• Journal of Information Processing Systems
• /
• 제16권4호
• /
• pp.809-819
• /
• 2020

Gender prediction accuracy increases as convolutional neural network (CNN) architecture evolves. This paper compares voting and ensemble schemes to utilize the already trained five CNN models to further improve gender prediction accuracy. The majority voting usually requires odd-numbered models while the proposed softmax-based voting can utilize any number of models to improve accuracy. The ensemble of CNN models combined with one more fully-connected layer requires further tuning or training of the models combined. With experiments, it is observed that the voting or ensemble of CNN models leads to further improvement of gender prediction accuracy and that especially softmax-based voters always show better gender prediction accuracy than majority voters. Also, compared with softmax-based voters, ensemble models show a slightly better or similar accuracy with added training of the combined CNN models. Softmax-based voting can be a fast and efficient way to get better accuracy without further training since the selection of the top accuracy models among available CNN pre-trained models usually leads to similar accuracy to that of the corresponding ensemble models.

• International Journal of Computer Science & Network Security
• /
• 제21권12spc호
• /
• pp.549-555
• /
• 2021

Machine Learning(ML) splits data into 3 parts, which are usually 60% for training, 20% for validation, and 20% for testing. It just splits quantitatively instead of selecting each set of data by a criterion, which is very important concept for the adequacy of test data. ML measures a model's accuracy by applying a set of validation data, and revises the model until the validation accuracy reaches on a certain level. After the validation process, the complete model is tested with the set of test data, which are not seen by the model yet. If the set of test data covers the model's attributes well, the test accuracy will be close to the validation accuracy of the model. To make sure that ML's set of test data works adequately, we design an experiment and see if the test accuracy of model is always close to its validation adequacy as expected. The experiment builds 100 different SVM models for each of six data sets published in UCI ML repository. From the test accuracy and its validation accuracy of 600 cases, we find some unexpected cases, where the test accuracy is very different from its validation accuracy. Consequently, it is not always true that ML's set of test data is adequate to assure a model's quality.

• Journal of Periodontal and Implant Science
• /
• 제52권3호
• /
• pp.220-229
• /
• 2022

Purpose: The aim of this study was to evaluate and compare the accuracy performance of dental professionals in the classification of different types of dental implant systems (DISs) using panoramic radiographic images with and without the assistance of a deep learning (DL) algorithm. Methods: Using a self-reported questionnaire, the classification accuracy of dental professionals (including 5 board-certified periodontists, 8 periodontology residents, and 31 dentists not specialized in implantology working at 3 dental hospitals) with and without the assistance of an automated DL algorithm were determined and compared. The accuracy, sensitivity, specificity, confusion matrix, receiver operating characteristic (ROC) curves, and area under the ROC curves were calculated to evaluate the classification performance of the DL algorithm and dental professionals. Results: Using the DL algorithm led to a statistically significant improvement in the average classification accuracy of DISs (mean accuracy: 78.88%) compared to that without the assistance of the DL algorithm (mean accuracy: 63.13%, P<0.05). In particular, when assisted by the DL algorithm, board-certified periodontists (mean accuracy: 88.56%) showed higher average accuracy than did the DL algorithm, and dentists not specialized in implantology (mean accuracy: 77.83%) showed the largest improvement, reaching an average accuracy similar to that of the algorithm (mean accuracy: 80.56%). Conclusions: The automated DL algorithm classified DISs with accuracy and performance comparable to those of board-certified periodontists, and it may be useful for dental professionals for the classification of various types of DISs encountered in clinical practice.

• 한국정밀공학회지
• /
• 제31권5호
• /
• pp.431-439
• /
• 2014

Recently, a demand for precision 5-axis machine tools is significantly increasing, and the maintenance of machine tool accuracy becomes more important. it is very difficult to evaluate to accuracy of 5-axis M/C in the production site since it needs expensive measuring equipment and skilled engineer. On the other hand, evaluation items of 5-axis M/C are not systematically organized in the existing KS and ISO standards. In this study, the evaluation items for 5-axis M/C were derived systematically and a test workpiece was developed to evaluate the machine tool accuracy more easily. The error sources of machine tool can be estimated by machining and measuring of the test workpiece. The correlation between the machine tool accuracy and the accuracy of machined test workpiece was analyzed. As a result, the accuracy of machined test workpiece represented the accuracy of machine tool and the error sources very effectively.

• International journal of advanced smart convergence
• /
• 제6권3호
• /
• pp.17-21
• /
• 2017

In this paper, the service execution accuracy was compared by ontology based rule inference method and machine learning method, and the amount of data at the point when the service execution accuracy of the machine learning method becomes equal to the service execution accuracy of the rule inference was found. The rule inference, which measures service execution accuracy and service execution accuracy using accumulated data and pattern matching on service results. And then machine learning method measures service execution accuracy using cross validation data. After creating a confusion matrix and measuring the accuracy of each service execution, the inference algorithm can be selected from the results.

• 한국정밀공학회지
• /
• 제14권7호
• /
• pp.15-21
• /
• 1997

This paper presented a linear motion accuracy by using two-dimensional probe with the master block and the square for NC machine tools. This measuring system could be measured motion error due to numerical control system. The results of measurement and simulation for motion error were similar, and so, this system had enough accuracy to measure a linear motion accuracy for NC machine tools. The experimental results are as follows. 1. This measuring system could be measured motion error due to mumerical control system. 2. The results of measurement and simulation for motion error were similar. 3. This measuring system had enough accuracy to measure a linear motion accuracy for NC machine tools.

• Journal of the Korean Data and Information Science Society
• /
• 제25권6호
• /
• pp.1345-1352
• /
• 2014

데이터마이닝 기법 중에서 연관성 규칙은 연관성 평가 기준을 기반으로 하여 데이터베이스에 포함되어 있는 항목들 간의 관련성을 탐색하는 기법이다. 일반적인 연관성 규칙 기법과는 달리 역의 연관성 규칙은 하나의 항목집합이 발생하지 않으면 다른 항목집합도 발생하지 않는다는 규칙을 찾아내는 것이다. 이러한 역의 연관성 규칙을 일반적인 연관성 규칙과 함께 생성하면 기업체에서 특정 제품을 판매하기 위해서는 그 제품만의 마케팅뿐만 아니라 더 나아가 어떤 제품의 마케팅이 필요한 지에 대한 정보를 파악할 수 있다. 이를 위해 본 논문에서는 이러한 두 종류의 연관성 규칙에 적용 가능한 균형화된 기여 상대적 규칙 정확도를 연관성 평가 기준으로 제안하고자 한다. 또한 Piatetsky-Shapiro (1991)가 제안한 흥미도 측도가 가져야 할 조건들을 점검한 후, 예제를 통하여 제안된 측도와 연관성 규칙에 적용 가능한 의학진단분야의 평가 측도들의 유용성을 비교하였다. 그 결과, 기여 상대적 정확도와 역의 기여 상대적 정확도의 크기가 다르게 나타나면 연관성의 정도를 명확하게 설명하기가 어려우므로 이들 두 측도를 동시에 고려한 균형화된 기여 상대적 규칙 정확도를 이용하는 것이 가장 바람직하다는 사실을 확인하였다.

• 대한원격탐사학회지
• /
• 제35권6_1호
• /
• pp.895-905
• /
• 2019

현재 현업에서 사용되고 있는 상용 무인기 영상처리 소프트웨어는 카메라 캘리브레이션 정보나 영상 전체에 대한 블록 번들조정 정확도만 제공할 뿐 스테레오 페어의 실제 도화 가능여부에 대한 정확도는 거의 제공하지 않는다. 본 논문에서는 무인기 영상처리 소프트웨어에서 산출된 표정요소를 사용하여 도화품질을 산출하고 실제 도화기에 적용하여 도화품질의 신뢰성에 대해서 분석하였다. 도화품질은 Y시차 정확도, 상대모델 정확도, 절대모델 정확도의 3가지 정확도로 정의하였다. Y시차 정확도는 스테레오 페어간 입체시 여부를 판단할 수 있는 정확도이다. 상대모델 정확도는 모델 좌표계 상에서 스테레오 페어간 상대적인 번들조정 정확도이다. 절대모델 정확도는 절대 좌표계에서 번들조정 정확도이다. 실험데이터는 도심지를 대상으로 회전익에서 취득된 GSD 5 cm급의 영상 723장을 사용하여 도화품질을 분석하였다. 연구진이 개발한 기술을 사용해 예측한 상대모델 정확도와 실제 도화기에서 관측한 정확도의 최대오차는 0.11 m로 정밀한 결과를 보여 주었다. 절대모델 정확도도 마찬가지로, 도화기에서 관측한 정확도의 최대오차는 0.16 m로 정밀한 결과를 보여주었다.

• 한국측량학회:학술대회논문집
• /
• 한국측량학회 2007년도 춘계학술발표회 논문집
• /
• pp.447-450
• /
• 2007

Positioning technology for moving object is an important and essential component of ubiquitous. Also RFID(Radio Frequency IDentification) is a core technology of ubiquitous wireless communication. In this study we adapted kalman-filter theory to RFID-based Positioning System in order to trace a time-variant moving object and verify the positioning accuracy using RMSE (Roong technology for moving object is an important and essential component of ubiquitous Mean Square Error). The purpose of this study is to verify an effect of kalman-filter on the positioning accuracy and to analyze what does each design factor have an effect on the positioning accuracy by means of simulations and to suggest a standard of optimal design factor of a RFID-based Positioning System. From the results of simulations, Kalman-filer improved the positioning accuracy remarkably; the detection range of RFID tag is not a determining factor. The smaller standard deviation of detection range improves the positioning accuracy. However it accompanies a smaller fluctuation of the positioning accuracy. The larger detection rate of RFID tag yields the smaller fluctuation in the positioning accuracy and has more stable system and improves the positioning accuracy;

• 지식경영연구
• /
• 제13권5호
• /
• pp.1-13
• /
• 2012

This study investigates whether firm-specifics affect forecast accuracy using a sample of IPO firms in Korea. The forecasts accuracy can be differentiated depending on firm specifics. This study uses the foreign investor, intangible asset and patents as firm specifics. The analysts are divided into two groups by firm-specifies(foreign investors ratio of low and high, intangible asset ratio of low and high, patents of acquisition) and also examine the degree of analysts's forecast accuracy over the two groups. and examined the degree of the analysts' forecast accuracy over the two groups. The sample is composed of 460 IPO (Initial Public Offering) firms listed on the KOSDAQ (Korean Securities Dealers Automated Quotations) for the period from 2001 to 2009. The analysts' forecast accuracy is much higher in the group of high foreign investor but is lower in the group of high intangible assets and patents. Also, the group of high foreign investors respectively interacts with group of high intangible assets ratio and group of patents of acquisition. In result, The analysts' forecast accuracy is higher because foreign investor is decreased information asymmetry. This study compares suggests that patents may be helpful for predicting forecast accuracy.