• 디지털융복합연구
• /
• 제16권12호
• /
• pp.217-229
• /
• 2018

이 연구에서는 미분양 아파트 단지의 세대별 계약 자료에 데이터 마이닝 기법인 의사결정나무, 신경망, 로지스틱 모형을 적용하여 세대별 초기계약을 예측하는 모형을 개발한다. 모형 개발에는 위례신도시 미분양 아파트 단지의 계약 자료가 이용되며, 이 자료는 훈련용 자료와 검정용 자료로 분할되어 분석에 투입된다. 훈련용 자료에서는 신경망, 의사결정나무, 로지스틱 모형 순으로 예측력이 뛰어났지만 검정용 자료에서는 로지스틱 모형이 가장 우수하게 나타났다. 이 같은 결과는 신경망이 훈련용 자료에 최적화된 모형으로 구축되면서 검정용 자료에 대한 적응성이 떨어져 나타난 결과로 판단된다. 의사결정나무와 로지스틱 모형을 병행 적용한 결과, 층수, 향, 세대 위치, 전기 및 발전기실의 소음, 청약자 거주지, 청약 종류가 초기계약에 영향을 주는 것으로 나타났다. 이는 두 가지 모형을 같이 사용하는 것이 초기계약 결정요인 발굴에 더 효과적이라는 것을 의미한다. 이 연구는 데이터 마이닝의 적용 범위를 주택 분양 예측까지 확장함으로써 융복합 분야 발전에 기여하고 있다.

• 응용통계연구
• /
• 제36권6호
• /
• pp.619-629
• /
• 2023

비료는 작물 생산성을 높이는데 중요한 역할을 하지만 작물의 양분요구량을 고려하지 않은 비료 과다 사용은 농가 경영비 부담과 환경 부하를 높힐 우려가 있다. 스마트 농업을 통해 작물의 생장 특성을 반영하여 시기별로 필요한 만큼 비료를 공급하면 비료 유실에 대한 부담을 줄이고, 경제적인 양분관리 효과를 기대할 수 있다. 본 논문에서는 다양한 재배환경에서 재배한 고추 및 대파의 정식일수별 전체 건중량을 기반으로 다양한 생장곡선(로지스틱(logistic), 곰페르츠(Gompertz), 리차드(Richards), 이중 로지스틱(double logistic curve)을 활용한 비선형 모형 기반 작물 생육 모형을 적합하고, 작물 성장률에 기반한 비료시비량 분배 알고리즘을 제안하고자 한다.

• 한국경영과학회지
• /
• 제26권4호
• /
• pp.55-70
• /
• 2001

Artificial neural network(ANN) models have been widely used for the classification problems in business such as bankruptcy prediction, credit evaluation, etc. Although the application of ANN to classification of consumers＇ choice behavior is a promising research area, there have been only a few researches. In general, most of the researches have reported that the classification performance of the ANN models were better than conventional statistical model Because the survey data on consumer behavior may include much noise and missing data, ANN model will be more robust than conventional statistical models welch need various assumptions. The purpose of this paper is to study the potential of the ANN model for forecasting consumers＇ choice behavior based on survey data. The data was collected by questionnaires to the shoppers of department stores and discount stores. Then the correct classification rates of the ANN models for the training and test sample with that of multiple discriminant analysis(MDA) and logistic regression(Logit) model. The performance of the ANN models were betted than the performance of the MDA and Logit model with respect to correct classification rate. By using input variables identified as significant in the stepwise MDA, the performance of the ANN models were improved.

• 사상체질의학회지
• /
• 제32권2호
• /
• pp.10-21
• /
• 2020

Objectives Genome-wide association studies(GWAS) is a useful method to identify genetic associations for various phenotypes. The purpose of this study was to develop predictive models for Sasang constitution types using genetic factors. Methods The genotypes of the 1,999 subjects was performed using Axiom Precision Medicine Research Array (PMRA) by Life Technologies. All participants were prescribed Sasang Constitution-specific herbal remedies for the treatment, and showed improvement of original symptoms as confirmed by Korean medicine doctor. The genotypes were imputed by using the IMPUTE program. Association analysis was conducted using a logistic regression model to discover Single Nucleotide Polymorphism (SNP), adjusting for age, sex, and BMI. Results & Conclusions We developed models to predict Korean medicine constitution types using identified genectic factors and sex, age, BMI using Random Forest (RF), Support Vector Machine (SVM), and Neural Network (NN). Each maximum Area Under the Curve (AUC) of Teaeum, Soeum, Soyang is 0.894, 0.868, 0.767, respectively. Each AUC of the models increased by 6~17% more than that of models except for genetic factors. By developing the predictive models, we confirmed usefulness of genetic factors related with types. It demonstrates a mechanism for more accurate prediction through genetic factors related with type.

• Journal of the Korean Data and Information Science Society
• /
• 제21권1호
• /
• pp.21-32
• /
• 2010

그래픽적 회귀는 모형에 대한 가정을 하지 않고 회귀정보를 모두 포함하는 충분요약그림을 찾아내는 분석 방법으로 모든 회귀정보를 저차원의 그림으로 표현할 수 있게 하는 데에 그 목적이 있다. 잔차산점도를 이용한 모형의 평가는 적용 범위가 선형회귀모형에 국한되는 문제점이 있기 때문에 일반화선형모형에서는 그 대안으로 주변모형 산점도를 이용하여 모형의 적절성을 평가한다. 본 논문에서는 일반화선형모형 중에서 이진반응변수를 갖는 로지스틱모형에서의 그래픽적 회귀 방법과 주변모형 산점도를 이용한 모형평가 방법을 알아본다.

• 한국건축시공학회지
• /
• 제7권4호
• /
• pp.145-151
• /
• 2007

In this paper, tests are carried out in order to investigate the strength development of concrete under various binder types, W/B and curing temperature ranged from $5{\sim}20^{\circ}C$. Fly ash and blast furnace slag were incorporated by as much as 30%, respectively. Strength development of concrete are estimated using Logistic model and strength ratio of concrete at 28days to that at early age are also investigated. According to experimental results, it is found that good agreements are obtained between measured values and calculated ones using logistic model below $20^{\circ}C$. Strength ratio of concrete at 28days to that at early age increases in case W/B decreases and curing temperature increases. Tables and graphs for strength ratio of concrete are provided in this paper. It is capable of obtaining and predicting the periods to attain design strength by considering increment factor of strength easily with the table and graphs presented in this paper. This paper presents the reference data to decide removal time of form, time to reach target strength and strength inspection of remicon whether the test specimens meet the specified criteria of compressive strength. Multi regression models with respect to the relationship between 7days compressive strength and 28 days compressive strength depending on W/B and admixture types are presented.

• 한국IT서비스학회지
• /
• 제10권3호
• /
• pp.113-126
• /
• 2011

The smelting and the continuous casting of steel are important processes that determine the quality of steel products. Especially most of quality defects occur during solidification of the steel continuous casting process. Although quality control techniques such as six sigma, SQC, and TQM can be applied to the continuous casting process for improving quality of steel products, these techniques don't provide real-time analysis to identify the causes of defect occurrence. To solve problems, we have developed a detection model using decision tree which identified abnormal transactions to have a coarse grain structure. And we have compared the proposed model with models using neural network and logistic regression. Experiments on steel data showed that the performance of the proposed model was higher than those of neural network model and logistic regression model. Thus, we expect that the suggested model will be helpful to control the quality of steel products in real-time in the continuous casting process.

• 한국산림과학회지
• /
• 제99권2호
• /
• pp.163-171
• /
• 2010

Forest fires have potentials to change the structure and function of forest ecosystems and significantly influence on atmosphere and biogeochemical cycles. Forest fire also affects the quality of public benefits such as carbon sequestration, soil fertility, grazing value, biodiversity, or tourism. The prediction of fire occurrence and its spread is critical to the forest managers for allocating resources and developing the forest fire danger rating system. Most of fires were human-caused fires in Korea, but meteorological factors are also big contributors to fire behaviors and its spread. Thus, meteorological factors as well as social factors were considered in the fire danger rating systems. A total of 298 forest fires occurred during the fall season from 2002 to 2006 in South Korea were considered for developing a logistic model of forest fire occurrence. The results of statistical analysis show that only effective humidity and temperature significantly affected the logistic models (p<0.05). The results of ROC curve analysis showed that the probability of randomly selected fires ranges from 0.739 to 0.876, which represent a relatively high accuracy of the developed model. These findings would be necessary for the policy makers in South Korea for the prevention of forest fires.

• 한국기계가공학회지
• /
• 제19권2호
• /
• pp.111-118
• /
• 2020

Recently, bioenergy research using microalgae, one of the most promising biofuel sources, has attracted much attention. Cell disruption, which can be classified as physical or chemical, is essential to extract functional ingredients from microalgae. In this study, we investigated the cell disruption efficiency of Chlorella sp. using low-frequency non-focused ultrasound (LFNFU). This is a continuously physical method that is superior to chemical methods with respect to environmental friendliness and low processing cost. A flat panel photobioreactor was employed to cultivate Chlorella sp. and its growth curve was fitted both with Logistic and Gompertz models. The temporal change in cell reduction by cell disruption using LFNFU was fitted with a Logistic model. The experimental conditions that were investigated were the initial concentration of microalgal cells, relative amplitude of output ultrasound waves, processing volume of microalgal cells, and initial pH value. The optimal conditions for the most efficient cell disruption were determined through the various tests.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권7호
• /
• pp.3608-3628
• /
• 2017

Several block cipher modes of operation have been proposed in the literature to protect sensitive information. However, different security analysis models have been presented for attacking them. The analysis indicated that most of the current modes of operation are vulnerable to several attacks such as known plaintext and chosen plaintext/cipher-text attacks. Therefore, this paper proposes a secure block cipher mode of operation to thwart such attacks. In general, the proposed mode combines one-time chain keys with each plaintext before its encryption. The challenge of the proposed mode is the generation of the chain keys. The proposed mode employs the logistic map together with a nonce to dynamically generate a unique set of chain keys for every plaintext. Utilizing the logistic map assures the dynamic behavior while employing the nonce guarantees the uniqueness of the chain keys even if the same message is encrypted again. In this way, the proposed mode called SPCBC can resist the most powerful attacks including the known plaintext and chosen plaintext/cipher-text attacks. In addition, the SPCBC mode improves encryption time performance through supporting parallelized implementation. Finally, the security analysis and experimental results demonstrate that the proposed mode is robust compared to the current modes of operation.