support vector machine (svm)

Over time, numerous studies have been conducted to read license plates and recognize license plates. However, it is noteworthy that these studies usually do not have the ability to learn complex structures in images with high accuracy. For this purpose, this paper uses the high capacities of deep neural networks to learn license plate identifiers. The proposed model in this paper includes two main steps: highlighting license plates and reading the ID. In the proposed model, the support vector machine (SVM) network is used to select the best range. After identifying the range of the license plate, its characters must be recognized. In this step, a gated convolutional neural network (GCNN) will be used. The proposed model is evaluated on two datasets, FZU Cars and Stanford Cars, and the results of the experiments show that this model has higher accuracy than other methods presented in both datasets.

The issue of fraud in insurance claims is one of the problems faced by insurance companies. Therefore, the issue of discovering such frauds in all types of insurances is one of the topics of interest for experts in various fields. Insurance fraud can be defined as taking damages from insurance companies by resorting to fraudulent means and documents. Losses caused through fraudulent activities affect the interests of insurers and potentially their financial stability. The current research uses data mining techniques to identify the fraudulent behavior of life insurance policyholders in insurance companies in order to identify the factors affecting these behaviors. The results of the article show that decision tree and support vector machine techniques are useful in identifying frauds and can be considered as the main center in business management to detect fraud. The results of the implementation of different methods on the studied dataset show the superiority of the neural network method over other methods. The neural network method has succeeded in classifying the desired classes in this research with an accuracy of 90.83, which is a good accuracy. Also, from the created decision tree, it is possible to detect frauds or the possibility of violations before issuing the insurance policy by using the data of the insurers under investigation, and if the violation is proven, it can be prevented from being issued.

The emulsion of salt water in crude oil causes reduction of crude oil quality, catalyst poisoning in downstream industries, and equipment corrosion in various parts of crude oil transportation, refining and storage systems. The most common industrial method for separating saline water from crude oil is the application of chemical compounds. Procuring the required chemical compounds will cost a lot. In addition, the performance of chemical demulsifiers is highly dependent on the type and structure of crude oil. The complexity of crude oil’s structure makes it difficult to develop performance models for demulsifiers. In order to reduce the number of important parameters of crude oil in modeling, the ratio of asphaltene to the total of resin and aromatic was used as the discriminating factor of crude oil emulsion process. Considering the complexity of the considered model, a support vector machine was used to predict the performance of commercial demulsifiers. The most important challenge in support vector machines is tuning hyperparameters. To tune the hyperparameters in this study, the risk criterion was used in predicting efficiencies higher than 85% and increasing the correlation coefficient. In order to collect modeling data, four samples of crude oil along with two samples of common commercial demulsifiers were prepared in the Production units of the southwestern fields of the country, and the performance of commercial demulsifiers at different operating conditions was checked using the bottle test method. The performance of the proposed algorithm for tuning hyperparameters was compared with the Bayesian optimization algorithm. The results show that adjusting the support vector machine hyperparameter with the risk criterion helps to design a model with better accuracy for predicting the performance of commercial demulsifiers. Considering the importance of modeling the performance of demulsifiers in petroleum industry, the evaluation of the model was investigated with a new emulsion sample. The results show that support vector machine can provide an efficient model for predicting the performance of commercial demulsifiers.

The Internet of Things is an emerging information architecture based on the Internet that develops interaction between things and services in a safe and reliable environment. In fact, the purpose of this structure is to reduce the distance between the things of the physical world and information systems. In the Internet of Things, it is expected that intelligent devices will become active members in business and informational and social processes, so that they are able to interact between themselves and the external environment through the exchange of data and sensed information. In fact, the Internet of Things is a network of devices in which various things can communicate with other equipment with the help of computers and through Internet connections.  Recommendation technologies can help to more easily identify relevant artifacts and thus will become one of the key technologies in future IoT solutions.  The main task of recommender systems is to recommend service providers that meet the different needs of users. The paper porposes a Support Vector Machine (SVM) based algorithm and the TOPSIS multi-criteria decision-making model in order to create an effective recommender system and provide suggestions to users based on their preferences and increase user satisfaction. The experimental results show that the proposed recommender system can produce a series of objective recommendations that are effective based on accuracy and variety, novelty and high coverage. Finally, the results confirm the improvement in making recommendations .

Pipeline network is one of the major agents of the economic growth and development of a country, which is used to transport fluid fuels, wastewaters, and any other fluids. When these pipelines meet seas and oceans, pipes are laid on the solid bed and it causes changes in the flow pattern around the pipes. In result of these changes, the shear stress of the bed under pipelines and turbulence of current will be increased, and scour will occur under pipelines and the scour hole will form and develop. These holes cause damage and failure to the pipe due to the pipe weight and oscillating loads. Therefore, it is very important to study the scour depth and effective variables to reduce scour and prevent damage. Researchers have conducted experimental and numerical studies on scour phenomenon, considering various types of flows and conditions, and have provided different relations over the years.In this research, the effect of various factors on this phenomenon in waves is investigated using Gaussian process regression (GPR) and support vector machine (SVM). To this end, several laboratory data were used and after defining several non-dimensional parameters the performance of these methods was evaluated. The result of this research demonstrated that these methods are better than experimental relations and have promising outcomes. This study has shown that an SVM model with KC, Re and  variables in wave-induced current has the best results. It is worth mentioning that the KC variable has the most significant effect on the scour below pipelines.

This study detects the activity of disabled people living in smart homes based on monitoring their behavior and desires to achieve the final model and store it in the database of the building management system for decision making and automatic operation of equipment used to meet the needs of residents. First, a suitable set of sensors are installed to record and process data correctly in different parts of the building and some to control vital signs depending on the type of disability are installed on the body. . First: selecting sensors and implementing recording techniques, and focusing on the major activities of people with disabilities in smart homes, second: classifying and processing the data collected using the support vector machine (svm) algorithm; svm was performed using the optimal data congestion algorithm (pso) and finally the final pattern for storage in the building management system database.

Pipelines are used all around the world to transport fluids from one location to another. When these pipelines meet rivers, seas and oceans pipes are laid on the solid bed and it causes changes in the flow pattern around the pipes. In result of these changes, the shear stress of bed under pipelines and turbulence of current will be increased, and scour will occur under pipelines and the scour hole will form and develop. These holes cause damage and failure to the pipe due to the pipe weight. In case of failure of the pipe, irreparable damages will incur to the environment and there will be huge financial costs. Therefore, it is very important to study the scour depth and effective variables to reduce scour and prevent damages. Researchers have conducted experimental and numerical studies on scour phenomenon and have provided relations over the years.
In this research the effect of various factors on this phenomenon in steady current are investigated using Gaussian process regression (GPR) and support vector machine (SVM) and it is compared with the previous presented relations. To this end several laboratory data were used and after defining several non-dimensional parameters the performance of these methods was evaluated. The result of this research demonstrated that these methods are better than experimental relations and have promising outcomes. This study have shown that an SVM model with ℎ/D, D/d, Re and S0 variables in steady current have the best results. It is worth mentioning that ℎ and variables in steady current have the most significant effect on the scour below pipelines.

Prediction of flow discharge coefficient, Cd, for a sluice gate under free and submerged flow conditions is one of the essential issues in hydraulics. In recent years, various semi-empirical equations have been developed in order to predict Cd for a sluice gate that application of those formulas under submerged flow conditions suffered from large errors. The aim of the present research is to use Gaussian Process Regression (GPR) and Support Vector Machine (SVM) used in soft computing techniques, so that estimating Cd in submerged flow conditions and comparing the results with quasi-experimental methods are of interest, herein. For this purpose, an experimental dataset comprised of 122 data points were used to feed the methods utilized. Different combinations of dimensionless parameters were then prepared and the performance of the afore mentioned methods were assessed. The results showed that SVM with input parameters of 𝑦𝑡⁄𝑤, 𝑦0⁄𝑤, 1/𝐹𝑟2 and S by the values of Root Mean Square Error (RMSE=0.017), correlation coefficient (R=0.97) and Nash-Sutcliffe Equivalent (NSE=0.95) had a better performance than GPR and other semi-empirical approaches, indeed.

Now , gait recognition is one of the biometric methods that has received more and more attention researchers in machine vision and pattern recognition. However, one of the main challenges of technology is variations caused by covariate factors such as fast and slow gait, clothing, carrying conditions.this reserch tried is to provide issues in three basic parts: preprocessing, feature extraction, and classification. Using the Cassia data set, which is a large set with different gestures and carrying conditions. First, organized a new Spatio-temporal database with horizontal profiling of the silhouette. next, feature extracting by PCA. Finally, the SVM algorithm is used and classifies and identifies. The test results show the efficiency of the proposed algorithm under different conditions.

Summary This paper presents statistical and intelligent models to predict stability of different surfaces in underground open stope mining method on the basis of the stability graph method. For this purpose, logistic regression and support vector machine were used to predict the stability state of walls and back, separately. Introduction The open stope mining method can be considered as the most common method for underground mining of hard rock metal mines around the world. The most common method for the design and evaluation of stopes in the mining technique is the stability graph method which was introduced by Mathews et al. in 1981. The stability graph method is based on data collected from different mines including hydraulic radius, stability number (tunneling modified quality index, rock stress factor, joint adjustment factor, and gravity factor), and related stability conditions. The empirical nature of this method, which results in uncertainties in the values of the parameters, and the existence of different types of charts with different stability regions, interpreting the results becomes a challenge. This increases the risk of error in interpreting the results. In addition to these problems, considering the same conditions for all surfaces, such as walls and back, causes obvious errors in the results of this method. In this regard, this paper incorporated statistical and intelligent methods to predict stability state for a database of open stopes by means of stability graph parameters. The novelty of this paper in addition to the used methods is developing models for evaluation of walls and back separately.   Methodology and Approaches New models to predict the stability state of walls and surface of underground open stopes were developed by using statistical and intelligent techniques. For this purpose, logistic regression (LR), as the most common statistical method for classification, was used. Besides, support vector machine (SVM), which is based on the machine learning theory, was applied as a powerful intelligent technique for classification problems.   Results and Conclusions Results of developed models were compared to those of the stability graph method. The following main conclusions were derived from this study: The accuracy of back stability prediction by using the stability graph method, LR and SVM were 29%, 86%, and 95%, respectively. The accuracy of wall stability prediction by using the stability graph method, LR and SVM were 71%, 81%, and 90%, respectively. In comparison with the stability graph method, LR models of back and walls increase the accuracy of predictions by 57% and 10%, respectively. In comparison with the stability graph method, SCM models of back and walls increase the accuracy of predictions by 66% and 19%, respectively. It was concluded that SVM models possess higher performance in the prediction of stable states when compared to the LR models.

Today easy data access through the network has made it possible to steal them. Therefore, the security of computer systems has become increasingly important. Intrusion Detection Systems . as the last line of computer defense, can play an important role in attack resistance and their efficiencies has direct impact on network security. The Intrusion Detection Systems must extract the necessary strategies based on the connections and use them to detect new connections. Support Vector Machine is a Machine Learning method that it is popular to extract intrusion strategies in past decade. Although simplification of SVM returned it to popular method but it has constraints such as senility to kernel selection and it has not any optimization mechanism to determine the best of them. We model it as using of several kernels simultaneously and different weighting to them and dynamic SVM parameters. Due to the high complexity of this problem, conventional optimization methods are not able to solve it. Therefore, we propose a Distributed Self Adaptive Genetic Algorithm with Migration. On the other hand, due to the high volume of data in such issues, Autoencoder has been used to reduce data. The proposed approach is a hybrid method based on Autoencoder and improved Support Vector Machine with Distributed Self Adaptive Genetic Algorithm with Migration that it is evaluated by its execution on data set. The experimental results have demonstrated that the proposed system exhibits a high performance for attack detection based on precision and recall and it low time for intrusion.

Nowadays, the rapid growth of the use of information technology and computer networks has increased the transfer of information in a digital form. For this reason, the protection of data has become one of the most important challenges in this field. Watermarking is introduced as one of the newest and most important techniques for data protection. Audio watermarking is considered to be the most challenging type of watermarking due to the nature of audio files. The most extraction methods used in audio watermarking algorithms, which mainly use non-intelligent techniques based on the reverse of embedding rules for extraction phase of audio watermarking, often they are not able to extract watermarking exactly and have a lot of errors in extracting. Our proposed solution to solve this problem is to use an intelligent algorithm to extract the watermark. The purpose of this article is to provide a method that covered the weakness of non-intelligent extraction methods using trained machine learning classifier and helped to improve system performance. For the embedding operation, the Lifting Wavelet Transform (LWT) has been used in the proposed method; in the extraction operation, the Support Vector Machine (SVM) classifier is also used. The trained classifier is able to detect the effects of various attacks on the audio files and consequently, intelligent and precise extraction of watermark. The results of various experiments under different conditions indicate that this intelligent method has achieved appropriate imperceptibility and high capacity along with high robustness.