• Journal of the Korean Data and Information Science Society
• /
• 제25권6호
• /
• pp.1521-1530
• /
• 2014

The ROC analysis is considered for multiple class diagnosis. There exist many criteria to find optimal thresholds and measure the accuracy of diagnostic tests for k dimensional ROC analysis. In this paper, we proposed a diagnostic accuracy measure called the correct classification simple rate, which is defined as the summation of true rates for each classification distribution and expressed as a function of summation of sequential true rates for two consecutive distributions. This measure does not weight accuracy across categories by the category prevalence and is comparable across populations for multiple class diagnosis. It is found that this accuracy measure does not only have a relationship with Kolmogorov - Smirnov statistics, but also can be represented as a linear function of some optimal threshold criteria. With these facts, the suggested measure could be applied to test for comparing multiple distributions.

• Journal of Multimedia Information System
• /
• 제5권2호
• /
• pp.99-104
• /
• 2018

Phishing website has become a crucial concern in cyber security applications. It is performed by fraudulently deceiving users with the aim of obtaining their sensitive information such as bank account information, credit card, username, and password. The threat has led to huge losses to online retailers, e-business platform, financial institutions, and to name but a few. One way to build anti-phishing detection mechanism is to construct classification algorithm based on machine learning techniques. The objective of this paper is to compare different classifier ensemble approaches, i.e. random forest, rotation forest, gradient boosted machine, and extreme gradient boosting against single classifiers, i.e. decision tree, classification and regression tree, and credal decision tree in the case of website phishing. Area under ROC curve (AUC) is employed as a performance metric, whilst statistical tests are used as baseline indicator of significance evaluation among classifiers. The paper contributes the existing literature on making a benchmark of classifier ensembles for web phishing detection.

• 한국지진공학회논문집
• /
• 제22권6호
• /
• pp.311-322
• /
• 2018

As a method of seismic-design for pile-supported wharves, equivalent static analysis, response spectrum analysis, and time history analysis method are applied. Among them, the response spectrum analysis is widely used to obtain the maximum response of a structure. Because the ground is not modeled in the response spectrum analysis of pile-supported wharves, the amplified input ground acceleration should be calculated by ground classification or seismic response analysis. However, it is difficult to calculate the input ground acceleration through ground classification because the pile-supported wharf is build on inclined ground, the methods to calculate the input ground acceleration proposed in the standards are different. Therefore, in this study, the dynamic centrifuge model tests and the response spectrum analysis were carried out to calculate the appropriate input ground acceleration. The pile moment in response spectrum analysis and the dynamic centrifuge model tests were compared. As a result of comparison, it was shown that the response spectrum analysis results using the amplified acceleration in the ground surface were appropriate.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권11호
• /
• pp.928-935
• /
• 2009

온톨로지 스키마 추론은 개념의 정합성을 테스트하고 계층구조를 자동으로 생성해 준다. 이를 위해 개념의 검색과정은 반드시 수행하여야 한다. 온톨로지 스키마 추론은 온톨로지에 정의된 모든 개념들간의 포함관계를 테스트함으로써 수행된다. 이러한 포함관계 테스트 결과는 개념의 완전 그래프 생성여부에 따라 결정되며 이는 추론의 전체적인 성능에 결정적인 영향을 준다. 일반적으로 완전 그래프 생성 과정은 복잡도가 높은 과정으로 알려져 있으며 이는 낮은 추론성능을 유발시키는 원인이 되고 있다. 본 논문에서는 개념들간의 포함관계 테스트시 개념의 향상된 검색방법을 통해 불필요한 포함관계 테스트를 생략함으로써 온톨로지 스키마 추론의 성능을 향상시키는 방법을 제시하고자 한다. 이 향상된 검색방법은 개념의 포함관계 테스트 결과를 정의상 연관된 다른 개념에 전파함으로써 이루어진다.

• 한국군사과학기술학회지
• /
• 제21권2호
• /
• pp.147-157
• /
• 2018

There are increasing demands to provide early warning against intruding drones and cope with potential threats. Commercial anti-drone systems are mostly based on simple target detection by radar reflections. In real scenario, however, it becomes essential to obtain drone radar signatures so that hostile targets are recognized in advance. We present experimental test results that micro-Doppler radar signature delivers partial information on multi-rotor platforms and exhibits limited performance in drone recognition and classification. Afterward, we attempt to generate high resolution profile of flying drone targets. To this purpose, wide bands radar signals are employed to carry out inverse synthetic aperture radar(ISAR) imaging against moving drones. Following theoretical analysis, experimental field tests are carried out to acquire real target signals. Our preliminary tests demonstrate that high resolution ISAR imaging provides effective measures to detect and classify multiple drone targets in air.

• 전기학회논문지
• /
• 제65권7호
• /
• pp.1263-1269
• /
• 2016

Hearing impaired persons are exposed to the danger since they can't be aware of many dangerous situations like fire alarms, car hones and so on. Therefore they need haptic or visual informations when they meet dangerous situations. In this paper, we design a dangerous sound detection engine for hearing impaired. We consider four dangerous indoor situations such as a boiled sound of kettle, a fire alarm, a door bell and a phone ringing. For outdoor, two dangerous situations such as a car horn and a siren of emergency vehicle are considered. For a test, 6 data sets are collected from those six situations. we extract LPC, LPCC and MFCC as feature vectors from the collected data and compare the vectors for feasibility. Finally we design a matching engine using an artificial neural network and perform classification tests. We perform classification tests for 3 times considering the use outdoors and indoors. The test result shows the feasibility for the dangerous sound detection.

• 한국IT서비스학회지
• /
• 제20권3호
• /
• pp.75-86
• /
• 2021

As artificial intelligence is actively studied, chatbot systems are being applied to various fields. In particular, many chatbot systems for psychological counseling have been studied that can comfort modern people. However, while most psychological counseling chatbots are studied as rule-base and deep learning-based chatbots, there are large limitations for each chatbot. To overcome the limitations of psychological counseling using such chatbots, we proposes a novel psychological counseling AI chatbot system. The proposed system consists of a GPT-2 model that generates output sentence for Korean input sentences and an Electra model that serves as sentiment analysis and anxiety cause classification, which can be provided with psychological tests and collective intelligence functions. At the same time as deep learning-based chatbots and conversations take place, sentiment analysis of input sentences simultaneously recognizes user's emotions and presents psychological tests and collective intelligence solutions to solve the limitations of psychological counseling that can only be done with chatbots. Since the role of sentiment analysis and anxiety cause classification, which are the links of each function, is important for the progression of the proposed system, we experiment the performance of those parts. We verify the novelty and accuracy of the proposed system. It also shows that the AI chatbot system can perform counseling excellently.

• 중환자간호학회지
• /
• 제5권2호
• /
• pp.49-60
• /
• 2012

Purpose: The purpose of this study was to explore health care workers awareness and performance of standard precautions. Methods: Participants were 296 health care workers including nurses, physicians, and medical technicians. Awareness and performance of standard precautions were measured with 4-point Likert scales. The data were analyzed with t-tests and one-way ANOVA by using SPSS 18.0. Results: The mean scores of awareness were 3.72 in nurses, 3.62 in physicians, and 3.47 in medical technicians. There was a significant difference of awareness by occupational classification (F=12.39, p<.001). The mean scores of performance of standard precautions were 3.45 in nurses, 3.19 in physicians, and 3.23 in medical technicians. There was a significant difference of performance by occupational classification (F=10.98, p<.001). In addition, the score of performance of standard precautions was significantly lower than that of awareness (t=11.89, p<.001). Conclusion: The results of this study indicated that awareness and performance of standard precautions were different by occupational classification. To improve performance of standard precautions in hospitals, it is necessary to provide a distinct infection control program by occupational classification.

• Molecules and Cells
• /
• 제22권1호
• /
• pp.78-88
• /
• 2006

The phylogeny of the subfamily Tephritinae (Diptera: Tephritidae) was reconstructed from mitochondrial 16S ribosomal RNA gene sequences using 53 species representing 11 currently recognized tribes of the Tephritinae and 10 outgroup species. The minimum evolution and Bayesian trees suggested the following phylogenetic relationships: (1) monophyly of the Tephritinae was strongly supported; (2) a sister group relationship between the Tephritinae and Plioreocepta was supported by the Bayesian tree; (3) the tribes Tephrellini, Myopitini, and Terelliini (excluding Neaspilota) were supported as monophyletic groups; (4) the non-monophyletic nature of the tribes Dithrycini, Eutretini, Noeetini, Tephritini, Cecidocharini, and Xyphosiini; and (5) recognition of 10 putative tribal groups, most of which were supported strongly by the statistical tests of the interior branches. Our results, therefore, convincingly suggest that an extensive rearrangement of the tribal classification of the Tephritinae is necessary. Since our sampling of taxa heavily relied on the current accepted classification, some lineages identified by the present study were severely under-sampled and other possible major lineages of the Tephritinae were probably not even represented in our dataset. We believe that our results provide baseline information for a more rigorous sampling of additional taxa representing all possible major lineages of the subfamily, which is essential for a comprehensive revision of the tephritine tribal classification.

• 한국통신학회논문지
• /
• 제24권12B호
• /
• pp.2447-2454
• /
• 1999

본 논문은 기존 분류형 신경망의 인식성능을 향상시키기 위하여 프레임 정규화와 비선형 사후확률 추정법(N-APPEM)을 제안하고 한국어 숫자음에 대하여 예측형과 분류형 신경망으로 인식성능을 평가하였다. 실험결과 예측형 신경망에서 최고 98.0%의 인식률을 얻었다. 예측형 신경망은 네트워크가 입력패턴의 카테고리 수만큼 마련되는 복잡한 네트워크를 가지는 반면에 분류형 신경망은 단일 네트워크로 구성되며 프레임 정규화와 비선형 사후확률 추정법으로 85.5%까지 인식률을 향상시킬 수 있었으며 이는 기존의 방법보다 인식률이 12.0% 향상된 것이다.