• Information Systems Review
• /
• v.14 no.3
• /
• pp.115-129
• /
• 2012

A matchmaking system is a type of recommender systems that provides a set of dating partners suitable for the user by online. Many matchmaking systems, which are widely used these days, require users to specify their preferences with regards to ideal dating partners based on criteria such as age, job and salary. However, some users are not aware of their exact preferences, or are reluctant to reveal this information even if they do know. Also, users' selection standards are not fixed and can change according to circumstances. This paper suggests a new matchmaking system called Decision Tree based Matchmaking System (DTMS) that automatically adjusts the stated standards of a user by analyzing the characteristics of the people the user chose to contact. AMMS provides recommendations for new users on the basis of their explicit preferences. However, as a user's behavioral records are accumulated, it begins to analyze their hidden implicit preferences using a decision tree technique. Subsequently, DTMS reflects these implicit preferences in proportion to their predictive accuracy. The DTMS is regularly updated when a user's data size increases by a set amount. This paper suggests an architecture for the DTMS and presents the results of the implementation of a prototype.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.11
• /
• pp.245-253
• /
• 2011

Due to the imprudent spending of the fossil fuels, the environment was contaminated seriously and the exhaustion problems of the fossil fuels loomed large. Therefore people become taking a great interest in alternative energy resources which can solve problems of fossil fuels. The wind power energy is one of the most interested energy in the new and renewable energy. However, the plants of wind power energy and the traditional power plants should be balanced between the power generation and the power consumption. Therefore, we need analysis and prediction to generate power efficiently using wind energy. In this paper, we have performed a research to predict power generation patterns using the wind power data. Prediction approaches of datamining area can be used for building a prediction model. The research steps are as follows: 1) we performed preprocessing to handle the missing values and anomalous data. And we extracted the characteristic vector data. 2) The representative patterns were found by the MIA(Mean Index Adequacy) measure and the SOM(Self-Organizing Feature Map) clustering approach using the normalized dataset. We assigned the class labels to each data. 3) We built a new predicting model about the wind power generation with classification approach. In this experiment, we built a forecasting model to predict wind power generation patterns using the decision tree.

• Spatial Information Research
• /
• v.20 no.6
• /
• pp.77-90
• /
• 2012

In this paper, we examined how to utilize QR codes for meeting the information demand and making better use of the paper map. By Decision Tree Analysis, we investigated whether to have any intention to use the paper map with QR codes for receiving more information and what decision variables affect the answers. Thus, we also surveyed the area of providing information and sectoral demand for deriving additional information demand to being provided through QR codes. In the results of our study, we confirmed that the decision variables, to make any intention to use the paper map with QR code, are the frequency of using the paper, the experience of using the paper map, the intention to buy the paper map, the experience of using QR codes and the experience of buying the paper map. In these variables, the frequency of using the paper map is a major factor to decide whether it is intended to use the paper map with QR codes. we also identified that there are various additional information demand using the paper map with QR codes in the area of 'Daily life', 'Real estate', 'Education', 'Travel and Leisure', and 'Entertainment'. Especially additional information demand is high in the area of 'Travel and Leisure'. These results could be used to find a way how to vitalize the usage of paper map by introduction of QR codes and how to develop QR codes for the paper map and concerning applications.

• Journal of Digital Convergence
• /
• v.11 no.10
• /
• pp.593-608
• /
• 2013

According to increasing number of injury claims, the challenge is reducing investigation of cases of injuries by selecting them more delicately, while also increasing the redemption rates and the amount of restitution. In this regards, we developed the fraud detection model for injury claims of self-employed insured by using decision tree after collecting medical claim data from 2006 to 2011 of the National Health Insurance in Korea. As a result of this model, subject types were classified into 18 types. If applying these types to the actual survey compared with if not applying, the redumption collecting rate will be increasing by 12.8%. Also, the effectiveness of this model will be maximize when the number of claims handlers considering their survey volume and management plans are examined thoroughly.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.571-573
• /
• 2022

최근 금융기관에서는 축적된 금융 빅데이터를 활용하여 차별화된 서비스를 강화하고 있다. 기업고객에 투자하기 위해서는 보다 정밀한 기업분석이 필요하다. 본 연구는 대만기업 6,819개의 95개 재무데이터를 가지고, 비대칭 데이터 문제해결, 데이터 표준화 등 데이터 전처리 작업을 하였다. 해당 데이터는 로지스틱 회기, SVM, K-NN, 나이브 베이즈, 의사결정나무, 랜덤포레스트 등 9가지 분류모델에 5겹 교차검증을 적용하여 학습한 후 모델 성능을 비교하였다. 이 중에서 성능이 가장 우수한 분류모델을 선택하여 예측 결정 이유를 판단하고자 설명 가능한 인공지능(XAI)을 적용하여 예측 결과에 대한 설명을 부여하여 이를 분석하였다. 본 연구를 통해 데이터 전처리에서부터 모델 예측 결과 설명에 이르는 분류예측모델의 전주기를 자동화하는 시스템을 제시하고자 한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.7
• /
• pp.3368-3373
• /
• 2013

The purpose of this study was to analyze the risk factors of self-reported voice problem. Data were from the Korea National Health and Nutritional Examination Survey 2008. Subjects were 3,600 persons (1,501 men, 2,099 women) aged 19 years and older. A prediction model was developed by the use of a exhaustive CHAID (Chi Squared Automatic Interaction Detection) algorism of decision tree model. In the decision tree analysis, pain and discomfort during the last 2 weeks, age, the longest occupation and thyroid disorders was significantly associated with self-reported voice problem. The findings of associated factors suggest potential ways of targeting counseling and prevention efforts to control self-reported voice problem.

• Journal of the military operations research society of Korea
• /
• v.36 no.3
• /
• pp.83-93
• /
• 2010

In this paper, we propose a new expert system for close combat in military war game model for training. Simulation logic for damage assesment is one of the main simulation functions in military war game. In Changcho 21's model which is the war game model for Republic of Korea Army corps and division, the main function of close combat's damage assessment has not been calculated by Changcho 21's model, but by COBRA which was made by US Army and has been the expert system for close combat. Results which were calculated in COBRA were sent to Changcho 21's model through a cable network. And Changcho 21's model finally calculated the value of damage assessment with the results. In this paper, we develop an new expert system for close combat using decision tree. The experimental results show that the proposed expert system has similar performance to COBRA and has less computing complexity. And it can substitute for COBRA and be applicable to battlefield.

• Communications for Statistical Applications and Methods
• /
• v.18 no.1
• /
• pp.23-34
• /
• 2011

Property auctions have become a new method for real estate investment because the property auction market grows in tandem with the growth of the real estate market. This study focused on the statistical model for predicting bid price rates which is the main index for participants in the real estate auction market. For estimating the monthly bid price rate, we proposed a new method to make up for the mean of regions and terms as well as to reduce the prediction error using a decision tree analysis. We also proposed a linear regression model to predict a bid price rate for individual auction property. We applied the proposed model to apartment auction property and tried to predict the bid price rate as well as categorize individual auction property into an auction grade.

• Journal of Korean Biological Nursing Science
• /
• v.20 no.2
• /
• pp.103-113
• /
• 2018

Purpose: This study was performed to assess problems associated with sleep (short and long sleep duration) and to identify risky subgroups with sleep problems among adult cancer survivors. The study is based on the Korea National Health and Nutrition Examination Survey (KNHANES VI and VII) from 2013 to 2016. Methods: The sociodemographic and clinical data of 504 Korean cancer survivors aged 20-64 years was extracted from the KNHANES VI and VII database. Descriptive statistics for complex samples was used, and decision-tree analyses were performed using the SPSS WIN 24.0 program. Results: The mean age for survivors was approximately 51 years. The mean sleep duration was 6.97 hours; 36.2% of participants had short (< 7 hours) and 9.9% had long (> 8 hours) sleep duration. From the decision-trees analyses, the characteristics of the adult cancer survivors related to sleep problems were presented with six different pathways. Sleep problems were analyzed according to the survivors' sociodemographic information (age, education, living status, and occupation), clinical characteristics (body mass index, hypercholesterolemia, and anemia) and health-related quality of life (HRQoL). The HRQoL (${\leq}0.5$ or > 0.5 cutoff point) was a significant predictor of the participants' sleep problems because all six pathways were started from this predictor in the model. Conclusion: Health care professionals could use the decision-tree model for screening adult cancer survivors with sleep problems in clinical or community settings. Nursing interventions considering these specific individual characteristics and HRQoL level should be developed to have adequate sleep duration for Korean adult cancer survivors.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.6
• /
• pp.1417-1426
• /
• 2015

We develop the high risk group predictive model for loan default by utilizing the direct loan data from 2012 to 2014 of the Korea Student Aid Foundation. We perform the decision tree analysis using the data mining methodology and use SAS Enterprise Miner 13.2. As a result of this model, subject types were classified into 25 types. This study shows that the major influencing factors for the loan default are household income, national grant, age, overdue record, level of schooling, field of study, monthly repayment. The high risk group predictive model in this study will be the basis for segmented management service for preventing loan default.