Prediction And Evaluation of Forest Fire In Yunnan of China Based On Geographically Weighted Logistic Regression Model

Establishing an effective forest fire forecasting mechanism is the premise of scientific planning and management of forest fires and forest fire prevention. In recent years, the forest fire prediction mechanism has been one of the key areas of concern for the government forestry management departments and forestry researchers. One of them, is logistic regression ( LR ). It is a relatively frequent prediction probability model used in forest fire prediction and prediction in China and abroad for the past few years. However, with the gradual deepening of research, it is found that the logistic regression model fails to fully consider the spatial non-stationary relationship between forest fires and driving factors, which leads to poor fitting effect and low prediction accuracy of the model. But its extended counterpart, the Geographically weighted logistic regression ( GWLR ) model, takes into account the spatial correlation between model variables, and effectively improves the fitting ability and prediction accuracy of the model. Therefore, this paper compares the ability of the logistic regression model and the geographically weighted logistic regression model in terms of fitting ability and prediction accuracy in order to obtain the ability of the two models to predict forest fires in Yunnan Province. In this paper, the samples were divided into 60% training samples and 40% test samples, and repeated sampling was carried out 5 times for training. Variables that appeared in the training model for 3 or more times were used to construct the final LR and GWLR models. Finally, the models with better fitting ability and higher prediction accuracy were used to classify the fire risks in Yunnan Province. The results show that the geographically weighted logistic regression model is superior to the logistic regression model in terms of fitting effect and accuracy. The geographically weighted logistic regression model is more suitable for the data structure of forest fires in Yunnan Province and has better prediction ability. The AUC value of the geographically weighted logistic regression model is 0.902, and the prediction accuracy is 82.7 %; The AUC value of logistic regression model was 0.891, and the prediction accuracy was 80.1%; Fully considering the spatial heterogeneity among model variables can, to some extent, predict forest fires more accurately. The fitting of the two models shows that the relative humidity, temperature, air pressure, sunshine hours, daily precipitation, wind speed, and other meteorological factors; Vegetation type; terrain factor; Population density, road network and other human activity factors become the cause of forest fires in Yunnan Province.

The use of entropy based fuzzy membership on weighted logistic regression for the unbalanced data

Logistic regression is a popular and powerful classification method. The addition of ridge regularization and optimization using a combination of linear conjugate gradients and IRLS, called Truncated Regularized Iteratively Re-weighted Least Square (TR-IRLS), can outperform Support Vector Machine (SVM) in terms of processing speed, especially when applied to large data and have competitive accuracy. However, neither SVM nor TR-IRLS is good enough when applied to unbalanced data. Fuzzy Support Vector Machine (FSVM) is an SVM development for unbalanced data that adds fuzzy membership to each observation. The fuzzy membership makes the interest of each observation in the minority class higher than the majority class. Meanwhile, TR-IRLS developed into a Rare Event Weighted Logistic Regression (RE-WLR) by adding weight to logistic regression and bias correction. The weighting of the RE-WLR depends on the undersampling scheme. It allows an “information loss”. Between FSVM and RE-WLR has a similarity, the weight based only on class differences (minority or majority). Entropy Based Fuzzy Support Vector Machine (EFSVM) is a method used to accommodate the weaknesses of FSVM by considering the class certainty of class observations. As a result, EFSVM is able to improve SVM performance for unbalanced data, even beating FSVM. For this reason, we use EF on the TR-IRLS algorithm to classify large and unbalanced data, as a proposed method. This method is called Entropy-Based Fuzzy Weighted Logistic Regression (EF-WLR). This Research shows the review of EF-WLR for unbalanced data classification.

Pemodelan Semiparametrik Geographical Weighted Logistic Regression pada Data Kemiskinan di Provinsi Sulawesi Selatan Tahun 2017

The level of poverty in a Regency/city in South Sulawesi in 2017 is different. The grouping of poverty status can be done based on the value of the HeadCount Index (HCI) of South Sulawesi. Factors affecting poverty will differ for each area being observed. The statistical modeling method developed for data analysis by taking into account the location factor is semiparametric Geographical Weighted Logistic Regression (GWLR). The GWLR semiparametric Model consists of parameters that are affected by the location and not affected by the location. The parameter estimator of the GWLR semiparametric model used in this research was obtained using the maximum method likelihood estimation. The result of a semiparametric model of GWLR each district/city in South Sulawesi in 2017 has the value Estimator parameter for global parameters is the same value for each location, namely, a3 = 0.1724, a4 = 0.0204, and a6 = 0.0261 whereas the parameter estimator for local parameters has different values so that GWLR semiparametric model of each district/city.

Association of obesity with diabetic retinopathy in US adults with diabetes in a national survey

Objective: There is a lack of consensus on whether a high body mass index (BMI) increases the risk of diabetic retinopathy (DR). We aimed to investigate the association between BMI, overweight, obesity, and DR using the data of diabetes respondents in the 2015 US Behavioral Risk Factor Surveillance System survey. Methods: Diabetes respondents aged over 18-year-old with complete information as well as undergone fundus examination in the past two years or had been diagnosed with DR were included. Weighted logistic regression analyses were used to identify the association of BMI with DR. Results: Among the 21,647 diabetes respondents, 4588 respondents had DR with a weighted prevalence of 22.5%. The mean BMI of all diabetes respondents was 31.50±6.95 kg/m2 with18,498 (86.5%) overweight and 11,353 (54.6%) obese. The mean BMI of the DR group (31.83±7.41kg/m2) was significantly higher than that of the non-DR group (31.41±6.81kg/m2, p<0.05). The proportion of obese respondents in the DR group was higher than the non-DR group (54.3%, p<0.001).The weighted prevalence of DR was 0.8%, 13.8%, 29.7%, and 55.7% for the emaciation group, the normal weight group, the overweight group, and the obesity group, respectively (p<0.001). Weighted logistic regression analysis showed that both BMI (adjusted OR=1.004, 95%CI 1.003-1.004) and obesity (adjusted OR=1.051, 95%CI 1.048-1.055) were associated with DR after adjusting for the confounding variables. However, overweight was not significantly associated with DR. Conclusion: The prevalence of DR in the normal weight, overweight, and obesity groups increased gradually. Obesity, rather than overweight, was significantly associated with increased DR prevalence.

Predicting Eurasian Watermilfoil and Curly-leaf Pondweed Presence-Absence in Adirondack Lakes Using Geographically Weighted Logistic Regression

Aquatic invasive species, Eurasian Watermilfoil (EWM) and Curly-leaf Pondweed (CLP), have been dispersing across New York, USA and are threatening the ecosystem of Adirondack Park, a state park with a large forest preserve and heavily frequented by tourists. In this study, the prediction of EWM and CLP invasion across Adirondack Park lakes is modeled using logistic regression (LR) and geographically weighted logistic regression (GWLR) with lake, landscape, and climate variable predictors. EWM presence-absence is found to be best predicted by nearby invaded lakes, human presence, and elevation. The presence-absence of CLP models have similar findings, with the addition of game-fish abundance being important. GWLR increases model performance and prediction, with explained variation of EWM and CLP increasing by 23% and 16% and the percent correctly predicted increasing by 2.6% and 0.9%. The study shows that GWLR, a relatively novel methodology, works better than common LR models for predicting invasion of EWM and CLP across Adirondack Park, and corroborates anthropogenic influences on dispersal of aquatic invaders.

Predicting Eurasian Watermilfoil and Curly-leaf Pondweed Presence-Absence in Adirondack Lakes Using Geographically Weighted Logistic Regression

Aquatic invasive species, Eurasian Watermilfoil (EWM) and Curly-leaf Pondweed (CLP), have been dispersing across New York, USA and are threatening the ecosystem of Adirondack Park, a state park with a large forest preserve and heavily frequented by tourists. In this study, the prediction of EWM and CLP invasion across Adirondack Park lakes is modeled using logistic regression (LR) and geographically weighted logistic regression (GWLR) with lake, landscape, and climate variable predictors. EWM presence-absence is found to be best predicted by nearby invaded lakes, human presence, and elevation. The presence-absence of CLP models have similar findings, with the addition of game-fish abundance being important. GWLR increases model performance and prediction, with explained variation of EWM and CLP increasing by 23% and 16% and the percent correctly predicted increasing by 2.6% and 0.9%. The study shows that GWLR, a relatively novel methodology, works better than common LR models for predicting invasion of EWM and CLP across Adirondack Park, and corroborates anthropogenic influences on dispersal of aquatic invaders.