‘모델하우스 단지’ 현상과 주택 정책의 관계 연구- 1970~1990년대 주택공급 및 택지개발정책들을 중심으로-

정수환,백진 한국문화공간건축학회 2020 한국문화공간건축학회논문집 Vol.- No.72

Since 1970, the Model-house has been created by the “pre-sale” system which is a special housing supply method of Korea. Under the situation in which apartment housing became the mainstream of Korean residence, not only the type of model-house has diversified, but also the ‘model-house clustered area’ has started to appear. The purpose of this paper is to discover what ‘the phenomenon of model-house cluster’ is and why it occurs. This paper finds out the agents of shaping the ‘model-house clusters’ from a literature review of laws, policies, and newspapers related to the ‘model-house cluster’ focused on the period from the 1970s to 1990s, when the model-house cluster started to appear. By doing so, this study uncovers the relationship between the formation of the ‘model-house cluster’ and the housing policies of the Korean government, especially “the Housing Site Development”, “the First New-town Development”, and “Simultaneous Sales” which fulfilled the condition of forming the ‘model-house cluster’. And this research confirms that the ‘model-house cluster’ appeared in all five cities of first new-town development after these policies were carried out. Also, it gives the importance of the 'model-house cluster' in three aspects: 1) the Korean government found the ‘model-house cluster’ important to housing supply in that period, 2) it caused a competition which led to the evolution of model-houses, 3) it entailed the advent of “the mega-scale collective model-house”. This study is significant in that it defines an unprecedented understanding of the phenomenon of ‘the model-house cluster’ and it expands the perspective of the model-house from a single facility into a collective and urban institution.

이미지 기반 AI 모델의 품질 고도화를 위한 이미지 증강 및 테스트 방법의 융복합 연구

서정민,이우진 한국전시산업융합연구원 2023 한국과학예술융합학회 Vol.41 No.4

This study was initiated to solve the difficulty of constructing a test set without bias for each image feature to verify the quality of an image-based artificial intelligence model. In previous studies, it was difficult to use because a large amount of labeling data was required for model learning. The purpose of this study is to improve the performance of artificial intelligence models by organizing learning and test datasets without bias and testing them by cluster without manual classification or labeling. Two experiments were conducted on the artificial intelligence model DC-VTON of the virtual try-on and the results were compared. First, the test set was divided into five clusters each using two clustering methods and the model was tested. In the two methods, the deviation of the model quality indicator SSIM values of the clusters was compared. The deviation of convolutional neural network-based clustering was 0.022 larger than that of pixel value-based clustering, resulting in more differences in performance by cluster. Second, two experiments were conducted to add re-learning data for each cluster to reach the target value of the model's quality performance index. Comparing the SSIM average of clusters that had low performance with after re-learning, the differential addition method by cluster increased by 3.2% and the equal addition method by cluster increased by 0.64%. In clusters with low quality, complementing the learning data resulted in improved performance improvement. Based on the results of these studies, it is expected that the quality of image-based artificial intelligence models can be improved by clustering images, which are unstructured data, according to characteristics, to refine test sets and enhance cluster-specific test results.

클러스터 적응주기 모델에 대한 비판적 검토

전지혜 ( Jihye Jeon ),이철우 ( Chulwoo Lee ) 한국경제지리학회 2017 한국경제지리학회지 Vol.20 No.2

본 연구는 클러스터 진화의 분석에 있어서 클러스터 적응주기 모델의 의의와 한계점을 비판적으로 검토하고, 이를 토대로 향후 클러스터 진화 분석을 위한 연구 과제를 제시하고자 하였다. 1980년대 이전까지 클러스터를 비롯한 산업집적지 연구는 특정 시점에서 경제 공간의 양상에 주목하는 `정태적 관점`을 기초로 이루어졌지만, 최근에는 `복잡적응계`의 `진화`에 주목하는 `동태적 연구`로 패러다임이 전환되었다. 이에 역동적·지속적으로 진화하는 클러스터에 적절한 분석도구로 적응주기 모델이 주목받게 되었으나, 클러스터 및 그 진화의 속성에 맞게 수정 및 보완되어 클러스터적응주기 모델이 등장하게 되었다. 클러스터 적응주기 모델은 자원축적, 상호의존성 그리고 회복력의 측면에서 클러스터 진화의 특성을 규명하고, 클러스터 진화 경로를 6가지로 구분하여 살펴 볼 수 있는 포괄적인 분석틀이지만, 모델의 확대 및 심화를 위해서 이론적·경험적 연구 측면에서 더욱 활발한 논의와 보완이 요구된다. 따라서 향후 클러스터 진화 분석에 있어서의 연구 과제로는 클러스터 진화 모델의 구체화 및 정교화, 회복력 개념의 강조 그리고 경험적 연구를 통한 모델의 적용가능성과 유용성의 검증을 제시하고자 한다. This study seeks to critically examine the significance and limits of the cluster adaptive cycle model for analysis of cluster evolution and to propose research issues for future analysis of cluster evo-lution based on this critical examination. Until the 1980s, research on industrial complexes including clusters was based on a `static perspective` that focuses on the aspect of economic space at a specific point in time, but the research paradigm has recently shifted to a `dynamic perspective` focusing on `evolution` of `complex adaptive systems`. As a result, the adaptive cycle model has attracted attention as an analysis tool of dynamically evolving clusters. However, the cluster adaptive cycle model has emerged by being ap-propriately modified and expanded according to the properties of the cluster and its evolution. The cluster adaptive cycle model is a comprehensive analysis framework that identifies the characteristics of cluster evolution in terms of resource accumulation, interdependence, and resilience and classifies cluster evolu-tion paths into six different categories. Nevertheless, there is still a need for further discussion and supple-mentation in terms of theoretical and empirical research to expand and deepen the model. Therefore, research issues for future analysis of cluster evolution are to specify and elaborate the cluster evolution model, to emphasize the concept of resilience, and to verify the applicability and usefulness of the model through empirical research.

부가정보를 이용한 조건부 Fuzzy C-Regression Model 클러스터링 알고리즘 기반 RBFNN 모델 설계

박상범(Sang-Beom Park),오성권(Sung-Kwun Oh) 한국지능시스템학회 2020 한국지능시스템학회논문지 Vol.30 No.2

Fuzzy C-Regression Model(FCRM) 클러스터링은 Fuzzy C-Means(FCM) 클러스터링과는 다르게 각 클러스터의 중심을 직선으로 나타내는 방법이다. 데이터의 실제 출력값을 사용한다는 것이 FCRM 클러스터링의 약점으로 작용한다. 본 논문에서는, 실제 출력값을 사용하지 않기 위해 FCRM 클러스터링의 목적함수를 개선하였고, 개선된 FCRM 클러스터링에 부가정보가 추가된 조건부 FCRM 클러스터링 기반 방사형 기저함수 신경회로망(Radial Basis Function Neural Network, RBFNN) 모델을 설계한다. 조건부 FCRM 클러스터링의 부가정보는 데이터의 실제 출력과 FCRM 클러스터링 기반 RBFNN 모델의 출력 차이(오차)의 절댓값을 정규화한 값을 사용한다. 조건부 FCRM 클러스터링을 사용하여 구한 멤버쉽값 (소속정도)은 은닉층 노드의 출력값으로 이용된다. 은닉층과 출력층 사이의 연결가중치는 최소제곱 추정 (Least Square Estimation, LSE)을 사용하여 학습한다. 제안된 모델의 성능지수는 평균 제곱근 오차(Root Mean Squared Error, RMSE)를 적용하여 평가한다. 조건부 FCRM 클러스터링 기반 RBFNN 모델은 다양한 모델의 성능과 비교해서 우수성을 보여준다. Unlike fuzzy c-means(FCM) clustering, fuzzy c-regression model(FCRM) clustering is a method that represents the center of each cluster as a line. The thing utilizing the actual value of the output data acts as a weakness of the FCRM clustering. In this paper, we improve the objective function of the FCRM clustering in order not to use the actual output value, and design a radial basis function neural network(RBFNN) model using conditional fuzzy c-regression model(CFCRM) clustering which added auxiliary information into the advanced FCRM clustering. The auxiliary information of CFCRM clustering uses a value, which normalizing absolute value of difference(error) of the actual output of the data and the output of the FCRM clustering-based RBFNN model. The membership grades(degree of belonging) obtained by using CFCRM clustering are exploited as the outputs of the nodes of hidden layer. The connection weights between hidden layer and output layer are trained with the aid of least square estimation(LSE). By applying root mean squared error (RMSE), the performance index of the proposed model is evaluated. CFCRM clustering-based RBFNN model shows the superiority when compared to the performance of the various models.

무선 센서 네트워크에서 클러스터 그룹 모델을 이용한 에너지 절약 방안

김진수(Kim, Jin-Su) 한국산학기술학회 2010 한국산학기술학회논문지 Vol.11 No.12

무선 센서 네트워크에서 클러스터링 기법은 클러스터를 형성하여 데이터를 통합한 후 한 번에 전송해서 에 너지를 효율적으로 사용하는 기법이다. 클러스터 그룹 모델은 클러스터링에 기반을 두지만 이전의 기법과 달리 클러 스터 헤드에 집중된 에너지 과부하를 클러스터 그룹 헤드와 클러스터 헤드로 분산시켜서 전체 에너지 소모량을 줄인 다. 본 논문에서는 이러한 클러스터 그룹 모델에서 에너지 소모 모델의 임계값에 따라 최적의 클러스터 그룹 수와 클 러스터 수를 구하고 이를 이용하여 센서 네트워크 전체 에너지 소모량을 최소화하고 네트워크 수명을 최대화한다. 실 험을 통하여 제안된 클러스터 그룹 모델이 이전의 클러스터링 기법보다 네트워크 에너지 효율이 향상되었음을 보였다. Clustering method in wireless sensor network is the technique that forms the cluster to aggregate the data and transmit them at the same time that they can use the energy efficiently. Even though cluster group model is based on clustering, it differs from previous method that reducing the total energy consumption by separating energy overload to cluster group head and cluster head. In this thesis, I calculate the optimal cluster group number and cluster number in this kind of cluster group model according to threshold of energy consumption model. By using that I can minimize the total energy consumption in sensor network and maximize the network lifetime. I also show that proposed cluster group model is better than previous clustering method at the point of network energy efficiency.

공유모델 인식 성능 향상을 위한 효율적인 연속 어휘 군집화 모델링

안찬식(Chan-Shik Ahn),오상엽(Sang-Yeob Oh) 한국컴퓨터정보학회 2010 韓國컴퓨터情報學會論文誌 Vol.15 No.1

연속 어휘 인식 시스템에서는 통계적 방법에 의한 어휘 인식을 수행하기 위하여 확률분포를 이용하며 이는 음소 단위의클러스터링을사용하여모델링하여샘플들을기반으로 확률 파라미터를 추정한다. 어휘 검색 시 추정된 확률 파라미터로부터 인식 결과를 나타내는데 미리 정의되지 않은 음소와 추가되어진 음소로부터 인식률이 저하되는 문제점이 발생하며, 하나의 클러스터링으로 모델링하므로 가우시안 모델이 정확성을 확보하지 못한다는 단점이 있다. 이를 개선하기 위하여 확률 분포의 혼합 가우시안 모델을 최적화하여 유사도를 기반으로 Euclidean과 Bhattacharyya 거리 측정 방법을 혼합한 군집화 모델을 제안하고, 군집화된 모델에서 음소 단위로 확률 모델을 탐색할 수 있는 시스템을 모델링하였다. 본 논문에서 제안한 시스템을 적용한 결과 시스템 성능에서 어휘 종속 인식률은 98.63%, 어휘 독립 인식률은 97.91%의 인식률을 나타내었다. In continuous vocabulary recognition system by statistical method vocabulary recognition to be performed using probability distribution it also modeling using phoneme clustering for based sample probability parameter presume. When vocabulary search that low recognition rate problem happened in express vocabulary result from presumed probability parameter by not defined phoneme and insert phoneme and it has it's bad points of gaussian model the accuracy unsecure for one clustering modeling. To improve suggested probability distribution mixed gaussian model to optimized for based resemble Euclidean and Bhattacharyya distance measurement method mixed clustering modeling that system modeling for be searching phoneme probability model in clustered model. System performance as a result of represent vocabulary dependence recognition rate of 98.63%, vocabulary independence recognition rate of 97.91%.

Implementation of delivery time prediction model that combines clustering and machine learning

Deok Ho An,So Yeon Woo,Da Woon Jeong,Yeong Hyeon Gu,Seong Joon Yoo 한국차세대컴퓨팅학회 2021 한국차세대컴퓨팅학회 학술대회 Vol.2021 No.11

Although there have been studies using various algorithms on the delivery time prediction in the logistics business, those studies did not reflect various features such as region or product. In the case of delivery time prediction of a single model that does not reflect the features, the accuracy of delivery time prediction for a region with a high distribution is high, but the prediction accuracy is low for a region with a low distribution. To solve this problem, this paper proposes a method of classifying logistic patterns using clustering and selecting an optimal model for each logistic pattern. The proposed method consists of four steps. First, the derived variables such as reception day, delivery speed and delivery distance are created. Second, the data with the same pattern goes through clustering using K-means. Third, by comparing the performance of each model using six regression algorithms for each classified logistic pattern, an optimal model is selected and the model is stored. Lastly, the logistic pattern of the data to be predicted is classified and the optimal model stored for each pattern is loaded, and the result of delivery time prediction is provided through the model. Two experiments were performed to verify the proposed method. The e-commerce data from Brazil is used as verification data. From the experiment, the delivery time prediction error of the proposed model was smaller than that of the single regression model.

A Conceptual Model of Port Clusters and Related Assemblages

Hong-Seung Roh 한국항해항만학회 2006 한국항해항만학회지 Vol.30 No.2

Recently we have seen a noticeable trend in ports to establish port clusters. Despite this trend, little research bas actually been undertaken to fully define and analyse port clusters. A couple of exceptions have descriptively defined port clusters but the boundaries of port clusters are not apparent. Therefore, the aim of this paper is to define port clusters in terms of set theory and in particular look at them in terms of their distinct characteristics and system boundaries. The main concern of this paper is that there is a need to distinguish, from a system and a competition perspective, between port clusters, ports, maritime clusters and port ranges. This paper proposes a conceptual model relevant to the relationship among port clusters related assemblages and that las been applied to the north western europe region This model suggests six levels of competition that will help port authorities and government to develop appropriate policies and strategies for port operation and port industry.

파렛트 이동량에 대한 시간 단위 군집화 예측 모델

문현지 ( Hyunji Moon ),송보미 ( Bomi Song ) 한국로지스틱스학회 2019 로지스틱스연구 Vol.27 No.4

데이터 수집 및 처리 기술이 발전하면서 빅데이터 분석 및 활용이 증가하고 있다. 물류 산업에서도 다양한 데이터가 더 높은 빈도로 수집되고 있으며, 이에 따라 데이터 분석 역시 더욱 세분화된 하위집단을 대상으로 시도되고 있다. 물류 데이터는 대표적인 시계열 데이터로서, 시계열 데이터에서 이러한 경향은 복잡한 계절성과 다양한 설명변수와 같은 새로운 데이터 특성을 의미한다. 그러나 기존의 많은 시계열 예측 모델은 구조적 한계로 인해 이러한 새로운 데이터 속성을 반영하지 못하여 낮은 예측 정확도를 보인다. 이러한 한계를 극복하기 위하여 본 논문은 시간 단위 군집화 모델을 제안한다. 제안한 모델은 시간 단위를 기준으로 시계열 데이터를 분류한 후 단위별로 분리된 시계열 집단들을 군집화한다. 또한 복잡한 계절성은 푸리에 방식으로, 다양한 설명변수는 베이즈 일반화된 선형 모델로 표현한다. 제안한 시간 단위 군집화 모델을 파렛트 대여기업의 실제 데이터에 적용하고, 기존의 프로펫 모델과 비교한 결과 더 높은 정확도로 예측값을 산출함을 확인하였다. As data collection and processing technologies evolve, big data analysis and utilization efforts are increasing. In the logistics industry, data are being collected with higher frequency and diversity, and analysis is conducted on subdivided groups. Logistics data are generally time series data, and such trends in time series data suggest new data characteristics: complex seasonality and various explanatory variables. However, a lot of existing time series models fail to reflect these new characteristics due to structural limitations, resulting in low prediction accuracy. To overcome this limitation, this paper proposes a time unit clustering model. This model classifies time series data based on time units, days of the week for example, and then clusters time series groups. In addition, Fourier series regression is used to model complex seasonality, and various explanatory variables are modeled through a Bayesian generalized linear model. The proposed model was applied to real data from the pallet rental company. This logistics flow amount data displayed the characteristics of recent time series data. Compared with prophet models, the proposed model yielded predictions with higher accuracy resulting from its capability to model the difference between time units through clustering.

자기구성 클러스터링 기반 뉴로-퍼지 모델링

김승석(Sung-Suk Kim),유정웅(Jeong-Woong Ryu),김용태(Yong-Tae Kim) 한국지능시스템학회 2005 한국지능시스템학회논문지 Vol.15 No.6

본 논문에서는 클러스터링을 뉴로-퍼지 모델에 직접 적용하여 모델을 최적화하는 방법을 제안하였다. 기존의 오차미분기반 학습을 통한 뉴로-퍼지 모델의 최적화 과정과는 달리 제안된 방법은 클러스터링 학습과 연계하여 모델을 구성하며 자율적으로 클러스터의 수를 추정하며 동시에 최적화를 수행한다. 순차적인 학습 기법에서는 각각의 학습 기법을 따로 적용하여 모델링을 실시하였으나 제안된 기법에서는 하나의 클러스터링 학습으로 전체 모델의 학습을 실시하였다. 또한 제안된 방법에서는 클러스터링이 수렴하는 만큼 전체 모델의 연산량이 감소하여 학습과정에서 발생하는 연산량 문제를 개선하였다. 시뮬레이션을 통하여 기존의 연구 결과들과 비교하여 제안된 기법의 유용성을 보였다. In this paper, we propose a new neuro-fuzzy modeling using clustering-based learning method. In the proposed clustering method, number of clusters is automatically inferred and its parameters are optimized simultaneously, Also, a neuro-fuzzy model is learned based on clustering information at same time. In the previous modelling methods, clustering and model learning are performed independently and have no exchange of its informations. However, in the proposed method, overall neuro-fuzzy model is generated by using both clustering and model learning, and the information of modelling output is used to clustering of input. The proposed method improve the computational load of modeling using Subtractive clustering method. Simulation results show that the proposed method has an effectiveness compared with the previous methods.