. In this regard, it is necessary to consider and analyze the Korean dataset to increase the accuracy of Multi-class Emotion Analysis for Korean. In this paper, we analyzed why Korean Emotion Analysis is difficult in the process of conducting Emotion Analysis through surveys and experiments, proposed a method for creating a dataset that can improve accuracy and can be used as a basis for Emotion Analysis of Korean sentences.

• Aerospace Engineering and Technology
• /
• v.13 no.2
• /
• pp.108-114
• /
• 2014

This paper describes the orbit accuracy of KOMPSAT-5, which has been in normal operations since the launch on Aug. 22, 2013. The analysis on the various GPS related data and the different methodologies for orbit estimation are carried out and compared with each other. The accuracy of precise orbit is confirmed to be 12.8cm($1{\sigma}$) on average using data from the in-flight dual frequency GPS receiver, GPS precise ephemeris, and IGS stations. In addition, the orbit estimation using single frequency GPS receiver provides the orbit solution around 2m level. And, the accuracy of orbit processing is 5m using on-board navigation solution, which has about 10m accuracy.

• Journal of the Ergonomics Society of Korea
• /
• v.31 no.2
• /
• pp.345-352
• /
• 2012

Objective: The aim of this study is to evaluate the accuracy and reliability of Spine-Pelvis Monitor(SPM) that was developed to measure 3-dimensional motion of spine and pelvis using tilt sensor and gyro sensor. Background: The main cause of low back pain is very much associated with the task using the low back and pelvis, but no measurement technique can quantify the both spine and pelvis. Method: For testing the SPM, 125 angles from three anatomical planes were measured three times in order to evaluate the accuracy and reliability. The accuracy of SPM in measuring dynamic motion was evaluated using digital motion analysis system. The motion pattern captured by two measuring methods was compared with each other. In result, the percentage error and Cronbach coefficient alpha were calculated to evaluate the accuracy and reliability. Results: The percentage error was 0.35% in flexion-extension on sagittal plane, 0.43% in lateral bending on coronal plane, and 0.40% in twisting on transverse plane. The Cronbach coefficient alpha was 1.00, 0.99 and 0.99 in sagittal, coronal and transvers plane, respectively. Conclusion: The SPM showed less than 1% error for static measurement, and showed reasonably similar pattern with the digital motion system. Application: The results of this study showed that the SPM can be the measuring method of spine pelvis motion that enhances the kinematic analysis of low back dynamics.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.6
• /
• pp.573-578
• /
• 2010

Precise determination of satellite positions is necessary to improve positioning accuracy in GNSS. In this study, GLONASS orbits were predicted from broadcast ephemeris using the 4th-order Runge-Kutta numerical integration method and their accuracy dependence on the integration step and the integration time was analyzed. The 3D RMS (Root Mean Square) differences between the results from I-second integration step and 300-second integration step was about 3 cm, but the processing time was one hundred times less for the I-second integration time case. For trials of different integration times, the 3D RMS errors were 8.3 m, 187.3 m, and 661.5 m for 30-, 150-, and 300-minutes of integration time, respectively. Though this integration-time analysis, we concluded that the accuracy gets higher with a shorter integration time. Thus we suggest forward and backward integration methods to improve GLONASS positioning accuracy, and with this method we can achieve a 5-meter level of 3-D orbit accuracy.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.5
• /
• pp.117-127
• /
• 2018

The author tried to propose the orthographic and DTM (digital terrain model) creation and evaluate the accuracy for an university campus using UAV (unmanned aerial vehicle) system. Most previous studies used GPS-based data, but in this paper, the observations of triangulation level measurements was used for comparison of accuracy. Accuracy analysis results showed that the operational requirements for aerial photographic standards are satisfied in all scenaries. The author confirmed availability in aviation photo measurements and applications using UAV (Drone). In order to create a sophisticated DTM and contour, we need to eliminate interference from building, trees, and artificial objects. The results of this study are expected to be used as the basis for future studies in the creation of DTM and the accuracy assessments using Drone.

• Korean Journal of Applied Biomechanics
• /
• v.19 no.4
• /
• pp.729-738
• /
• 2009

The purpose of this study was to analyze the impact accuracy and kinematic parameters of skill level and eye-tracking type during putting strokes. For comparison, five elite golfers and five novice golfers participated in this study. Three-dimensional kinematic data were collected for each subject while 10 putting trials were performed for each skill level and eye-tracking type. The APAS system was used to compute the impact accuracy and kinematic parameters of putter heads. The putting stroke was divided into three phases: back swing, downswing, and follow-through. The findings indicated that significant differences were found in skill level as it affected the rate of success. For impact accuracy and the displacement of putter heads, a significant difference was found for the skill level, particularly in backs-wing and follow-through. In addition, the displacement of the putter head had a greater influence on stroke accuracy than on velocity.

• The Journal of Advanced Prosthodontics
• /
• v.13 no.5
• /
• pp.327-332
• /
• 2021

PURPOSE. The purpose of this in vitro study was to evaluate the accuracy of digitally designed removable partial denture (RPD) frameworks, constructed by additive and subtractive methods castable resin patterns, using comparative 3D analysis. MATERIALS AND METHODS. A Kennedy class III mod. 1 educational maxillary model was used in this study. The cast was scanned after modification, and a removable partial denture framework was digitally designed. Twelve frameworks were constructed. Two groups were defined: Group A: six frameworks were milled with castable resin, then casted by the lost wax technique into Co-Cr frameworks; Group B: six frameworks were printed with castable resin, then casted by the lost wax technique into Co-Cr frameworks. Comparative 3D analysis was used to measure the accuracy of the fabricated frameworks using Geomagic Control X software. Student's t-test was used for comparing data. P value ≤ .05 was considered statistically significant. RESULTS. Regarding the accuracy of the occlusal rests, group A (milled) (0.1417 ± 0.0224) showed significantly higher accuracy than group B (printed) (0.02347 ± 0.0221). The same results were found regarding the 3D comparison of the overall accuracy, in which group A (0.1501 ± 0.0205) was significantly more accurate than group B (0.179 ± 0.0137). CONCLUSION. In indirect fabrication techniques, subtractive manufacturing yields more accurate RPDs than additive manufacturing.