• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권1호
• /
• pp.40-49
• /
• 2004

In this paper, we propose a new architecture of Genetic Algorithms(GAs)-based Self-Organizing Polynomial Neural Networks(SOPNN), discuss a comprehensive design methodology and carry out a series of numeric experiments. The conventional SOPNN is based on the extended Group Method of Data Handling(GMDH) method and utilized the polynomial order (viz. linear, quadratic, and modified quadratic) as well as the number of node inputs fixed (selected in advance by designer) at Polynomial Neurons (or nodes) located in each layer through a growth process of the network. Moreover it does not guarantee that the SOPNN generated through learning has the optimal network architecture. But the proposed GA-based SOPNN enable the architecture to be a structurally more optimized network, and to be much more flexible and preferable neural network than the conventional SOPNN. In order to generate the structurally optimized SOPNN, GA-based design procedure at each stage (layer) of SOPNN leads to the selection of preferred nodes (or PNs) with optimal parameters- such as the number of input variables, input variables, and the order of the polynomial-available within SOPNN. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. A detailed design procedure is discussed in detail. To evaluate the performance of the GA-based SOPNN, the model is experimented with using two time series data (gas furnace and NOx emission process data of gas turbine power plant). A comparative analysis shows that the proposed GA-based SOPNN is model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• Membrane and Water Treatment
• /
• 제7권2호
• /
• pp.127-141
• /
• 2016

Electrodialysis (ED) is known to be a useful membrane process for desalination, concentration, separation, and purification in many fields. In this process, it is desirable to work at high current density in order to achieve fast desalination with the lowest possible effective membrane area. In practice, however, operating currents are restricted by the occurrence of concentration polarization phenomena. Many studies showed the occurrence of a limiting current density (LCD). The limiting current density in the electrodialysis process is an important parameter which determines the electrical resistance and the current utilization. Therefore, its reliable determination is required for designing an efficient electrodialysis plant. The purpose of this study is the development of a predictive model of the limiting current density in an electrodialysis process using response surface methodology (RSM). A two-factor central composite design (CCD) of RSM was used to analyze the effect of operation conditions (the initial salt concentration (C) and the linear flow velocity of solution to be treated (u)) on the limiting current density and to establish a regression model. All experiments were carried out on synthetic brackish water solutions using a laboratory scale electrodialysis cell. The limiting current density for each experiment was determined using the Cowan-Brown method. A suitable regression model for predicting LCD within the ranges of variables used was developed based on experimental results. The proposed mathematical quadratic model was simple. Its quality was evaluated by regression analysis and by the Analysis Of Variance, popularly known as the ANOVA.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권2호
• /
• pp.867-870
• /
• 2014

Purpose: This study used receiver operating characteristic curve to analyze Surveillance, Epidemiology and End Results (SEER) ependymoma data to identify predictive models and potential disparity in outcome. Materials and Methods: This study analyzed socio-economic, staging and treatment factors available in the SEER database for ependymoma. For the risk modeling, each factor was fitted by a Generalized Linear Model to predict the outcome ('brain and other nervous systems' specific death in yes/no). The area under the receiver operating characteristic curve (ROC) was computed. Similar strata were combined to construct the most parsimonious models. A random sampling algorithm was used to estimate the modeling errors. Risk of ependymoma death was computed for the predictors for comparison. Results: A total of 3,500 patients diagnosed from 1973 to 2009 were included in this study. The mean follow up time (S.D.) was 79.8 (82.3) months. Some 46% of the patients were female. The mean (S.D.) age was 34.4 (22.8) years. Age was the most predictive factor of outcome. Unknown grade demonstrated a 15% risk of cause specific death compared to 9% for grades I and II, and 36% for grades III and IV. A 5-tiered grade model (with a ROC area 0.48) was optimized to a 3-tiered model (with ROC area of 0.53). This ROC area tied for the second with that for surgery. African-American patients had 21.5% risk of death compared with 16.6% for the others. Some 72.7% of patient who did not get RT had cerebellar or spinal ependymoma. Patients undergoing surgery had 16.3% risk of death, as compared to 23.7% among those who did not have surgery. Conclusion: Grading ependymoma may dramatically improve modeling of data. RT is under used for cerebellum and spinal cord ependymoma and it may be a potential way to improve outcome.

• Journal of Korean Neurosurgical Society
• /
• 제52권6호
• /
• pp.528-533
• /
• 2012

Objective : In the pediatric population the skull has not yet undergone ossification and it is assumed that the diagnostic rate of skull fractures by simple X-rays are lower than that of adults. It has been recently proposed that the diagnostic rates of skull fractures by 3-dimensional computer tomography (3D-CT) are higher than simple X-rays. The authors therefore attempted to compare the diagnostic rates of pediatric skull fractures by simple X-rays and 3D-CTs with respect to the type of fracture. Methods : One-hundred patients aged less than 12 years who visited the Emergency Center for cranial injury were subject to simple X-rays and 3D-CTs. The type and location of the fractures were compared and Kappa statistical analysis and the t-test were conducted. Results : Among the 100 pediatric patients, 65 were male and 35 were female. The mean age was $50{\pm}45$ months. 63 patients had simple skull fractures and 22 had complex fractures, and the types of fractures were linear fractures in 74, diastatic fractures 15, depressed fractures in 10, penetrating fracture in 1, and greenstick fractures in 3 patients. Statistical difference was observed for the predictive value of simple skull fractures' diagnostic rate depending on the method for diagnosis. A significant difference of the Kappa value was noted in the diagnosis of depressed skull fractures and diastatic skull fractures. Conclusion : In the majority of pediatric skull fractures, 3D-CT showed superior diagnosis rates compared to simple skull X-rays and therefore 3D-CT is recommended whenever skull fractures are suspected. This is especially true for depressed skull fractures and diastatic skull fractures.

• 한국환경복원기술학회지
• /
• 제11권3호
• /
• pp.28-38
• /
• 2008

The purpose of this study are to compare existing presence-absence predictive models and to predict suitable habitat for Goral (Nemorhaedus caudatus raddeanus) that is an endangered and protected species in Seoraksan national park using the best model among existing predictive models. The methods of this study are as follows. First, 375 location data and 9 environmental data layers were implemented to build a model. Secondly, 4 existing presence-absence models : Generalized Linear Model (GLM), Generalized Addictive Model (GAM), Classification and Regression Tree (CART), and Artificial Neural Network (ANN) were tested to predict the Goal habitat. Thirdly, ROC (Receiver Operating Characteristic) and Kappa statistics were used to calculate a model performance. Lastly, we verified models and created habitat suitability maps. The ROC AUC (Area Under the Curve) and Kappa values were 0.697/0.266 (GLM), 0.729/0.313 (GAM), 0.776/0.453 (CART), and 0.858/0.559 (ANN). Therefore, ANN was selected as the best model among 4 models. The models showed that elevation, slope, and distance to stream were the significant factors for Goal habitat. The ratio of predicted area of ANN using a threshold was 31.29%, but the area decreased when human effect was considered. We need to investigate the difference of various models to build a suitable wildlife habitat model under a given condition.

• 한국인터넷방송통신학회논문지
• /
• 제18권4호
• /
• pp.227-235
• /
• 2018

본 연구는 예비유아교사의 로봇활용교육 수용의도에 영향을 미치는 요인을 파악하기 위해 이루어졌다. 이를 위해 3년제 전문대학 재학생 259명을 대상으로 설문조사를 실시하고 수집된 자료는 SPSS 23.0을 이용하여 분석하였다. 예비유아교사의 배경변인별 수용의도와 각 예측요인의 차이를 살펴보기 위해 t-test와 일원분산분석을 실시하고, 지각된 용이성, 지각된 유용성, 혁신의지, 사회적 영향력이 예비유아교사의 로봇활용교육 수용의도에 어떠한 영향을 미치는지 알아보기 위해 중다선형 회귀분석을 실시하였다. 연구결과, 배경변인에 따라 수용의도와 각 예측요인은 의미있는 차이를 보이지 않았으며, 여러 예측요인 중 지각된 용이성과 지각된 유용성만 수용의도에 영향을 미치는 것으로 분석되었다. 또한, 지각된 용이성에는 혁신의지와 사회적 영향력이, 지각된 유용성에는 지각된 용이성과 사회적 영향력이 영향을 미치는 것으로 분석되었다.

• 제어로봇시스템학회논문지
• /
• 제8권2호
• /
• pp.126-135
• /
• 2002

The study is concerned with an approach to the design of new architectures of fuzzy neural networks and the discussion of comprehensive design methodology supporting their development. We propose an Adaptive Fuzzy Polynomial Neural Networks(APFNN) based on Fuzzy Neural Networks(FNN) and Self-organizing Networks(SON) for model identification of complex and nonlinear systems. The proposed AFPNN is generated from the mutually combined structure of both FNN and SON. The one and the other are considered as the premise and the consequence part of AFPNN, respectively. As the premise structure of AFPNN, FNN uses both the simplified fuzzy inference and error back-propagation teaming rule. The parameters of FNN are refined(optimized) using genetic algorithms(GAs). As the consequence structure of AFPNN, SON is realized by a polynomial type of mapping(linear, quadratic and modified quadratic) between input and output variables. In this study, we introduce two kinds of AFPNN architectures, namely the basic and the modified one. The basic and the modified architectures depend on the number of input variables and the order of polynomial in each layer of consequence structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the AFPNN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed AFPNN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Parasites, Hosts and Diseases
• /
• 제45권4호
• /
• pp.273-282
• /
• 2007

In acute uncomplicated falciparum malaria, there is a continuum from mild to severe malaria. However, no mathematical system is available to predict uncomplicated falciparum malaria patients turning to severe malaria. This study aimed to devise a simple and reliable model of Malaria Severity Prognostic Score (MSPS). The study was performed in adult patients with acute uncomplicated falciparum malaria admitted to the Bangkok Hospital for Tropical Diseases between 2000 and 2005. Total 38 initial clinical parameters were identified to predict the usual recovery or deterioration to severe malaria. The stepwise multiple discriminant analysis was performed to get a linear discriminant equation. The results showed that 4.3% of study patients turned to severe malaria. The MSPS = 4.38 (schizontemia) + 1.62 (gametocytemia) + 1.17 (dehydration) + 0.14 (overweight by body mass index; BMI) + 0.05 (initial pulse rate) + 0.04 (duration of fever before admission)-0.50 (past history of malaria in last 1 year). 0.48 (initial serum albumin)-5.66. Based on the validation study in other malaria patients, the sensitivity and specificity were 88.8% and 88.4%, respectively. We conclude that the MSPS is a simple screening tool for predicting uncomplicated falciparum malaria patients turning to severe malaria. However, the MSPS may need revalidation indifferent geographical areas before utilized at specific places.

• 한국정보통신학회논문지
• /
• 제23권1호
• /
• pp.20-26
• /
• 2019

미세먼지의 인체 영향으로 인해 기존 대기 환경 모니터링 네트워크에서 측정된 과거 데이터를 활용하여 미세먼지를 예측하려는 다양한 연구가 진행되고 있다. 하지만 기존 설계된 예측 모델의 측정 환경, 세부 조건을 정확히 설정하기 어려우며, 측정된 기상 데이터의 누락과 같은 문제로 기존 연구 결과에 기반 한 새로운 예측 모델의 설계가 필요하다. 본 논문에서는 미세먼지 예측을 위한 선행 연구로서 다수의 연구에서 사용된 기계 학습 알고리즘인 다중 선형 회귀와 인공 신경망을 통해 예측 모델을 설계하여 미세먼지 예측을 위한 알고리즘의 적합성을 평가하였다. RMSE를 통한 예측 성능 비교 결과, MLR 모델의 경우 18.13, MLP 모델의 경우 14.31의 값을 보여 미세먼지 농도를 예측함에 있어 인공 신경망 모델이 예측에 더 적합함을 보였다.

• Imaging Science in Dentistry
• /
• 제49권2호
• /
• pp.123-130
• /
• 2019

Purpose: This study was conducted to characterize the relationship of the angulation between the tooth root axis and alveolar bone axis with anterior alveolar(AA) arch forms and sagittal root position (SRP) in the anterior esthetic region using cone-beam computed tomography (CBCT) images. Materials and Methods: CBCT images that met the inclusion and exclusion criteria were categorized using a recent classification of AA arch forms and a SRP classification. Then, the angulation of the root axis and the alveolar bone axis was measured using mid-sagittal CBCT images of each tooth. The relationships of the angulation with each AA arch form and SRP classification were evaluated using 1-way analysis of variance and a linear regression model. Results: Ninety-eight CBCT images were included in this study. SRP had a greater influence than the AA arch form on the angulation of the root axis and the alveolar bone axis(P<0.05). However, the combination of AA arch form and SRP was more predictive of the angulation of the root axis and the alveolar bone axis than either parameter individually. Conclusion: The angulation of the root axis and alveolar bone axis demonstrated a relationship with the AA arch form and SRP in teeth in the anterior esthetic region. The influence of SRP was greater, but the combination of both parameters was more predictive of root-to-bone angulation than either parameter individually, implying that clinicians should account for both the AA arch form and SRP when planning implant placement procedures in this region.