• Journal of the Korea Safety Management & Science
• /
• 제8권1호
• /
• pp.53-64
• /
• 2006

Many researches and analyses have been focused on industrial accidents in order to predict and reduce them. As a similar endeavor, this paper is to develop an expert system for prevention of industrial accidents. Although various previous studies have been performed to prevent industrial accidents, these studies only provide managerial and educational policies using frequency analysis and comparative analysis based on data from past industrial accidents. As an initial step for the purpose of this study, this paper provides a comparative analysis of 4 kinds of algorithms including CHAID, CART, C4.5, and QUEST. Decision tree algorithm is utilized to predict results using objective and quantified data as a typical technique of data mining. Enterprise Miner of SAS and Answer Tree of SPSS will be used to evaluate the validity of the results of the four algorithms. The sample for this work was chosen from 10,536 data related to manufacturing industries during three years$(2002\sim2004)$ in korea. The initial sample includes a range of different businesses including the construction and manufacturing industries, which are typically vulnerable to industrial accidents.

• Journal of the Korea Safety Management & Science
• /
• 제16권4호
• /
• pp.275-284
• /
• 2014

Decision Tree is one of analysis techniques which conducts grouping and prediction into several sub-groups from interested groups. Researcher can easily understand this progress and explain than other techniques. Because Decision Tree is easy technique to see results. This paper uses CART algorithm which is one of data mining technique. It used 273 variables and 70094 data(2010-2011) of working environment survey conducted by Korea Occupational Safety and Health Agency(KOSHA). And then refines this data, uses final 12 variables and 35447 data. To find satisfaction factor in working environment, this page has grouped employee to 3 types (under 30 age, 30 ~ 49age, over 50 age) and analyzed factor. Using CART algorithm, finds the best grouping variables in 155 data. It appeared that 'comfortable in organization' and 'proper reward' is the best grouping factor.

• Journal of Digital Contents Society
• /
• 제5권3호
• /
• pp.205-212
• /
• 2004

This paper presents a design and implementation example of intelligent society member management system that is constructed to induce various research activity. Based on members data and society activity record, the system executed data mining. In the process of data mining useful society activity rules was produced and in result members could effectively interact with the system. Decision Tree Algorithm was used in the process, which is one of the methods of data mining. We presemts a plan for personalization website to provide user oriented administration policy and dynamic interface by using analyzed information of society activity rules produced.

• Journal of Korea Society of Digital Industry and Information Management
• /
• 제11권4호
• /
• pp.33-45
• /
• 2015

Currently, Internet is used an essential tool in the business area. Despite this importance, there is a risk of network attacks attempting collection of fraudulence, private information, and cyber terrorism. Firewalls and IDS(Intrusion Detection System) are tools against those attacks. IDS is used to determine whether a network data is a network attack. IDS analyzes the network data using various techniques including expert system, data mining, and state transition analysis. This paper tries to compare the performance of two data mining models in detecting network attacks. They are decision tree (C4.5), and neural network (FANN model). I trained and tested these models with data and measured the effectiveness in terms of detection accuracy, detection rate, and false alarm rate. This paper tries to find out which model is effective in intrusion detection. In the analysis, I used KDD Cup 99 data which is a benchmark data in intrusion detection research. I used an open source Weka software for C4.5 model, and C++ code available for FANN model.

• Journal of Korea Society of Digital Industry and Information Management
• /
• 제17권3호
• /
• pp.1-7
• /
• 2021

The number of new students at many domestic universities is declining. In particular, private universities, which are highly dependent on tuition, are experiencing a crisis of existence. Amid the declining school-age population, universities are striving to fill new students by improving the quality of education and increasing the student employment rate. Recently, there is an increasing number of cases of using the accumulated big data of universities to prepare measures to fill new students. A representative example of this is the analysis of factors that affect student employment. Existing employment-influencing factor analysis studies have applied quantitative models such as regression analysis to university big data. However, since the factors affecting employment differ by major, it is necessary to reflect this. In this paper, the factors affecting employment by major are analyzed using the data of University C and the decision tree model. In addition, based on the analysis results, a roadmap for student employment by major is recommended. As a result of the experiment, four decision tree models were constructed for each major, and factors affecting employment by major and roadmap were derived.

• Journal of Forest and Environmental Science
• /
• 제38권4호
• /
• pp.275-283
• /
• 2022

A study was designed to investigate the pattern of sexual dimorphism in a plantation of Garcinia kola. Twenty trees were randomly selected for the study and have been observed to flower regularly. A total of 100 inflorescence were randomly collected from the crown of each tree and 500 flowers randomly assessed within the period of four (4) flowering seasons. Floral sex assessment was done visually and with a hand magnifying lens; floral morphometric measurements (i.e. pedicel and perianth length and breadth), inflorescence length, and breadth) was taken using a veneer caliper; number of flowers per inflorescence and inflorescence per twig was counted; while, data analysis was conducted on excel using analysis of variance and pairwise t-test comparison. Four floral sexes were identified in the G. kola plantation studied which were unisexual male flowers, unisexual female flowers, cosexual unisexual male flowers, and cosexual hermaphrodite flowers. Three tree sexes were identified viz: inconstant male, invariant female, and cosexual trees. The plantation was significantly sexually dimorphic in floral sex and phenotypic traits (i.e. pedicel and perianth size), and as well as sexually dimorphic in tree sex and reproductive phenotypic traits (i.e. inflorescence size, number of inflorescences per twig, and number of flower bud per inflorescence). The sexual system of the plantation was therefore trioecious with features suggestive of evolving dioecy through the gynodioecious pathway.

• Journal of Korea Society of Digital Industry and Information Management
• /
• 제19권1호
• /
• pp.109-118
• /
• 2023

Data-based analysis methods have become used more for estimating or predicting housing prices, and neural network models and decision trees in the field of big data are also widely used more and more. Neural network models are often evaluated to be superior to existing statistical models in terms of estimation or prediction accuracy. However, there is ambiguity in determining the input feature of the input layer of the neural network model, that is, the type and number of input features, and decision trees are sometimes used to overcome these disadvantages. In this paper, we evaluate the existing methods of using decision trees and propose the method of using decision trees to prioritize input feature selection in neural network models. This can be a complementary or combined analysis method of the neural network model and decision tree, and the validity was confirmed by applying the proposed method to house price estimation. Through several comparisons, it has been summarized that the selection of appropriate input characteristics according to priority can increase the estimation power of the model.

• Journal of the Korea Institute of Military Science and Technology
• /
• 제27권1호
• /
• pp.80-93
• /
• 2024

This study attempted to analyze the factors that influence the participation of beneficiary companies in the government's defense industry promotion support project. To this end, experimental data were analyzed by constructing a prediction model consisting of highly important variables in beneficiary company decisions among various company information using the decision tree model, one of the data mining techniques. In addition, various rules were derived to determine the beneficiary companies of the government's support project using the analysis results expressed as decision trees. Three policy measures were presented based on the important rules that repeatedly appear in different predictive models to increase the effect of the government's industrial development. Using the analysis methods presented in this study and the determinants of the beneficiary companies of the government support project will help create a sustainable future defense industry growth environment.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 한국건설관리학회 2003년도 학술대회지
• /
• pp.485-490
• /
• 2003

Nowadays the rapid change in construction environment getting more globalized and complicated has caused lots of unexpected risks from inside and out of the country, so more sophisticated construction management strategies are being strongly needed. This paper suggests a risk management model with which we can estimate the appropriate contingency by quantifying the amount of probable risks immanent in large construction projects, which have a high degree of uncertainty in the anticipation of the total construction cost. To develop the model, the risk factors that make cost variations are elicited based on the real data of the contingencies assigned to the past projects. Furthermore, the influential relationship of risk factors is structured by applying the CRM(Cost Risk Model) which is the synthetic model of Monte Carlo Simulation, Influence Diagram and Decision Tree. The ultimate outcome of this research can by validated by tile case study with a large construction project performed.

• Journal of the Korean Operations Research and Management Science Society
• /
• 제40권1호
• /
• pp.91-116
• /
• 2015

This study aims to provide public sectors with eco-efficiency information. To implement the purposes of the study, environmental and economic variables of Eco-Efficiency were identified through decision tree model, then the relative Eco-Efficiencies of 243 public sectors were evaluated through input-oriented DEA (Data Envelopment Analysis) model. Specifically, the amount of public purchasing per a staff and the amount of energy use per a staff were considered as input factors. Sales per a staff was considered as output factor. The result shows that most of the public sectors (94.2%) were evaluated as "inefficient" taking into consideration of average value, 0.501 from market-based public corporations, 0.288 from local public corporations, 0.28 from quasi-market-based public corporations, 0.269 from fund-management-based quasi-governmental institutions, 0.09 from non-classified public institutions, and 0.078 from commissioned-service-based quasi-governmental institutions. Furthermore, it is possible to establish a plan for internal Eco-Efficiency improvement based on information of the reference set. In order to improve the Eco-Efficiency in the public sectors in the long term, environmental impacts of the overall public sectors' operations (e.g., energy saving, water saving, waste reduction, and purchasing of green products) needs to be properly proposed in consideration of BSC (Balanced Scorecard) indicators of public sectors.