• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.121-139
• /
• 2014

Predicting corporate failure has been an important topic in accounting and finance. The costs associated with bankruptcy are high, so the accuracy of bankruptcy prediction is greatly important for financial institutions. Lots of researchers have dealt with the topic associated with bankruptcy prediction in the past three decades. The current research attempts to use ensemble models for improving the performance of bankruptcy prediction. Ensemble classification is to combine individually trained classifiers in order to gain more accurate prediction than individual models. Ensemble techniques are shown to be very useful for improving the generalization ability of the classifier. Bagging is the most commonly used methods for constructing ensemble classifiers. In bagging, the different training data subsets are randomly drawn with replacement from the original training dataset. Base classifiers are trained on the different bootstrap samples. Instance selection is to select critical instances while deleting and removing irrelevant and harmful instances from the original set. Instance selection and bagging are quite well known in data mining. However, few studies have dealt with the integration of instance selection and bagging. This study proposes an improved bagging ensemble based on instance selection using genetic algorithms (GA) for improving the performance of SVM. GA is an efficient optimization procedure based on the theory of natural selection and evolution. GA uses the idea of survival of the fittest by progressively accepting better solutions to the problems. GA searches by maintaining a population of solutions from which better solutions are created rather than making incremental changes to a single solution to the problem. The initial solution population is generated randomly and evolves into the next generation by genetic operators such as selection, crossover and mutation. The solutions coded by strings are evaluated by the fitness function. The proposed model consists of two phases: GA based Instance Selection and Instance based Bagging. In the first phase, GA is used to select optimal instance subset that is used as input data of bagging model. In this study, the chromosome is encoded as a form of binary string for the instance subset. In this phase, the population size was set to 100 while maximum number of generations was set to 150. We set the crossover rate and mutation rate to 0.7 and 0.1 respectively. We used the prediction accuracy of model as the fitness function of GA. SVM model is trained on training data set using the selected instance subset. The prediction accuracy of SVM model over test data set is used as fitness value in order to avoid overfitting. In the second phase, we used the optimal instance subset selected in the first phase as input data of bagging model. We used SVM model as base classifier for bagging ensemble. The majority voting scheme was used as a combining method in this study. This study applies the proposed model to the bankruptcy prediction problem using a real data set from Korean companies. The research data used in this study contains 1832 externally non-audited firms which filed for bankruptcy (916 cases) and non-bankruptcy (916 cases). Financial ratios categorized as stability, profitability, growth, activity and cash flow were investigated through literature review and basic statistical methods and we selected 8 financial ratios as the final input variables. We separated the whole data into three subsets as training, test and validation data set. In this study, we compared the proposed model with several comparative models including the simple individual SVM model, the simple bagging model and the instance selection based SVM model. The McNemar tests were used to examine whether the proposed model significantly outperforms the other models. The experimental results show that the proposed model outperforms the other models.

• Journal of radiological science and technology
• /
• v.30 no.2
• /
• pp.167-175
• /
• 2007

There are some positive effects by the introduction of PACS(Picture Archiving Communication System). This study is to analyze the mutual relation between before and after of the introduction of PACS in terms of the environment effect. It is supposed to cause the reduction of developing and fixing wastewater according to the increase in the rate of a non-film. This study will also show the amount of wastewater. Target places were the department of image medicine(diagnostic radiation) of the general hospitals in Seoul and Gyeonggi-Do, which are equiped with full PACS. The authors examined questionnaires on the number of projection, the number of indirect projection, the amount of the film used, the number of radiation image CD loan, the amount of the developing and fixing solution used, the change of the amount of fixing wastewater. According to the analysis, we analyzed the change of the amount of developing and fixing solution per a film and the change of the amount of developing and fixing wastewater which is supposed to be reduced proportionally by the introduction of PACS. We got conclusion as below after analyzing 8 hospitals except the largest and the least amount of examination, film used, developing and fixing solution and the amount of developing and fixing wastewater in order to decrease the deviation from 10 general hospitals located in Seoul and Gyeonggi-Do. We compared data one year before adopting PACS Versus 3 years after adopting PACS. 1. The frequences of examination increased to 7,357.7 cases per month but the amount of film used decreased to 90%, from 42,774.4 to 4,181.88 after adopting the PACS. 2. 3 years after adopting PACS, monthly average amount of developing solution used decreased to 92% and the monthly average amount of fixing solution decreased to 86%. 3. Monthly average amount of developing solution used per film increased to 1.49 times and fixing solution increased as much as three times. 4. Monthly average wastewater for developing decreased to 88% and wastewater for fixing decreased up to 87%. 5. Monthly average wastewater for developing per film increased to 3.77 times and wastewater for fixing increased to 3.85 times. Although the amount of film used and the amount of developing and fixing wastewater affected by the reduction of the developing and fixing solution became less on the whole by introduction of PACS, they did not decrease proportionally. Moreover the amount of the developing and fixing solution used and the amount of developing and fixing wastewater per a film increased. That means the expectation for an environmental improvement differs from the actual condition.

• Journal of the Korean Applied Science and Technology
• /
• v.35 no.1
• /
• pp.235-246
• /
• 2018

Paeony has pharmacological activities such as anti-inflammatory, anti-allergic, anti-bacterial, central inhibitory, gastric secretion inhibition, and antispasmodic activities. In addition, its antioxidant activity and whitening effect being reported, thus it is being explored as raw materials for cosmetics. We compared the changes in the contents of paeoniflorin and paeonol in Peony extracts, depending on the changes of extracting solvents, temperature and time. The HPLC method was set up for simultaneous analysis, the system suitabilities were confirmed by using the calibration curves and the QC samples for each assay batch. Paeonol was detected only in roots, and paeoniflorin was higher in leaf and flower than root. Higher concentrations of both ingredients were extracted when the root was used after grinding to a suitable size, and when 30% 1,3-butylene glycol was used as the extraction solvent. Also the concentrations tended to increase at higher temperature and longer time, but the increase was gradual at over $75^{\circ}C$ and 4 hours. The ratio of root, leaf and flower was determined to be 2+2+1g/0.5kg of batch, reaching the contents criteria of paeoniflorin and paeonol. Finally, we selected as the best extraction condition when the raw materials are mixed with 2+2+1g/0.5kg and extracted with 30% 1,3-butylene glycol as an extraction solvent at $75^{\circ}C$ for 4 hours, considering both the concentrations of two components and the cost of raw materials and manufacturing process, The extraction units were scaled up to 10 kg under this condition.

• The Journal of the Korean life insurance medical association
• /
• v.22
• /
• pp.139-169
• /
• 2003

국내 보험 산업은 경제발전에 따라 많은 성장을 이루어왔으나 단체보험은 보험 회사의 외부적 또는 내부적 환경으로 인해 활성화가 미흡한 상황이다. 개인보험시장의 포화, 해외시장에서 단체보험의 지속적인 성장, 사회보험 민영화 논의, 방카슈랑스와 보험시장의 개방 등 급격한 변화를 겪고 있는 보험영업환경을 고려한다면 향후 확대될 기업복지시장에서의 성공적 역할수행을 위해서는 단체보험 영업과 지원시스템의 보완이 시급한 과제라고 할 수 있다. 특히 언더라이팅 기법은 단체보험의 핵심역량이며 회사의 경쟁력 강화와 사차익 안정화를 위해서는 단체보험 인수기법의 선진화가 필수요건이다. 단체보험은 대수의 법칙, 수지상등의 원칙 등 보험의 기본원리라는 측면에서는 개인보험과 다른 점이 없지만 하나의 계약을 통해 집단의 피보험자에게 보장을 제공하기 때문에 가지게 되는 역선택 축소, 비용절감, 기업에 의한 1차선택 등 몇 가지 특성들은 인수기법에서의 차이를 필요로 한다. 하지만 국내의 단체보험 언더라이팅은 기본개념조차 제대로 정립되어 있지 못한 초기단계로 단체보험의 특성을 제대로 반영할 수 있는 위험평가를 위해서는 선진기법의 도입이 절실하다. 첫째, 자유보장한도(FREE COVER LIMIT)의 도입이다. 자유보장한도는 단체에 대한 위험과 피보험자 개인의 위험을 구분하는 기준으로 자유보장한도내의 피보험자에 대해서는 고지나 의적 검사를 요구하지 않고 개인별 언더라이팅을 하지 않으며 거절체나 표준하체이더라도 자유보장한도 금액까지는 나머지 정상 피보험자들과 동일한 보장을 제공하는 것을 말한다. 이는 피보험자별 위험을 중심으로 심사하고 있는 현재의 국내 단체보험 인수방법에서 발생되는 고객측 불만과 심사의 비효율성을 개선할 수 있는 방안이다. 둘째, 단체별 특성에 따른 보험요율 차등화이다 단체는 산업의 종류, 피보험자의 직무, 지역적 위치, 크기(피보험자수), 성별구성비 등 여러 가지 특성들을 가지고 있으며 이런 특성들은 보장급부에 따라 발생위험에 영향을 미칠 수 있다. 하지만 현재 나이와 성별에 따라서 피보험자별로 정해지는 요율체계만으로는 이런 위험들을 적절하게 반영하지 못하고 단체별 형평성에도 문제가 생긴다. 따라서 정확한 data 구축을 통해 단체 특성별로 어떤 보장에 어느 정도로 위험 발생에 영향을 미치는지를 분석하여 보험요율을 차등부과 할 수 있는 체계를 갖추어야 한다. 셋째, 경험을 활용한 보험료 산정기법이다. 이것은 개인보험과 구분되는 가장 큰 단체보험만의 특성이라 할 수 있는데 해당 단체의 과거 경험 data 즉 청구로 인한 지급금액을 토대로 당해 계약 보험료를 결정하는 방법이다. 이를 위해서는 과거 경험 data가 얼마나 신뢰할 수 있는 지의 정도(신뢰도)를 측정하는 것이 필요하다. 과학적이고 효율적인 단체보험 인수를 위해서 경험율의 활용은 반드시 이루어져야 할 선결과제이다. 넷째, 관련규정의 정비가 필요하다 단체보험의 특성을 갖기 위해서는 가입가능한 피보험자들의 자격규정(eligibility), 활동적근무 조건(actively at work)이 요구되어야 하며 참여비율(가입비율)과 보장수준의 구성방법에 대해서도 가이드라인 설정을 통해 역선택을 최소화 할 수 있는 제도적 보완이 마련되어야 한다. 이런 선진인수기법의 도입과 함께 단체보험 언더라이터의 필요역량을 개발하고 향상시키는 노력도 병행되어야만 종합금융화, 대형화가 진전되면서 대경쟁의 구도로 바뀌고 있는 국내 보험시장에서의 경쟁력 확보가 가능할 것이다.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.131-142
• /
• 2011

In recent years, Social Network Service, which is defined as a web-based service that allows an individual to construct a public or a semi-public profile within a bounded system, articulates a list of other users with whom they share connections, and traverses their list of connections. For example, Facebook and Twitter are the representative sites of Social Network Service, and these sites are the big issue in the world. A lot of people use Social Network Services to connect and maintain social relationship. Recently the users of Social Network Services have increased dramatically. Accordingly, many organizations become interested in Social Network Services as means of marketing, media, communication with their customers, and so on, because social network services can offer a variety of benefits to organizations such as companies and associations. In other words, organizations can use Social Network Services to respond rapidly to various user's behaviors because Social Network Services can make it possible to communicate between the users more easily and faster. And marketing cost of the Social Network Service is lower than that of existing tools such as broadcasts, news papers, and direct mails. In addition, Social network Services are growing in market place. So, the organizations such as companies and associations can acquire potential customers for the future. However, organizations uniformly communicate with users through Social Network Service without consideration of the characteristics of the networks although networks have different effects on information deliveries. For example, members' cohesion in an offline communication is higher than that in an online communication because the members of the offline communication are very close. that is, the network of the offline communication has a strong tie. Accordingly, information delivery is fast in the network of the offline communication. In this study, we compose two networks which have different characteristic of communication in Twitter. First network is constructed with data based on an offline communication such as friend, family, senior and junior in school. Second network is constructed with randomly selected data from users who want to associate with friends in online. Each network size is 250 people who divide with three groups. The first group is an ego which means a person in the center of the network. The second group is the ego's followers. The last group is composed of the ego's follower's followers. We compare the networks through social network analysis and follower's reaction analysis. We investigate density and centrality to analyze the characteristic of each network. And we analyze the follower's reactions such as replies and retweets to find differences of information delivery in each network. Our experiment results indicate that density and centrality of the offline communicationbased network are higher than those of the online-based network. Also the number of replies are larger than that of retweets in the offline communication-based network. On the other hand, the number of retweets are larger than that of replies in the online based network. We identified that the effect of information delivery in the offline communication-based network was different from those in the online communication-based network through experiments. So, you configure the appropriate network types considering the characteristics of the network if you want to use social network as an effective marketing tool.

• Journal of Intelligence and Information Systems
• /
• v.17 no.3
• /
• pp.25-41
• /
• 2011

Recently, support vector machines (SVMs) are being recognized as competitive tools as compared with other data mining techniques for solving pattern recognition or classification decision problems. Furthermore, many researches, in particular, have proved them more powerful than traditional artificial neural networks (ANNs) (Amendolia et al., 2003; Huang et al., 2004, Huang et al., 2005; Tay and Cao, 2001; Min and Lee, 2005; Shin et al., 2005; Kim, 2003).The classification decision, such as a binary or multi-class decision problem, used by any classifier, i.e. data mining techniques is so cost-sensitive particularly in financial classification problems such as the credit ratings that if the credit ratings are misclassified, a terrible economic loss for investors or financial decision makers may happen. Therefore, it is necessary to convert the outputs of the classifier into wellcalibrated posterior probabilities-based multiclass credit ratings according to the bankruptcy probabilities. However, SVMs basically do not provide such probabilities. So it required to use any method to create the probabilities (Platt, 1999; Drish, 2001). This paper applied AdaBoost algorithm-based support vector machines (SVMs) into a bankruptcy prediction as a binary classification problem for the IT companies in Korea and then performed the multi-class credit ratings of the companies by making a normal distribution shape of posterior bankruptcy probabilities from the loss functions extracted from the SVMs. Our proposed approach also showed that their methods can minimize the misclassification problems by adjusting the credit grade interval ranges on condition that each credit grade for credit loan borrowers has its own credit risk, i.e. bankruptcy probability.

• Journal of Korean Society of Environmental Engineers
• /
• v.33 no.12
• /
• pp.913-921
• /
• 2011

This study examined the overall feasibility of low carbon green village formed in rural area. The check method is analyzing its environmental and economic feasibility and energy self-reliance. The biomass of the villages was set as 28 ton/day of livestock feces and 2 ton/day of cut fruit tree branches which make up the total of 30 ton/day. The facility consisted of a bio gasfication facility using wet (livestock feces) biomass and combined heat power generator, composting facility and wood boiler using dry (cut fruit tree branches) biomass. When operating the system, 540,540 kWh/yr of electricity and 1,762 Gcal/yr of heat energy was produced. The region's electricity energy and heat energy self-reliance rate will be 100%. The economic feasibility was found as a loss of 140 million won where the facility installation cost is 5.04 billion won, operation cost is 485.09 million won and profit is 337.12 million won. There will be a loss of about 2.2 billion won in 15 years but in the environmental analysis, it was found that crude replacement effect is about 178 million won, greenhouse gas reduction effect is about 92 million won making up the total environmental benefit of 270 million won. This means, there will be a yearly profit of about 130 million won. In terms of its environmental and economic feasibility and energy self-reliance, this project seemed to be a feasible project in overall even if it manages to get help from the government or local government.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.3
• /
• pp.176-188
• /
• 2018

Modern urban problems are increasingly becoming a market mix that can not be solved by the power of a single department and the necessity of establishing a cooperation system based on data communication between departments is increasing. Therefore, this study analyzed Busan metropolitan city's IT projects from 2014 to 2018 in order to understand the utilization and sharing status of departmental data from the viewpoint that cooperation between departments can start from the sharing of data with high common utilization. In addition, based on the results of the FGI(Focus Group Interview) conducted for the officials of the department responsible for the informatization project, we verified the results of data status analysis. At the same time, we figured out the necessity of data link between departments through SNA(Social Network Analysis) and presented data that should be shared first in the future. As a result, most of the information systems currently use limited data only within the department that produced the data. Most of the linked data was concentrated in the information department. Therefore, this study suggested the following solutions. First, in order to prevent overlapping investments caused by the operation of individual departments and share information, it is necessary to build a small platform to tie the departments, which have high connectivity with each other, into small blocks. Second, a local level process is needed to develop data standards as an extension of national standards in order to expand the information to be used in various fields. Third, as another solution, we proposed a system that can integrate various types of information based on address and location information through application of cloud-based GIS platform. The results of this study are expected to contribute to build a cooperation system between departments through expansion of information sharing with cost reduction.

• The Korean Journal of Air & Space Law and Policy
• /
• v.8
• /
• pp.231-248
• /
• 1996

이 논문(論文)은 한국(韓國)의 우주(宇宙) 산업(産業) 발달(發達)과 그 과정(過程)을 고찰(考察)하고 현재 진행 중인 통신위성(通信衛星)의 개발(開發)과 그 사업(事業)을 뒷받침해 주는 국가(國家)의 정책(政策)을 살펴보고 있다. KOREASAT라고 명명(命名)된 통신위성(通信衛星) 개발(開發)에 현재 여러 기관이 관계(關係)하고 있으나 그 중 KARI(한국항공우주연구소(韓國航空宇宙硏究所)), ETRI(전자통신연구소(電子通信硏究所)), SERI(시스템공학연구소(硏究所)), KAIST(한국과학기술연구소(韓國科學技術硏究所)) 등이 중추적인 연구(硏究)를 하고 있다. 특히 이 논문(論文)에서는 아래와 같은 문제(問題)를 다루고 있다 첫째, 최근(最近) 한국우주개발(韓國宇宙開發) 상황(狀況) 둘째, 장기(長期) 우주개발(宇宙開發) 계획(計劃) 셋째, 현재(現在) 우주개발(宇宙開發) 상황(狀況)과 미래(未來) 우주개발(宇宙開發)에 관련한 정책(政策)적 문제(問題) 넷째, 한국(韓國) 우주개발(宇宙開發)과 정책(政策) 방향(方向)에 대한 의견(意見) 최근의 한국우주개발(韓國宇宙開發) 상황(狀況)은 크게 한국(韓國)의 통신위성(通信衛星) 사업(事業)인 Koreasat program과 다목적 위성사업(衛星事業)인 KOMSAT로 나타나는데 한국(韓國)의 최초 상업용(商業用) 위성(衛星)인 Koreasat는 1호가 1995년도에 발사(發射)되었으나 정상궤도(正常軌道) 진입(進入)에 문제(問題)가 발생하여 발사업체(發射業體)로부터 보상문제(補償問題)가 제기(提起)되기도 하였으나 2호는 성공리에 발사(發射)되었다. 미국항공우주회사(美國航空宇宙會社)와 공동(共同)으로 개발(開發)중인 새로운 과학위성(科學衛星)인 KOMSAT는 한국우주과학기술(韓國宇宙科學技術)을 한단계 더 발전(發展)시킬 수 있을 것이고 1999년도에 발사(發射) 계획(計劃)이다. 한국항공우주연구소(韓國航空宇宙硏究所) 중심(中心)의 장기(長期) 우주개발(宇宙開發) 계획(計劃)에서 제시(提示)하는 4가지 우주개발(宇宙開發)의 기본목표(基本目標)는 첫째, 우주산업응용산업(宇宙産業應用産業)의 상호협조개발(相互協助開發)에 의한 우주산업(宇宙産業) 육성(育成) 둘째, 한국(韓國)의 현재 우주산업상황(宇宙産業狀況)에 알맞은 특정분야(特定分野)를 선정(選定)하고 이 분야(分野)에서 최단기에 세계(世界) 최고의 기술수준을 성취(成就)하도록 집중(集中) 셋째, 외국과 긴밀한 협조(協助)로 선진기술(先進技術)의 습득(習得) 넷째, 체계적이고 통합(統合)된 장기우주산업발전(長期宇宙産業發展) 계획(計劃) 성립(成立) 등이다. 이러한 계획(計劃)에 주요 사안(事案)으로는 2015년까지 19기의 인공위성(人工衛星) 보유(保有)를 위한 제작계획(製作計劃)과 2010년까지 발사체(發射體) 개발(開發)을 마련하는 것이다 현재 우주활동(宇宙活動)에 관련된 문제(問題)는 주로 Koreasat의 서비스와 사용계획(使用計劃)에 대한 것으로 위성개발사업(衛星開發事業)에 있어서 관련 정부부서간(政府部署間)에 의견(意見)을 달리하고 있는 형태이다. 한국통신(韓國通信)과 정보통신부(情報通信部)는 위성(衛星)의 DBS 트랜스폰더에 대해 디지털 방식(方式)을 적용(適用)할 것을 제안(提案)했지만 공보처(共報處)는 반대(反對)의 입장(立場)을 표명(表明)한 것과 방송국(放送局)의 관리(管理)와 통제(統制)는 공보처(共報處)에 있고 무선통신표준(無線通信標準)에 대한 면허(免許)는 정보통신부(情報通信部)에 있기 때문에 방송국(放送局)에 대한 면허(免許)는 각기 다른 두 단계(段階)로 구성(構成)되는 문제(問題)가 발생(發生)한다. 또한 DBS 서비스에서 사기업(私企業)의 참여(參與)와 관련하여 재벌(財閥)의 참여(參與)를 허용(許容)하느냐의 여부(與否)의 논쟁(論爭)이다. 다음으로 미래(未來) 우주산업개발(宇宙産業開發)에 관한 정책문제(政策問題)를 살펴보면 국가적(國家的) 차원(次元)에서 조직적(組織的)인 육성책(育成策)에 대한 문제(問題)로 현재 주관 부처가 과학기술처(科學技術處)와 통상산업부(通商産業部)로 나뉘어 추진(推進)되고 있다는 점이다. 그리고 차세대(次世代) 통신위성개발(通信衛星開發) 계획(計劃)에 대한 문제(問題)로 최소 2${\sim}$4개의 궤도확보(軌道確保)와 이미 정상궤도(正常軌道) 진입(進入) 실패(失敗)에 따른 Koreasat 1호의 생명단축(生命短縮)으로 새로운 통신위성(通信衛星)을 4년이내에 발사(發射)해야 한다는 문제(問題)이다. 결론적으로 장기(長期) 우주개발계획정책(宇宙開發計劃政策)에 있어서 첫째, 국제적 우주개발사업(宇宙開發事業)에 대한 적극적(積極的) 참여(參與), 둘째, 우주(宇宙)에서 독립적(獨立的)인 활동(活動)을 할 수 있는 국가안보체제(國家安保體制)의 개발(開發), 셋째, 국가(國家) 위상(位相)의 발전(發展)과 우주개발(宇宙開發)을 위한 인력활용(人力活用)의 개발(開發), 넷째, 무한한 우주(宇宙)에 도전(挑戰)할 수 있는 우수(優秀)한 인재(人才)의 교육(敎育), 다섯째, 21세기를 대비(對備)하여 한국(韓國)의 우주개발정책(宇宙開發政策)의 결정(決定) 등이 고려(考慮)되어야 할 5가지 요소(要所)들이다. 그리고 막대(莫大)한 비용(費用)이 드는 우주개발사업(宇宙開發事業)을 효율적으로 추진(推進)하기 위해서는 국가(國家) 최고(最高) 지도자(指導者)의 직접지휘(直接指揮)를 받는 정부기구(政府機構)가 수립(樹立)되어 정부차원(政府次元)에서 추진(推進)하되 산학연(産學硏)이 협조(協助)하여 우주개발계획(宇宙開發計劃)을 추진(推進)하여야 할 것으로 본다.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.243-264
• /
• 2018

Although the worldwide battery market is recently spurring the development of lithium secondary battery, lead acid batteries (rechargeable batteries) which have good-performance and can be reused are consumed in a wide range of industry fields. However, lead-acid batteries have a serious problem in that deterioration of a battery makes progress quickly in the presence of that degradation of only one cell among several cells which is packed in a battery begins. To overcome this problem, previous researches have attempted to identify the mechanism of deterioration of a battery in many ways. However, most of previous researches have used data obtained in a laboratory to analyze the mechanism of deterioration of a battery but not used data obtained in a real world. The usage of real data can increase the feasibility and the applicability of the findings of a research. Therefore, this study aims to develop a model which predicts the battery deterioration using data obtained in real world. To this end, we collected data which presents change of battery state by attaching sensors enabling to monitor the battery condition in real time to dozens of golf carts operated in the real golf field. As a result, total 16,883 samples were obtained. And then, we developed a model which predicts a precursor phenomenon representing deterioration of a battery by analyzing the data collected from the sensors using machine learning techniques. As initial independent variables, we used 1) inbound time of a cart, 2) outbound time of a cart, 3) duration(from outbound time to charge time), 4) charge amount, 5) used amount, 6) charge efficiency, 7) lowest temperature of battery cell 1 to 6, 8) lowest voltage of battery cell 1 to 6, 9) highest voltage of battery cell 1 to 6, 10) voltage of battery cell 1 to 6 at the beginning of operation, 11) voltage of battery cell 1 to 6 at the end of charge, 12) used amount of battery cell 1 to 6 during operation, 13) used amount of battery during operation(Max-Min), 14) duration of battery use, and 15) highest current during operation. Since the values of the independent variables, lowest temperature of battery cell 1 to 6, lowest voltage of battery cell 1 to 6, highest voltage of battery cell 1 to 6, voltage of battery cell 1 to 6 at the beginning of operation, voltage of battery cell 1 to 6 at the end of charge, and used amount of battery cell 1 to 6 during operation are similar to that of each battery cell, we conducted principal component analysis using verimax orthogonal rotation in order to mitigate the multiple collinearity problem. According to the results, we made new variables by averaging the values of independent variables clustered together, and used them as final independent variables instead of origin variables, thereby reducing the dimension. We used decision tree, logistic regression, Bayesian network as algorithms for building prediction models. And also, we built prediction models using the bagging of each of them, the boosting of each of them, and RandomForest. Experimental results show that the prediction model using the bagging of decision tree yields the best accuracy of 89.3923%. This study has some limitations in that the additional variables which affect the deterioration of battery such as weather (temperature, humidity) and driving habits, did not considered, therefore, we would like to consider the them in the future research. However, the battery deterioration prediction model proposed in the present study is expected to enable effective and efficient management of battery used in the real filed by dramatically and to reduce the cost caused by not detecting battery deterioration accordingly.