• 한국연소학회지
• /
• 제22권4호
• /
• pp.27-34
• /
• 2017

Various approaches have been proposed to predict the ash deposition (slagging and fouling) propensity of coal, which is essential in maintaining high efficiency and preventing corrosion/damage of a coal-fired boiler. The common method is to establish an index of the ash deposition propensity based on elementary coal composition and advanced characterization of ash properties, which is readily applicable to design, operation and maintenance of coal-fired boilers. Although many indexes have been developed for this purpose, their validity is still not satisfactory in actual applications to particular coal types or operating conditions. This paper reviews the status of predictive approaches for the ash deposition propensity, and assesses the performance of existing indexes by comparing the results for selected coals. This work will contribute to the development of a comprehensive and practical method for prediction of the ash deposition propensity.

• 산업경영시스템학회지
• /
• 제45권1호
• /
• pp.71-82
• /
• 2022

Most of the predictions using machine learning are neutral predictions considering the symmetrical situation where the predicted value is not smaller or larger than the actual value. However, in some situations, asymmetric prediction such as over-prediction or under-prediction may be better than neutral prediction, and it can induce better judgment by providing various predictions to decision makers. A method called Asymmetric Twin Support Vector Regression (ATSVR) using TSVR(Twin Support Vector Regression), which has a fast calculation time, was proposed by controlling the asymmetry of the upper and lower widths of the ε-tube and the asymmetry of the penalty with two parameters. In addition, by applying the existing GSVQR and the proposed ATSVR, prediction using the prediction propensities of over-prediction, under-prediction, and neutral prediction was performed. When two parameters were used for both GSVQR and ATSVR, it was possible to predict according to the prediction propensity, and ATSVR was found to be more than twice as fast in terms of calculation time. On the other hand, in terms of accuracy, there was no significant difference between ATSVR and GSVQR, but it was found that GSVQR reflected the prediction propensity better than ATSVR when checking the figures. The accuracy of under-prediction or over-prediction was lower than that of neutral prediction. It seems that using both parameters rather than using one of the two parameters (p_1,p_2) increases the change in the prediction tendency. However, depending on the situation, it may be better to use only one of the two parameters.

• Communications for Statistical Applications and Methods
• /
• 제16권1호
• /
• pp.103-113
• /
• 2009

표본조사에서 무응답은 비 표본추출오차를 발생시키는 중요한 원인 중 하나이다. 단위무응답이 발생하는 경우 무응답에 의한 편향을 줄이는 동시에 추정의 정도를 향상시키기 위해 단위무응답 조정층을 구성해 무응답 가중치 조정을 하는 것이 일반적이다. 본 연구에서는 무응답 조정층 구성과 관련된 기존의 이론들을 정리하고 어업총조사 자료를 이용한 실증적인 모의실험을 통해 효과적으로 무응답 조정층을 구성하는 방안에 대해 살펴본다. 모의실험결과 응답성향에 따른 조정층 구성보다는 예측평균을 기준으로 한 조정층 구성이 효율성 측면에서 효과적인 것으로 나타났으며, 아울러 다른 관심변수에도 적용될 수 있는 로버스트한 조정층 구성을 위해서는 예측평균만을 고려하는 것보다 응답성향과 예측평균을 모두 고려한 조정층 구성방법이 효과적인 것으로 나타났다. 한편 무응답 조정을 위한 응답률 산출에 있어서 설계가중치의 적용 필요성에 대해 살펴본 결과 설계가중치 적용 여부는 추정결과에 거의 영향을 주지 않는다는 사실을 확인할 수 있었다.

• 대한토목학회논문집
• /
• 제32권4D호
• /
• pp.313-322
• /
• 2012

본 연구는 택시운수업 종사자의 위험성향을 위험감행행동과 교통 통제소재의 측면에서 분석하였다. 위험감행행동 측면의 분석을 위해 운전자의 위험감행성향의 위험감행행동에 대한 예측모형을 제시하였다. 이를 통해 운전자의 위험감행성향은 위험감행행동에 유의미한 영향을 주고 있음을 알 수 있었으며, 지각운전능력이 높고 준법의식이 결여될수록 위반행동을 많이 하는 것으로 나타났다. 둘째, 교통 통제소재(외적통제형, 내적통제형)에 따라 위험감행 수준의 차이가 있는지 다변량 분석을 통해 알아보았다. 외적통제형(External Control) 운전자의 경우 위험감행수준이 높고 내적통재형(Internal Control)의 경우 위험감행수준이 낮게 분석되어 사고책임을 귀인(歸因)하는 태도에 따라 위험감행성향에 차이가 있는 것을 볼 수 있었다. 본 연구에서 구조방정식 모형을 통해 구축한 위험감행성향과 관련된 위험감행 예측모형에서 ${\chi}^2$값은 279.7, ${\chi}^2$/df=1.55, RMSEA=.44, GFI=.911, TLI=.916, CFI=.929로 적합한 부합도를 보이고 있다.

• 한국학교ㆍ지역보건교육학회지
• /
• 제22권2호
• /
• pp.1-14
• /
• 2021

Objectives: The study aimed to analyze the effects of household financial difficulties caused by COVID-19 (2019 coronavirus disease) on suicidal tendencies of adolescents. Methods: We selected 54,948 middle and high school students who were surveyed based on the Korean Youth Risk Behavior Web-based Survey 2020. To analyze the data, we used the STATA 16.0 program to conduct propensity score matching (PSM). Results: After controlling for selection effects by using PSM, the household financial difficulties caused by COVID-19 maintained a significant predictive effect on increasing suicidal ideation, suicide attempts, and decreasing tendency in hospital-treated after suicide attempts. However, depressive symptoms and suicide plan did not show a significant correlation with household financial difficulties associated with COVID-19. Conclusions: It was found that the rate of suicidal ideation and suicide attempts among adolescents who experienced a household financial difficulties due to COVID-19. Therefore, It can provide empirical evidence for estimating the impact of COVID-19 on adolescent suicide rates.

• 한국조사연구학회지:조사연구
• /
• 제9권1호
• /
• pp.87-100
• /
• 2008

200여개의 지역구 선거가 동시에 치러지는 국회의원 선거에서 주요정당의 의석 수를 예상해야 할 필요가 있는데, 이제까지는 정당별로 당선확실 선거구 수에 경합 선거구 수를 적당히 더하는 상식적 수준의 셈에 의존하여 왔다. 그러나 선거 예측 조사 자료를 베이즈 추론의 틀에 넣어 활용함으로써 정당 의석 수에 대한 합리적 점 예측과 구간 예측이 가능하다. 2004년의 제 17대 국회의원 선거에 적용하여 이 방법의 실용성을 살펴보았다.

• 대한가정학회지
• /
• 제34권4호
• /
• pp.103-117
• /
• 1996

This study is a theoretical approach by literature about life style and housing life style. The purpose of this research is making an analysis model of housing life style which can be used as a conceptual framework in empirical study. the theme of‘Life style’was studied mainly in the area of consumerism and housing. Model was made to explore the sub-domain of new analysis model by the microsociological approach. As results, 1) Independent variable of housing life style research model must be selected to be possible of comparison in effect of objective and subjective variable and in effect of predictive function and well-being function of housing life style. 2) Sub-domain of housing life style must be consisted of value orientation of family life, consumption in house, consumer durables, furniture., and propensity to using space. 3) Conceptual model of housing life style must be tested in empirical study to know what is the housing adjustment behavior of individual family, to improve quality of housing life and to suggest housing policy for family as a consumer.

• International Journal of Advanced Culture Technology
• /
• 제11권3호
• /
• pp.190-198
• /
• 2023

Sentiment analysis can be applied to all texts generated from websites, blogs, messengers, etc. The study fulfills an artificial intelligence sentiment analysis estimating visiting evaluation opinions (reviews) and visitor ratings, and suggests a deep learning model which foretells either an affirmative or a negative inclination for new reviews. This study operates review big data about Jeju tourist attractions which are extracted from Google from October 1st, 2021 to November 30th, 2021. The normalization data used in the propensity prediction modeling of this study were divided into training data and test data at a 7.5:2.5 ratio, and the CNN classification neural network was used for learning. The predictive model of the research indicates an accuracy of approximately 84.72%, which shows that it can upgrade performance in the future as evaluating its error rate and learning precision.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권4호
• /
• pp.1789-1795
• /
• 2016

Background: Breast cancer is the commonest female cancer worldwide and its propensity to metastasize negatively impacts on therapeutic outcome. Several clinicopathological parameters with prognostic/predictive significance have been associated with metastatic suppressor expression levels. The role of metastatic suppressor gene (MSG) KiSS1 in breast cancer remains unclear. Our goal was to investigate the possible clinical significance of KiSS1 breast cancer. Materials and Methods: The study was conducted on 87 histologically proven cases of breast cancer and background normal tiisue. Quantitative reverse transcriptase polymerase chain reaction (qRT PCR) and immunohistochemistry (IHC) were used to investigate KiSS1 at gene and protein levels, respectively, for correlation with several patient characteristics including age, family history, hormonal receptor status, stage, tumor size, nodal involvement and metastatic manifestation and finally with median overall survival (OS). Results: Our study revealed (i) KiSS1 levels were generally elevated in breast cancer vs normal tissue (P < 0.05). (ii) however, a statistically significant lower expression of KiSS1 was observed in metastatic vs non metastatic cases (P = 0.04). (iii) KiSS1 levels strongly correlated with T,N,M category, histological grade and advanced stage (p<0.001) but not other studied parameters. (iv) Lastly, a significant correlation between expression of KiSS1 and median OS was found (P = 0.04). Conclusions: Conclusively, less elevated KiSS1 expression is a negative prognostic factor for OS, advancing tumor stage, axillary lymph node status, metastatic propensity and advancing grade of the breast cancer patient. Patients with negative KiSS1 expression may require a more intensive therapeutic strategy.

• Radiation Oncology Journal
• /
• 제36권4호
• /
• pp.276-284
• /
• 2018

Purpose: Traditionally, three-dimensional conformal radiation therapy (3D-CRT) is used for neoadjuvant chemoradiation in locally advanced rectal cancer. Intensity-modulated radiation therapy (IMRT) was later developed for more conformal dose distribution, with the potential for reduced toxicity across many disease sites. We sought to use the National Cancer Database (NCDB) to examine trends and predictors for IMRT use in rectal cancer. Materials and Methods: We queried the NCDB from 2004 to 2015 for patients with rectal adenocarcinoma treated with neoadjuvant concurrent chemoradiation to standard doses followed by surgical resection. Odds ratios were used to determine predictors of IMRT use. Univariable and multivariable Cox regressions were used to determine potential predictors of overall survival (OS). Propensity matching was used to account for any indication bias. Results: Among 21,490 eligible patients, 3,131 were treated with IMRT. IMRT use increased from 1% in 2004 to 22% in 2014. Predictors for IMRT use included increased N stage, higher comorbidity score, more recent year, treatment at an academic facility, increased income, and higher educational level. On propensity-adjusted, multivariable analysis, male gender, increased distance to facility, higher comorbidity score, IMRT technique, government insurance, African-American race, and non-metro location were predictive of worse OS. Of note, the complete response rate at time of surgery was 28% with non-IMRT and 21% with IMRT. Conclusion: IMRT use has steadily increased in the treatment of rectal cancer, but still remains only a fraction of overall treatment technique, more often reserved for higher disease burden.