• 제목/요약/키워드: multiple classification analysis

검색결과 468건 처리시간 0.024초

Analysis of Weights and Feature Patterns in Popular 2D Deep Neural Networks Models for MRI Image Classification

  • Khagi, Bijen;Kwon, Goo-Rak
    • Journal of Multimedia Information System
    • /
    • 제9권3호
    • /
    • pp.177-182
    • /
    • 2022
  • A deep neural network (DNN) includes variables whose values keep on changing with the training process until it reaches the final point of convergence. These variables are the co-efficient of a polynomial expression to relate to the feature extraction process. In general, DNNs work in multiple 'dimensions' depending upon the number of channels and batches accounted for training. However, after the execution of feature extraction and before entering the SoftMax or other classifier, there is a conversion of features from multiple N-dimensions to a single vector form, where 'N' represents the number of activation channels. This usually happens in a Fully connected layer (FCL) or a dense layer. This reduced 2D feature is the subject of study for our analysis. For this, we have used the FCL, so the trained weights of this FCL will be used for the weight-class correlation analysis. The popular DNN models selected for our study are ResNet-101, VGG-19, and GoogleNet. These models' weights are directly used for fine-tuning (with all trained weights initially transferred) and scratch trained (with no weights transferred). Then the comparison is done by plotting the graph of feature distribution and the final FCL weights.

동일 어종(갈치, 참조기) 어획에 대한 다수 어업별 온실가스 배출량 정량적 분석 (A quantitative analysis of greenhouse gases emissions by multiple fisheries for catching the same species (hairtail and small yellow croaker))

  • 강경미;이지훈;신동원
    • 수산해양기술연구
    • /
    • 제57권2호
    • /
    • pp.149-161
    • /
    • 2021
  • The concern on the greenhouse gas emission is strongly increasing globally. In fishery industry section, the greenhouse gas emissions are an important issue according to The Paris Climate Change Accord in 2015. The Korean government has a plan to reduce the GHG emissions as 4.8% compared to the BAU in fisheries until 2020. Furthermore, the Korean government has also declared to achieve the carbon neutrality in 2050 at the Climate Adaptation Summit 2021. However, the investigation on the GHG emissions from Korean fisheries did not carry out extensively. Most studies on GHG emissions from Korean fishery have dealt with the GHG emissions by fishery classification so far. However, follow-up studies related to GHG emissions from fisheries need to evaluate the GHG emission level by species to prepare the adoption of environmental labels and declarations (ISO 14020). The purpose of this research is to investigate which degree of GHG emitted to produce the species (hairtail and small yellow croaker) from various fisheries. Here, we calculated the GHG emission to produce the species from the fisheries using the Life Cycle Assessment method. The system boundary and input parameters for each process level are defined for the LCA analysis. The fuel use coefficients of the fisheries for the species are also calculated according to the fuel type. The GHG emissions from sea activities by the fisheries will be dealt with. Furthermore, the GHG emissions for producing the unit weight species and annual production are calculated by fishery classification. The results will be helpful to understand the circumstances of GHG emissions from Korean fisheries.

구인구직사이트의 구인정보 기반 지능형 직무분류체계의 구축 (Development of Intelligent Job Classification System based on Job Posting on Job Sites)

  • 이정승
    • 지능정보연구
    • /
    • 제25권4호
    • /
    • pp.123-139
    • /
    • 2019
  • 주요 구인구직사이트의 직무분류체계가 사이트마다 상이하고 SW분야에서 제안한 'SQF(Sectoral Qualifications Framework)'의 직무분류체계와도 달라 SW산업에서 SW기업, SW구직자, 구인구직사이트가 모두 납득할 수 있는 새로운 직무분류체계가 필요하다. 본 연구의 목적은 주요 구인구직사이트의 구인정보와 'NCS(National Competaency Standars)'에 기반을 둔 SQF를 분석하여 시장 수요를 반영한 표준 직무분류체계를 구축하는 것이다. 이를 위해 주요 구인구직사이트의 직종 간 연관분석과 SQF와 직종 간 연관분석을 실시하여 직종 간 연관규칙을 도출하고자 한다. 이 연관규칙을 이용하여 주요 구인구직사이트의 직무분류체계를 맵핑하고 SQF와 직무 분류체계를 맵핑함으로써 데이터 기반의 지능형 직무분류체계를 제안하였다. 연구 결과 국내 주요 구인구직사이트인 '워크넷,' '잡코리아,' '사람인'에서 3만여 건의 구인정보를 open API를 이용하여 XML 형태로 수집하여 데이터베이스에 저장했다. 이 중 복수의 구인구직사이트에 동시 게시된 구인정보 900여 건을 필터링한 후 빈발 패턴 마이닝(frequent pattern mining)인 Apriori 알고리즘을 적용하여 800여 개의 연관규칙을 도출하였다. 800여 개의 연관규칙을 바탕으로 워크넷, 잡코리아, 사람인의 직무분류체계와 SQF의 직무분류체계를 맵핑하여 1~4차로 분류하되 분류의 단계가 유연한 표준 직무분류체계를 새롭게 구축했다. 본 연구는 일부 전문가의 직관이 아닌 직종 간 연관분석을 통해 데이터를 기반으로 직종 간 맵핑을 시도함으로써 시장 수요를 반영하는 새로운 직무분류체계를 제안했다는데 의의가 있다. 다만 본 연구는 데이터 수집 시점이 일시적이기 때문에 시간의 흐름에 따라 변화하는 시장의 수요를 충분히 반영하지 못하는 한계가 있다. 계절적 요인과 주요 공채 시기 등 시간에 따라 시장의 요구하는 변해갈 것이기에 더욱 정확한 매칭을 얻기 위해서는 지속적인 데이터 모니터링과 반복적인 실험이 필요하다. 본 연구 결과는 향후 SW산업 분야에서 SQF의 개선방향을 제시하는데 활용될 수 있고, SW산업 분야에서 성공을 경험삼아 타 산업으로 확장 이전될 수 있을 것으로 기대한다.

Hierarchical Attention Network를 활용한 주제에 따른 온라인 고객 리뷰 분석 모델 (Analysis of the Online Review Based on the Theme Using the Hierarchical Attention Network)

  • 장인호;박기연;이준기
    • 한국IT서비스학회지
    • /
    • 제17권2호
    • /
    • pp.165-177
    • /
    • 2018
  • Recently, online commerces are becoming more common due to factors such as mobile technology development and smart device dissemination, and online review has a big influence on potential buyer's purchase decision. This study presents a set of analytical methodologies for understanding the meaning of customer reviews of products in online transaction. Using techniques currently developed in deep learning are implemented Hierarchical Attention Network for analyze meaning in online reviews. By using these techniques, we could solve time consuming pre-data analysis time problem and multiple topic problems. To this end, this study analyzes customer reviews of laptops sold in domestic online shopping malls. Our result successfully demonstrates over 90% classification accuracy. Therefore, this study classified the unstructured text data in the semantic analysis and confirmed the practical application possibility of the review analysis process.

Development and testing of a composite system for bridge health monitoring utilising computer vision and deep learning

  • Lydon, Darragh;Taylor, S.E.;Lydon, Myra;Martinez del Rincon, Jesus;Hester, David
    • Smart Structures and Systems
    • /
    • 제24권6호
    • /
    • pp.723-732
    • /
    • 2019
  • Globally road transport networks are subjected to continuous levels of stress from increasing loading and environmental effects. As the most popular mean of transport in the UK the condition of this civil infrastructure is a key indicator of economic growth and productivity. Structural Health Monitoring (SHM) systems can provide a valuable insight to the true condition of our aging infrastructure. In particular, monitoring of the displacement of a bridge structure under live loading can provide an accurate descriptor of bridge condition. In the past B-WIM systems have been used to collect traffic data and hence provide an indicator of bridge condition, however the use of such systems can be restricted by bridge type, assess issues and cost limitations. This research provides a non-contact low cost AI based solution for vehicle classification and associated bridge displacement using computer vision methods. Convolutional neural networks (CNNs) have been adapted to develop the QUBYOLO vehicle classification method from recorded traffic images. This vehicle classification was then accurately related to the corresponding bridge response obtained under live loading using non-contact methods. The successful identification of multiple vehicle types during field testing has shown that QUBYOLO is suitable for the fine-grained vehicle classification required to identify applied load to a bridge structure. The process of displacement analysis and vehicle classification for the purposes of load identification which was used in this research adds to the body of knowledge on the monitoring of existing bridge structures, particularly long span bridges, and establishes the significant potential of computer vision and Deep Learning to provide dependable results on the real response of our infrastructure to existing and potential increased loading.

인터넷중독 측정도구와 한국형 인터넷중독지표의 개발 (Development of Internet Addiction Measurement Scales and Korean Internet Addiction Index)

  • 박재성
    • Journal of Preventive Medicine and Public Health
    • /
    • 제38권3호
    • /
    • pp.298-306
    • /
    • 2005
  • Objectives : To develop measurement scales of Internet addiction, and propose a Korean Internet Addiction Index (K-IAI) and classification criteria for Internet addiction from the threshold scores developed. Methods : The identification of the concept of 'Internet addiction' was based on the literature review. To select the scales, an exploratory factor analysis was applied. A construct validation was tested by a confirmatory factor analysis (CFA) with a structured equation model (SEM). In testing the validity of the classification criteria, ANOVA and non-recursive models with SEM were applied. Results : Out of 1,080 questionnaires distributed, 1,037 were returned,; a response rate of 96%. The Cronbach-$\alpha$ of all items was over 0.75. Using an exploratory factor analysis in the condition of a 6 factor constrain as the study model proposed, 23 of the initial 28 items were identified. In testing the discriminant and convergent validity of the selected 23 scales using CFA with SEM, the Internet addiction model explained about 93% of all variances of the data collected, and all the latent variables significantly explained the designated scales. A K-IAI was proposed using the T-scores of the sum of all factor averages. In the classification of users, the basic concept was a twostandard deviation approach of the K-IAI as the criteria of MMPI. The addiction group had a score ${\geq}70$ in the K-IAI, the pre-addiction group between ${\geq}50$ and <70, and the average user group <50. The Internet use times of the classified groups were statistically different in the ANOVA and multiple comparisons. Conclusions : The K-IAI is a reliable and valid instrument for measuring Internet addiction. Moreover, the taxonomy of the groups was also verified using various methods.

Development of a Novel Endoscopic Scoring System to Predict Relapse after Surgery in Intestinal Behçet's Disease

  • Park, Jung Won;Park, Yehyun;Park, Soo Jung;Kim, Tae Il;Kim, Won Ho;Cheon, Jae Hee
    • Gut and Liver
    • /
    • 제12권6호
    • /
    • pp.674-681
    • /
    • 2018
  • Background/Aims: The cumulative surgery rate and postoperative relapse of intestinal Behçet's disease (BD) have been reported to be high. This study aimed to establish a scoring system based on follow-up endoscopic findings that can predict intestinal BD recurrence after surgery. Methods: Fifty-four patients with intestinal BD who underwent surgery due to bowel complications and underwent follow-up colonoscopy were retrospectively investigated. Their clinical data, including colonoscopic findings, were retrieved. Classification and regression tree analysis was used to develop an appropriate endoscopic classification model that can explain the postsurgical recurrence of intestinal BD most accurately based on the following classification: e0, no lesions; e1, solitary ulcer <20 mm in size; e2, solitary ulcer ${\geq}20mm$ in size; and e3, multiple ulcers regardless of size. Results: Clinical relapse occurred in 37 patients (68.5%). Among 38 patients with colonoscopic recurrence, only 29 patients had clinically relapsed. Multivariate analysis identified higher disease activity index for intestinal BD at colonoscopy (hazard ratio [HR], 1.013; 95% confidence interval [CI], 1.005 to 1.021; p=0.002) and colonoscopic recurrence (HR, 2.829; 95% CI, 1.223 to 6.545; p=0.015) as independent risk factors for clinical relapse of intestinal BD. Endoscopic findings were classified into four groups, and multivariate analysis showed that the endoscopic score was an independent risk factor of clinical relapse (p=0.012). The risk of clinical relapse was higher in the e3 group compared to the e0 group (HR, 6.284; 95% CI, 2.036 to 19.391; p=0.001). Conclusions: This new endoscopic scoring system could predict clinical relapse in patients after surgical resection of intestinal BD.

Unsupervised one-class classification for condition assessment of bridge cables using Bayesian factor analysis

  • Wang, Xiaoyou;Li, Lingfang;Tian, Wei;Du, Yao;Hou, Rongrong;Xia, Yong
    • Smart Structures and Systems
    • /
    • 제29권1호
    • /
    • pp.41-51
    • /
    • 2022
  • Cables are critical components of cable-stayed bridges. A structural health monitoring system provides real-time cable tension recording for cable health monitoring. However, the measurement data involve multiple sources of variability, i.e., varying environmental and operational factors, which increase the complexity of cable condition monitoring. In this study, a one-class classification method is developed for cable condition assessment using Bayesian factor analysis (FA). The single-peaked vehicle-induced cable tension is assumed to be relevant to vehicle positions and weights. The Bayesian FA is adopted to establish the correlation model between cable tensions and vehicles. Vehicle weights are assumed to be latent variables and the influences of different transverse positions are quantified by coefficient parameters. The Bayesian theorem is employed to estimate the parameters and variables automatically, and the damage index is defined on the basis of the well-trained model. The proposed method is applied to one cable-stayed bridge for cable damage detection. Significant deviations of the damage indices of Cable SJS11 were observed, indicating a damaged condition in 2011. This study develops a novel method to evaluate the health condition of individual cable using the FA in the Bayesian framework. Only vehicle-induced cable tensions are used and there is no need to monitor the vehicles. The entire process, including the data pre-processing, model training and damage index calculation of one cable, takes only 35 s, which is highly efficient.

패싯 기반 민원 다차원 분석을 위한 자동 분류 모델 (A Study on an Automatic Classification Model for Facet-Based Multidimensional Analysis of Civil Complaints)

  • 김나랑
    • 한국산업정보학회논문지
    • /
    • 제29권1호
    • /
    • pp.135-144
    • /
    • 2024
  • 시민의 의견인 민원은 다양한 사람들이 여러 주제에 대하여 반복·지속적으로 실시간 쏟아내기 때문에 담당자가 이를 읽고 분석하는데 한계가 있다. 이에 본 연구에서는 빅데이터 분석을 통해 주요 현안에 대한 여론 및 요구 사항을 파악하기 위하여 정성적인 분석에 패싯을 기반으로 한 정량적인 다차원 분석을 위한 자동 분류 모델을 제안하였다. 구체적으로 첫째, 패싯 이론과 정치분석모형을 기반으로 민원 특성을 분석하고 이를 정책 단계에 활용할 수 있는 새로운 분류 프레임워크를 제시하였다. 둘째, 민원 분석 및 처리에 따른 행정 업무를 감소시키고, 시민들의 정책참여를 용이하게 하기 위해 딥러닝을 활용하여 패싯 분석 프레임에 의해 자동으로 속성을 추출하고 분류 하였다. 본 연구결과는 학문적으로 민원 빅데이터의 특성을 이해하고 분석하는데 중요한 단초를 제공하여 향후 많은 후속 연구를 창출할 수 있을 것으로 기대되며, 공공분야를 넘어 교육, 산업, 의료 등 다른 분야에서의 비정형 데이터의 계량화를 위한 가이드 라인과 다차원 분석의 활용에 대한 이론적 근거를 제시할 수 있다. 실무적으로 대용량 전자 민원에 대한 처리체계 개선 및 딥러닝을 통한 자동화로 민원처리 업무의 효율성과 신속성을 높일 수 있으며, 다른 분야의 텍스트 데이터의 처리에 활용될 수 있을 것이다.

경남 일부지역의 농부증 및 하우스증 발생에 미치는 요인분석 (A Study on the causes of farmer's disease and greenhouse disease in a rural area of Kyungnam province)

  • 홍대용;김장락;이명순;강경희;하호성
    • 농촌의학ㆍ지역보건
    • /
    • 제21권2호
    • /
    • pp.173-193
    • /
    • 1996
  • This study was carried out to elucidate the causes of farmer's disease and greenhouse disease in the rural area of Kyungnam province during 1996. 2,171 (69.1%) of the 3,140 adults above 20 years old who had lived in the selected 20 villages in Uiryong County of Kyungnam Province were selected in order to over both residents who live in the green house distracts and conventional rural area. The results were as follows: 1. Among the subjects, the male was 42.2% and female was 57.6%. The average age for the male was 52.3, 55.6 for the female. 2. The proportion of the farmers in the subjects was 81.5%. Among these 78.0% were engaged in the greenhouse farming. 3. Among the eight symptoms of Nofusho(japanese farmer's comples), lumbago was the most frequently complained and followed by shoulder stiffness and parenthesis of hand or foot. 4. The total score of farmer's disease was evidently higher in the female and the older the score was clearly bigger. 5. The prevalence of Nofusho(Japanese farmer's comples)was 26.8% and 43.15% was for the suspected Nofusho. In the male, the prevalence of Nofusho was 13.6% and 40.9% for the suspected Nofusho and in the female 36.5% and 44.7% were shown, comparatively. 6. According to the multiple classification analysis, sex, age, and occupation were selected as significant variables to explain the total score of Nofusho. 7. The correlations between the total score of Nofusho and the number of sick day, working years, and age were significant, comparatively. 8. According to the multiple classification analysis corolling interaction between independent variables, age was the only variable which was significant in the male and age, pesticide work in the female. 9. The score of greenhouse disease was highest in the group who engaged in greenhouse farming and conventional farming(2.76 for male, 3.77 for female) followed by the group who engaged only in greening house farming(2.66 for male, 3.49 for female) and by the group who engaged only in conventional farming(2.27 for male, 3.05 for female) 10. According to the multiple classification analysis with the total score of greenhouse disease as dependent variable, corolling interaction between independent variables, age and pesticide work were revealed as significant variables in the male, while, pesticide work and farming type were significant in the female. According to the above results, the following could be suggested. Because lumbago, shoulder stiffness, paresthesia of hand and foot were the most frequently complained symptoms in the respondent, the development of farming tool to reduce the body burden and periodical physical exercise and rest is highly recommended. It is revealed that both in the farmer' disease and greenhouse disease the score was higher in the female than in the male. So the reasonables measures is recommended to reduce the working hours of the female. Pesticide work was revealed as the significant variable in the female in farmer's disease and both in the male and the female in greenhouse disease. So the development of the safe method of pesticide spraying including safety education should be introduced. Particularly the female should be excluded in pesticide spraying.

  • PDF