• ETRI Journal
• /
• 제41권3호
• /
• pp.358-370
• /
• 2019

The main aim of this study is to select the optimal set of genes from microarray cancer datasets that contribute to the prediction of specific cancer types. This study proposes the enhancement of the feature selection filter algorithm based on Joe's normalized mutual information and its use for gene selection. The proposed algorithm is implemented and evaluated on seven benchmark microarray cancer datasets, namely, central nervous system, leukemia (binary), leukemia (3 class), leukemia (4 class), lymphoma, mixed lineage leukemia, and small round blue cell tumor, using five well-known classifiers, including the naive Bayes, radial basis function network, instance-based classifier, decision-based table, and decision tree. An average increase in the prediction accuracy of 5.1% is observed on all seven datasets averaged over all five classifiers. The average reduction in training time is 2.86 seconds. The performance of the proposed method is also compared with those of three other popular mutual information-based feature selection filters, namely, information gain, gain ratio, and symmetric uncertainty. The results are impressive when all five classifiers are used on all the datasets.

• Computers and Concrete
• /
• 제24권4호
• /
• pp.329-345
• /
• 2019

Despite the extensive use of mortar materials in constructions over the last decades, there is not yet a robust quantitative method, available in the literature, which can reliably predict mortar strength based on its mix components. This limitation is due to the highly nonlinear relation between the mortar's compressive strength and the mixed components. In this paper, the application of artificial neural networks for predicting the compressive strength of mortars has been investigated. Specifically, surrogate models (such as artificial neural network models) have been used for the prediction of the compressive strength of mortars (based on experimental data available in the literature). Furthermore, compressive strength maps are presented for the first time, aiming to facilitate mortar mix design. The comparison of the derived results with the experimental findings demonstrates the ability of artificial neural networks to approximate the compressive strength of mortars in a reliable and robust manner.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제12권1호
• /
• pp.25-40
• /
• 2023

추천 시스템은 지능적인 자동 결정을 생성하기 위해 사용자가 자주 사용한다. 영화 추천 시스템의 연구에서, 기존 접근 방식은 협업 및 콘텐츠 기반 필터링 기술을 사용한다. 협업 필터링은 사용자 유사성을 고려하는 반면, 콘텐츠 기반 필터링은 단일 사용자의 활동에 중점을 두고 있다. 또한 협업 필터링과 콘텐츠 기반 필터링을 결합한 혼합 필터링 접근법은 서로의 한계를 보완하기 위해 사용되고 있다. 최근엔 더 나은 추천 서비스를 제공하기 위해 사용자 간의 유사성을 찾는데 몇 가지 AI 기반 유사성 기법을 사용하고 있다. 본 논문은 기존의 다양한 영화 추천 시스템과 문제점 분석을 통해 가능한 해결책을 도출하여 유용한 확장 방안을 제공하는 것을 목표로 한다.

• 한국전기전자재료학회논문지
• /
• 제36권2호
• /
• pp.175-185
• /
• 2023

A hybrid supercapacitor is a promising energy storage device in view of its excellent capacitive performance. Commercial three-dimensional foam nickel (Ni) can be used as an ideal framework due to an interconnected network structure. However, its application as an electrode material for supercapacitors is limited due to its low specific capacity. Herein, we report a successful growth of MnO₂ on the surface of graphene by a one-step hydrothermal method; thus, forming a three-dimensional MnO₂-graphene-Ni hybrid foam. Our results show that the mixed structure of MnO₂ with nanoflowers and nanorods grown on the graphene/Ni foam as a hybrid electrode delivers the maximum specific capacitance of 193 F·g^-1 at a current density 0.1 A·g^-1. More importantly, the hybrid electrode retains 104% of its initial capacitance after 1,000 charge-discharge cycles at 1 A·g^-1; thus, showing the potential application as a stable supercapacitor electrode.

• IMMUNE NETWORK
• /
• 제16권2호
• /
• pp.116-125
• /
• 2016

Medical records of 139 tumors from 114 dogs that underwent surgery from May 2010 through March 2015 were retrospectively reviewed. Among 114 dogs, females (64.9%) were significantly more common than males (35.1%) (p<0.05). Dogs aged 6 to 10 years were more presented than non-tumor patients, however, there was no significant difference. The mean age (±SD) was 10.3±3.0 years. Although we found no significant difference of breed predisposition, the most common breed was Maltese (19.3%), followed by Shih-Tzu (14.0%), and Yorkshire terrier (13.2%). Proportional morbidity ratios (PMRs) of mammary gland, oral cavity, and skin tumors were high in Poodles, Yorkshire terriers, and Golden retrievers, respectively. Mammary gland (36.0%) was the most common site, followed by skin and soft tissues (12.2%), oral cavity (10.8%), and digestive organs (8.6%), but there was no significant difference. The objectives of surgery were curative surgery (86.2%), biopsy (4.9%), and palliative surgery (6.5%). In this study, 123 of 139 tumors had histopathological diagnoses. Adenocarcinoma was the most common type (n=24), followed by adenoma (n=17), soft tissue sarcoma (n=13), benign mixed tumor (n=5), and others (n=64). Recurrence or suspected metastasis was identified in 26 dogs. Median survival times of malignant mammary gland tumors, skin and subcutaneous tumors, and splenic tumors were 1,563.0±1,201.7, 469, and 128 days, respectively.

• 대한의용생체공학회:의공학회지
• /
• 제23권3호
• /
• pp.197-205
• /
• 2002

본 연구에서는 잡음이 존재하는 공간에서 획득한 음성신호로부터 후두질환을 감별진단 할 수 있는 분류기를 구현하였다. 이를 위해 후두질환 환자로부터 수집한 /아/ 모음에 잡음을 혼입하여 음성 신호를 획득하였고, 여러 가지 후두질환을 감별진단 할 수 있는 파라미터를 추출하였으며. 이를 입력으로 하는 계층적 신경회로망을 구성하여 후두질환을 감별진단 하도록 하였다. 감별진단용 분류기는 다섯 단계의 계층적 신경회로망으로 구성하였다 첫 번째 신경회로망은 정상 양성 후두질환과 악성 후두질환을. 두 번째 신경회로망은 정상과 양성 후두질환을 감별진단 하도록 하였다 그리고 세 번째 신경회로망은 양성 후두질환 중 후두용. 성대결절 후두마비를 감별진단 하도록 하였으며. 네 번째와 다섯 번째 신경회로망은 성문암 1-4기를 감별진단 하도록 구성하였다. 분류기에 적용된 신경회로망은 다층퍼셉트론 구조로써 역전파 알고리듬으로 학습시켰으며, 선형변환 표준점수변환 등 전처리과정을 적용하여 분류기의 성능을 개선하였다. 후두질환의 감별진단 결과 후두용 88.23%. 정상. 성대결절. 후두마비 100%. 성문암 1기 90%, 성문암 2-4기 100%의 감별진단율을 관찰할 수 있었다.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• 제23권2호
• /
• pp.57-62
• /
• 2012

Objectives : Characteristic symptoms, including hyperactivity and easy distractibility, in children with attention-deficit hyperactivity disorder (ADHD) suggest that their brain status, even at rest, might differ from that of healthy children. This study was conducted in order to determine whether resting state brain activity is compromised in medication-naive children with ADHD. Methods : Twenty medication-naive children with ADHD (mean age $10.3{\pm}2.5$) and 28 age- and gender-matched healthy volunteers (mean age $10.3{\pm}2.0$) underwent measurements for resting state brain activity using functional magnetic resonance imaging (fMRI). Among resting state related-independent components (RSICs) extracted from fMRI data using independent component analysis, a significant difference in RSICs was observed between groups, using a mixed Gaussian/gamma model. Results : Except for IQ, which was higher in the healthy control group, no demographic difference was observed between the two groups (p<.001). Significantly less activation of one RSIC, which includes the bilateral precuneus/posterior cingulate cortex, occipito-temporal junction, and anterior cingulate cortex, was observed in the ADHD group, compared with the control group (p<.05). Conclusion : An abnormal RSIC, posterior default mode network (DMN), was observed in the medication-naive ADHD group. Results of our study suggest that abnormality of posterior DMN is one of the main pathophysiologies of ADHD.

• 한국GIS학회:학술대회논문집
• /
• 한국GIS학회 1995년도 추계학술대회 발표요약문
• /
• pp.9-21
• /
• 1995

계획지원체계는 도시성장관리수단으로 간선시설 시설확장을 이용하는 우리의 능력을 증진시킨다. 이연구는 그러한 계획지원체계 (PEGASUS: 공간도시체계의 생성 및 분석을 위한 계획환경)의 개발계획의 일환으로서 상수도 시설확장모델의 개발에 관한 연구이다. 이 연구는 토지이용과 개발은 간선시설의 수요를 유발하고 간선시설은 토지가 이용되고 개발되는 방향에 영향을 미친다는 전제하에서 시작된다. 상수도 시설 확장은 2 방법으로 해결될 수 있다: 1) 최적통제이론, 2) MINLP 방법. 이 방법들은 각각장단점을 가지고 있다. 이 연구에서는 동적 시설 확장 크기 및 시기를 동시에 결정할 수 있는 MINLP방법이 채택이 되었다. 상수도 관망해석모델과 동적 상수도 시설 최적화 모델이 상수도계획과 토지이용계획의 동적 연관관계를 해결할 수 있다. 상수도 관망해석모델은 생성된 관경의 적정성을 분석하며 동적 상수도 시설 최적화모델은 변화하는 상수도 수요량을 충족할 수 있는 대안을 작성한다. 표준화기법에 의한 비용편익분석은 가장 경제적인 대안을 선정한다. GIS는 필요 상수도 수요량을 산정하고 시설확장모델의 결과를 이용자에게 보여주는데 훌륭한 역활을 할 수 있다.

• 한국측량학회지
• /
• 제38권6호
• /
• pp.663-670
• /
• 2020

본 연구에서는 건물이 혼재한 준 도심 지역에서 발생할 수 있는 재난/재해를 가정하여 네트워크 RTK (Real Time Kinematic) 측위가 가능한 무인기를 이용한 항공삼각측량의 정확도를 평가하였다. 검사점 측위의 신뢰성을 확보하기 위해 검사점을 건물의 옥상에 설치하여 네 시간 이상의 GNSS (Global Navigation Satellite System) 정적 측위를 수행하였다. 객관적인 정확도 평가를 위해 소프트웨어에서 자동으로 인식 가능한 코드화된 대공 타겟을 사용하였다. 무인기에서는 네트워크 RTK 측위의 일종인 VRS (Virtual Reference Station) 방식을 이용하여 영상 취득 당시 카메라의 3차원 좌표를 측정하였고, IMU (Inertial Measurement Unit)와 짐벌 회전각 측정을 통해 카메라의 3축 회전각을 측정하였다. Agisoft Metashape를 이용하여 내·외부 표정요소를 추정·갱신한 결과, 항공삼각측량의 3차원 RMSE (Root Mean Square Error)는 영상의 중복도와 촬영 각도의 조합에 따라 크게는 0.153 m에서 작게는 0.102 m로 나타났다. 더욱 높은 수준의 항공삼각측량 정확도를 확보하기 위해서는 연직 영상의 중복도를 높이는 것이 일반적이나 경사 영상을 추가하는 것이 효과적인 것으로 나타났다. 따라서 대응 단계의 재난/재해 현장에서 긴급하게 무인기 매핑을 수행할 경우 중복도를 높이기 보다는 경사 영상도 함께 취득할 필요가 있다.

• 대한건축학회논문집:계획계
• /
• 제34권11호
• /
• pp.85-94
• /
• 2018

This study aims to investigate how easy pedestrians get around within/through the "Apartment Complexes (AC), " a common style of high-rise multi-family housing in Korea. Over the past six decades, the AC has been the most conventional way to provide standardized housing efficiently to address the problems of the shortage of housing and the substandard housing, due to the explosion of urban population with the rapid industrialization. The AC is a huge chunk of homeogenous multi-family housing, mostly condos with decent infrastructure, including parks, pedestrian passages, schools, ect. Both in the new town development and urban renewal programs have utilized the advantages of the AC. Since the design principals of AC tend to adopt the "protective design" to prevent cars and pedestrians coming outside from passing it, it has been criticised for dissecting the continuity of socioeconomic context in neighborhoods. The neo-traditional planning urbanists, including Jane Jacobs, emphasize that smaller blocks and grid road newtworks are the key in improving social, cultural, and economic vitality of the neighborhoods, because these design concepts allow more pedestrians and different types of people to be mixed in a neighborhood. In this study, we first adopted objective measures for pedestrian accessibility and pedestrian efficiency. These measures were used to calculate the lengths of shortest paths from residential buildings to the edges of AC. We tested the difference in shortest paths between the current pedestrian networks of AC and hypothetical grid networks on the AC, and the relative difference is considered as the pedestrian efficiency, using the network analysis function of Geographic Information Systems (GIS) and Python programming. We found from the randomly selected 30 ACs that the existing non-grid road networks in ACs are worse than the hypothesized grid networks, in terms of pedestrian efficiency. In average, pedestrians in AC with the conventional road networks have to walk than 25%, 26%, and 27% longer than the networks of $125{\times}45m$, $100{\times}45m$, and $75{\times}45m$, respectively. With the t-test analysis, we found the pedestrian efficiency of AC with the conventional network is lower than grid-networks. Many new urbanists stress, easiness of walking is one of the most import elements for community building and social bonds. With the findings from the objective measures of pedestrian accessibility and efficiency, the AC would have limitations to attract people outside into the AC itself, which would increase dis-connectivity with adjacent areas.