• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.129-136
• /
• 2023

The study aims to examine the determinants that affect satisfaction, revisit intention, and recommendation intention with foreign tourists who visited Korea despite the threat of COVID-19. This study employs the survey data collected by the Korea Tourism Organization from 8,135 foreign tourists who visited Korea in 2020. As the survey data contains a mixture of continuous and categorical variables, decision tree analysis can ensure analytical validity for the research. According to the analytical results, the determinants affecting satisfaction are the purpose of the visit and acceptance of self-quarantine during their stay. The factors influencing revisit intention are the purpose of the visit, frequency of the visit, and acceptance of self-quarantine during their stay. The determinants affecting recommendation intention are the purpose of the visit, length of stay, and gender. Based on the results of this analysis, this study not only explains the relationship between these determinants and tourism satisfaction, revisit intention, and recommendation intention, but also suggests implications for revitalizing tourism activities.

• The Korean Journal of Applied Statistics
• /
• v.19 no.1
• /
• pp.149-161
• /
• 2006

The most important component in decision tree algorithm is the rule for split variable selection. Many earlier algorithms such as CART and C4.5 use greedy search algorithm for variable selection. Recently, many methods were developed to cope with the weakness of greedy search algorithm. Most algorithms have different selection criteria depending on the type of variables: continuous or nominal. However, ordinal type variables are usually treated as continuous ones. This approach did not cause any trouble for the methods using greedy search algorithm. However, it may cause problems for the newer algorithms because they use statistical methods valid for continuous or nominal types only. In this paper, we propose a ordinal variable selection method that uses Cramer-von Mises testing procedure. We performed comparisons among CART, C4.5, QUEST, CRUISE, and the new method. It was shown that the new method has a good variable selection power for ordinal type variables.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.4
• /
• pp.777-786
• /
• 2018

This study propose a new method to detect Cochlodinium polykrikoides on satellite images using logistic regression and decision tree. We used spectral profiles(918) extracted from red tide, clear water and turbid water as training data. The 70% of the entire data set was extracted and used for model training, and the classification accuracy of the model was evaluated by using the remaining 30%. As a result of the accuracy evaluation, the logistic regression model showed about 97% classification accuracy, and the decision tree model showed about 86% classification accuracy.

• Journal of Digital Convergence
• /
• v.16 no.11
• /
• pp.363-372
• /
• 2018

The objective of this study is to derive specific classification rules that could be used to prevent individuals with Metabolic Syndrome (MS) from developing diabetes. Specifically, we aim to identify rules which classify individuals with MS into those without diabetes (class 0) and those with diabetes (class 1). In this study we collected data from Korean National Health and Nutrition Examination Survey and built a decision tree after data pre-processing. The decision tree brings about five useful rules and their average classification accuracy is quite high (75.8%). In addition, the decision tree showed that high blood pressure and waist circumference are the most influential factors on the classification of the two groups. Our research results will serve as good guidelines for clinicians to provide better treatment for patients with MS, such that they do not develop diabetes.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.517-525
• /
• 2022

In the era of big data, various data mining techniques have been proposed as major analysis methodologies. As complex and diverse data is mass-produced, data mining techniques have attracted attention as a method that forms the foundation of data science. In this paper, we focused on the decision tree, which is frequently used in practice and easy to understand as one of representative data mining methods. Specifically, we analyzed the effect of the splitting method of decision trees on the model performance. We compared the prediction power and structures of decision tree models with different split methods based on various simulated data. The results show that the linear combination split method can improve the prediction accuracy of decision trees in the case of data simulated from nonlinear models with complex structure.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.227-228
• /
• 2023

정부의 인공지능 국가전략을 통해 인공지능 교육은 초등학교에서도 필수교육으로 대두되고 있다. 또한 인공지능 소양을 습득하기 위해 타 교과와 융합한 인공지능 융합 교육의 필요성이 증가하고 있고, 인공지능 발달에 대한 수학의 역할을 고려하여 수학 교과를 통해 인공지능의 이해를 기르는 것이 강조되고 있다. 따라서 본 연구에서는 수학 교과와 인공지능 교과가 융합한 인공지능 융합 교육 프로그램을 개발하기 위해 초등학교 3~4학년 수학 교과의 도형 분류를 의사 결정 나무 모델을 활용하여 가르치는 인공지능 융합 교육 프로그램을 개발하였다. 본 연구를 통해 개발된 프로그램은 초등학생의 인공지능 개념학습을 통한 인공지능 기초소양 함양뿐만 아니라 수학 교과의 이해 및 성취도 향상에 도움이 될 것으로 기대된다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.239-242
• /
• 2007

최근 데이터의 양이 많아지고 다양해짐에 따라서 데이터를 활용하기 위한 데이터 마이닝에 관한 관심이 중대되고 있다. 데이터 분석을 위한 수집 데이터에는 수집 과정에서 분석가가 원치 않은 데이터 잡음이 발생하는 경우가 있고 그 데이터가 다른 데이터들과 같은 가중치로 데이터 마이닝에 반영되는 경우 예상과 다른 결과를 얻을 수 있다. 따라서 데이터 분석 시 데이터와 전문가 의견이 고려된 데이터 엔트로피(Entropy)를 사용하여 잡음 데이터를 다를 필요가 있다. 본 논문에서는 전문가의견을 이용한 전문가 의견 목록을 만들고 이를 데이터와 비교하여 유사한 정도에 따라 각 데이터에 가중치를 부여한다. 그리고 이 데이터를 활용한 의사결정나무(Decision Tree)를 사용하여 기존 데이터를 이용한 의사결정나무 보다 데이터 잡음의 영향을 줄이는 방법을 제안한다. 제안한 방법은 학습자의 학습 활동에서 수집된 학습 행위 데이터를 사용하여 실험하였다.

• Journal of the Korea Convergence Society
• /
• v.9 no.5
• /
• pp.179-189
• /
• 2018

Currently, research on smart factories is steadily being carried out in terms of implementation strategies and considerations in construction. Various studies have not been conducted on companies that introduced smart factories. This study conducted a questionnaire survey for SMEs applying the basic stage of smart factory. And the cluster analysis was conducted to examine the characteristics of the company. In addition, we conducted Decision Tree and Naive Bay to examine how the characteristics of a company are derived and compare the results. As a result of the cluster analysis, it was confirmed that the group was divided into the high satisfaction group and the low satisfaction group. The decision tree and the Naive Bay analysis showed that the higher satisfaction group has high productivity.

• Korean Journal of Construction Engineering and Management
• /
• v.22 no.6
• /
• pp.76-86
• /
• 2021

Construction projects have high potentials of claims and disputes due to inherent risks where a variety of stakeholders are involved. Since disputes could cause losses in terms of cost and time, it is a critical issue for contractors to forecast and pro-actively manage disputes in advance in order to secure project efficiency and higher profits. The objective of the study is to develop a decision tree-based predictive analytics model for forecasting dispute types and their probabilities according to construction project conditions. It can be a useful tool to forecast potential disputes and thus provide opportunities for proactive management.

• International Journal of Highway Engineering
• /
• v.13 no.3
• /
• pp.167-176
• /
• 2011

This study's purpose is to analyse factors of determination about detouring for makinga standard model in regard of unfavorableness and uncertainty when unspecified individual recipients make a decision at the time of course detour. In order to achieve this, we surveyed SP investigation whether making a detour or not for drivers as a target who take a high way and National highway. Based on this result, we analysed detour determination factors of drivers, establishing a combination model of Decision Tree and Neural Network model. The result demonstrates the effected factors on drivers' detour determination are in ordering of the recognition of alternative routevs, reliable and frequency of using traffic information, frequency of transition routes and age. Moreover, from the outcome in comparison with an existing model and prediction through undistributed data, the rate of combination model 8.7% illustrates the most predictable way in contrast with logit model 12.8%, and Individual Model of Decision Tree 13.8% which are existed. This reveals that the analysis of drivers' detour determination factors is valid to apply. Hence, overall study considers as a practical foundation to make effective detour strategies for increasing the utility of route networking and dispersion in the volume of traffic from now on.