In order to control and minimize the damaging impacts of floods, flood modeling or simulation is a fundamental solution. Identifying effective models for this purpose is crucial in watershed management. This study evaluates the accuracy of support vector machine models combined with the support vector machine (SVM), Grasshopper algorithm (SVM-GOA) and least square support vector machine (LS-SVM) in simulating the flood peak discharge of Poldokhtar station in the Karkheh basin. For this study, 74 flood events from 2009 to 2016 at the Poldokhtar station and data from 13 daily rainfall stations in the upstream area for the same period were utilized. Subsequently, 52 events were allocated for training, and 22 for validation. The comparison of results was conducted using three statistical indicators: Correlation coefficient (R2), Root mean square error (RMSE), Nash efficiency (Ns), and Standard error (SE). Additionally, uncertainty analysis was performed using two indexes: ARIL and POC. The results indicate the relative superiority of the LS-SVM model with SE=0.407, RMSE=110.16, NS= 0.91 and R2=0.92 compared to the SVM model with SE=0.5, RMSE=137.70, NS= 0.87 and R2=0.88 and SVM-GOA model with SE=0.519, RMSE=144.53, NS= 0.83  and R2=0.9. The study's overall conclusion is that the LS-SVM model is more accurate, faster, and easier to implement compared to the SVM and SVM-GOA models. As a result, it can be confidently preferred over the SVM and SVM-GOA models due to its significant advantages. The research emphasizes the critical importance of precise flood modeling and simulation in watershed management for mitigating the destructive impact of floods.

One of the major errors that affect GPS accurately is the multi-path effect of each receiver. Multi-paths is receiving an antenna signal from more than one path, multi-path effect is a major source of unknown error in positioning and is not eliminated by differential methods. This effect is largely dependent on the environment specific to each receiver and it is low-frequency effect. The geometry between GPS satellites and the specific location of each receiver is repeated on astronomical days, the multi-path effects tend to behave similarly on consecutive days. In this paper, a method for extracting the multi-path effects behavior was applied to the GPS-code observations, multi-path error mitigation increases the accuracy of positioning. In the proposed method, the residual signal is generated based on the dual difference (DD) and is used as the input of the proposed algorithm. Support Vector Machine (SVM) is used for multi-path approximation. To determine the basic parameters of SVM and its kernel function, particle optimization algorithms (PSO) and genetic algorithm (GA) were used. In order to evaluate the accuracy of the proposed method, simulation and experimental based on two stations (reference and user) and two low-cost receivers were designed. The proposed methods were tested based on practical data. The experiments showed that the multi-path error of the receiver of the user's station decreased by 70% in the static test based on the RMS criterion. Models of this paper have been compared with some recent models presented in the context of multi-path error reduction. The results showed that the proposed model had better performance than other methods. The result is high accuracy and stability in positioning results. Three-dimensional position accuracy improved by about 56% after using the proposed method, reaching 1.60 m.

In the banking industry, one issue that must always be considered by the credit policy makers is risk management. Among various risks which banks are dealing with, credit risk is most important. It is caused by the losses of disability or lack of tendency of borrowers to pay their credit obligations. To manage and control the mentioned risk, classification systems are undeniable requirement. Such systems, according to existent documents and information, determine the class of customers. It is evident that use of these systems helps bank to choose customers in a good way and through the control and reduction the credit risk, improves efficiency level of providing bank facilities. In This research artificial intelligent based classification model consist of support vector machine is used to predict bank legal customers financial performance. Indeed, in this paper SVM is used with other mechanisms like F-score and Grid search to increase the accuracy of the model and classify the legal customers. The results justify the improvements in the classification accuracy and demonstrate that SVM can provide the better accuracy than other models.

Tuberculosis (TB) is a dangerous disease caused by mycobacterium leads to mortality. Early detection and identification of tuberculosis is crucial for managing tuberculosis infections. Recent technological improvements use a machine learning-based SVM and Modified CNN to identify specific diseases more accurately, as demonstrated in this research. The modified CNN's improved feature extraction and classification accuracy are maintained throughout construction. To obtain good performance a TBX11K publicly accessible dataset is used it consists of 11000 images of which 4600 chest x-ray (CXR) images are considered in this research, and the suggested model is verified. This approach significantly increases the accuracy of categorizing TB symptoms.  The PCA in this system locates the elements and extracts a large amount of variance technique applied to the full chest radiograph for pulmonary tuberculosis identification accuracy using SVM is 93.14% and modified CNN 96.72% respectively. When it comes to helping radiologists diagnose patients and public health professionals screen for tuberculosis in places where the disease is endemic, the proposed system SVM and modified CNN perform better than existing methods.