Enhanced Support Vector Machine Research Articles

Diabetes Prediction using Enhanced SVM and Deep Neural Network Learning Techniques – An Algorithmic Approach for Early Screening of Diabetes

Diabetes, caused by the rise in level of glucose in blood, has many latest devices to identify from blood samples. Diabetes, when unnoticed, may bring many serious diseases like heart attack, kidney disease. In this way, there is a requirement for solid research and learning model’s enhancement in the field of gestational diabetes identification and analysis. SVM is one of the powerful classification models in machine learning, and similarly, Deep Neural Network is powerful under deep learning models. In this work, we applied Enhanced Support Vector Machine and Deep Learning model Deep Neural Network for diabetes prediction and screening. The proposed method uses Deep Neural Network obtaining its input from the output of Enhanced Support Vector Machine, thus having a combined efficacy. The dataset we considered includes 768 patients’ data with eight major features and a target column with result “Positive” or “Negative”. Experiment is done with Python and the outcome of our demonstration shows that the deep Learning model gives more efficiency for diabetes prediction.

Twitter sentimental analysis from time series facts: the implementation of enhanced support vector machine

Sentiment analysis through textual data mining is an indispensable system used to extract the contextual social information from the texts submitted by the intended users. Now days, world wide web is playing a vital source of textual content being shared in different communities by the people sharing their own sentiments through the websites or web blogs. Sentiment analysis has become a vital field of study since based on the extracted expressions, individuals or the businesses can access or update their reviews and take significant decisions. Sentimental mining is typically used to classify these reviews depending on its assessment as whether these reviews come out to be neutral, positive or negative. In our study, we have boosted feature selection technique with strong feature normalization for classifying the sentiments into negative, positive or neutral. Afterwards, support vector machine (SVM) classifier powered with radial basis kernel with adjusted hyper plane parameters, was employed to categorize reviews. Grid search with cross validation as well as logarithmic scale were employed for optimal values of hyper parameters. The classification results of this proposed system provides optimal results when compared to other state of art classification methods.

Diabetes Prediction Using Enhanced SVM and Deep Neural Network Learning Techniques

Diabetes, caused by the rise in level of glucose in blood, has many latest devices to identify from blood samples. Diabetes, when unnoticed, may bring many serious diseases like heart attack, kidney disease. In this way, there is a requirement for solid research and learning model’s enhancement in the field of gestational diabetes identification and analysis. SVM is one of the powerful classification models in machine learning, and similarly, Deep Neural Network is powerful under deep learning models. In this work, we applied Enhanced Support Vector Machine and Deep Learning model Deep Neural Network for diabetes prediction and screening. The proposed method uses Deep Neural Network obtaining its input from the output of Enhanced Support Vector Machine, thus having a combined efficacy. The dataset we considered includes 768 patients’ data with eight major features and a target column with result “Positive” or “Negative”. Experiment is done with Python and the outcome of our demonstration shows that the deep Learning model gives more efficiency for diabetes prediction.

Intensive analysis of intrusion detection methodology over Mobile Adhoc Network using machine learning strategies

Mobile Adhoc Network (MANET) recently gained prominence due to the prevalence of handheld connectivity and their flexibility to supportability in specific non-permanent and instantaneous applications like floods and military situations. MANET offers great network utility, but comes with specific security challenges due to the fact that there is no central control, changing network topology, transient existence and uncoordinated communication. There are numerous proposals to use encryption and authentication measures to decrease the risk of security issues, especially as a first-line protection options. Although these risks cannot be removed entirely, an effective intrusion detection scheme is vital to keep unauthorized intrusion out of Mobile Adhoc Network. The role of intrusion identification on Mobile Adhoc Network is extremely difficult due to open medium, complex topology, dispersion, lack of centralized administration, and resource-constrained node groups. There is no direct analogue of a traditional intrusion detection system designed for Mobile AdHoc Networks Technology that can be used on the wireless network. The technology used in it must be flexible enough to accommodate ad hoc changes. This system implements new machine learning architecture that enhances detection to be much more. Intelligent Decision Support incorporates the high accuracy of Enhanced Support Vector Machine (eSVM) with the improved scalability of Rough Set Theory (RST).

An Intelligent Clustering Technique for Analysing the Performance of Students during Lockdown Period of Covid-19

Corona virus or simply Corona is the current leading pandemic of the world. It has affected students and their in education in higher numbers than any other sector putting them into a depression. Hence this research attempts to suggest solutions for reducing depression amongst students amidst the pandemic. This work proposes ESVMs (Enhanced Support Vector Machines) model for its predictions. Identifying student performances is complex issue as the numbers are voluminous and hence the objective of this research is to assess student performance prediction model by using an efficient clustering method. Missing values and irrelevant data are resolved in this work using SCCs (Statistical correlation Coefficients) which work on subject wise manner or student wise data. This work also provides a novel solution for data pre-processing. IFCM (Improved Fuzzy C-means clustering) proposed in this work identifies high quality clusters with robustness. Further, the use of PSO (Particle Swarm Optimization) in feature selections improves its efficiency of the given data. Classifications are executed by the proposed ESVMs which predicts student's grade with accuracy. The evaluation results of this study improve classification accuracy significantly when compared to existing prediction models. © 2021 Karadeniz Technical University. All rights reserved.

Short Term Wind Power Prediction Based on Data Regression and Enhanced Support Vector Machine

This paper presents a short-term wind power forecasting model for the next day based on historical marine weather and corresponding wind power output data. Due the large amount of historical marine weather and wind power data, we divided the data into clusters using the data regression (DR) algorithm to get meaningful training data, so as to reduce the number of modeling data and improve the efficiency of computing. The regression model was constructed based on the principle of the least squares support vector machine (LSSVM). We carried out wind speed forecasting for one hour and one day and used the correlation between marine wind speed and the corresponding wind power regression model to realize an indirect wind power forecasting model. Proper parameter settings for LSSVM are important to ensure its efficiency and accuracy. In this paper, we used an enhanced bee swarm optimization (EBSO) to perform the parameter optimization for LSSVM, which not only improved the forecast model availability, but also improved the forecasting accuracy.

Enhanced Support Vector Machine based on Metaheuristic Search for Multiclass Problems

Machine learning is an important field of artificial intelligent researches and it highly growing for real intelligent applications systems that relate brain computer interface to human brain activities. Support Vector Machine (SVM) is a popular machine learning approach, which can be used for pattern recognition, prediction and classification with many diverse applications. However, the SVM has many parameters, which have significant influences on the performance of SVM in terms of its prediction accuracy that is very important measure specifically with critical applications such that used in Medical applications. This paper proposed an enhanced SVM, which employs a meta-heuristic method, called scatter search to determine the optimal or near optimal values of the SVM parameters and its kernel parameters in multi-classification problem. Scatter search has the potential to determine the appropriate values of parameters for machine learning algorithms due its flexibility and sophistication. Therefore, the proposed method integrates the advantages of scatter search method with SVM to specify the appropriate setting of SVM parameters. The experimental results on lung cancer datasets and other standard datasets prove that the scatter search is practical method for tuning SVM parameters and enhance its performance, where the achieved results are better and comparable to other related methods.

Optimized Attribute Selection Using Artificial Plant (AP) Algorithm with ESVM Classifier (AP-ESVM) and Improved Singular Value Decomposition (ISVD)-Based Dimensionality Reduction for Large Micro-array Biological Data.

In the tremendous field of the bioinformatics look into, enormous volume of genetic information has been produced. Higher throughput gadgets are made accessible at lower cost made the age of Big data. In a time of developing information multifaceted nature and volume and the approach of huge information, feature selection has a key task to carry out in decreasing high dimensionality in AI issues. Dealing with such huge data has turned out to be incredibly testing strategy for choosing the exact features in enormous medical databases. Large clinical data frequently comprise of an enormous number of identifiers of the disease. Data mining when applied to clinical data for identification of diseases, a few identifiers are will not be much useful and sometimes may even have negative impacts. Consequently, when the FS is applied, it is vital as it can expel those insignificant disease identifiers. It likewise builds the adequacy of decision by a physician emotionally supportive network by viably diminishing the time of learning of the framework. In this paper, a unique approach is presented for the feature selection utilizing the Artificial Plant algorithm which uses the Enhanced Support Vector Machine classifier. The features got are additionally dimensionally decreased by presenting the Improved Singular Value Decomposition strategy; finally, enhancement is done by the outstanding BAT streamlining method. The examinations are completed with real-time large cervical cancer data and it demonstrated to be more effective than the current methods.

Market Basket Analysis with Enhanced Support Vector Machine (ESVM) Classifier for Key Security in Organization

Market Basket Analysis is considered to be one among the highly popular and efficient sort of data analysis exploited in the marketing and retailing field. The objective of market basket analysis lies in deciding the products purchased together by the customers. Its name has originated from the concept of customers filling into a shopping cart everything of all they had purchased (a "market basket") while doing shopping in the grocery. Having a knowledge of the products that customers buy in group can be quiteusefulfor a retailer or to any other organization. A store could make the best use of this information to keep the products that are often sold together in the same place, whereas a catalog or World Wide Web (WWW) merchant could utilize it for deciding the structure of their catalog and order form. Since several applications such as market basket analysis, fraud detection in web, medical diagnosis, census data, Customer Relationship Management of business that makes use of association rules exists, the process involving Decision making can be improved. Security is also regarded to bean important facet for transactions done individually and frequent itemsets for database that are horizontally partitioned. In order to render security for lastly bough often used itemsets for transaction purposes, this research work introduces a novel key security algorithm that uses RSA cryptographic technique which is classifier based. The classifier makes use of information about several often utilized itemsets and it provides a key value to the actual company. For instance, in case if there are any reliance users, only the valid users can obtain that market info. The rest of the users belonging to the reliance organization are not allowed to select the data’s key value. First, the frequent itemsets are mined with the help of association rule mining employing Probabilistic Graphical Model techniques. Then the Enhanced Support Vector Machine (ESVM) classifier checks the key values of the mined frequent itemsets.

Enhanced Support Vector Machine Based Leukaemia Cancer Classification

Leukaemia is a blood cancer that is characterized by the bone marrow’s unregulated and irregular generation of white blood cells (leukocytes) in the blood. By testing the microscopic blood cell images, diseases can be detected and its early diagnosis can be done. In the current work,at smart decision support systemutilizing microscopic images is proposedto make a diagnosisof acute lymphoblastic leukaemia. Identification through images is a quick and inexpensive technique since there is no equipment is specially needed for lab testing. The improved Bare-Bones Particle Swarm Optimization algorithm is usedto identify the important distinguishing featuresof blast cells and the healthy cells to help in carry out the acute lymphoblastic leukaemia classification with efficiency. To avoid the poor accuracy, the modified median filtering has been introduced in this novel, which helps in removing the noise from the images and maintains the edge which is used to improve the identification of leukocytes and lymphocytes during the segmentation process. Enhanced linear contrast stretching is introduced in image enhancement for enriching the image. In the next step, feature extraction is carried out through the second order statistical features. Bare-bones with Adaptive Bat Optimization (BBABO) is used for feature selection. At last, the classification is carried out by using the combined Fuzzy Neural Network (FNN) and here the Enhanced Support Vector Machine (ESVM) is a classifier called as FNESVM.

A Multimodal Biometrics for Aadhaar Based Secured E-Voting System

False polling is still a significant issue in elections in the latest moments. In this job, an effort is made to fix this issue using current Aadhaar card database and electoral biometrics. This scheme authenticates electors by combining multimodal biometrics such as picture, eye, and palm printing, after which registration is verified by verifying the age that enables only qualified applicants to register. The time needed for authentication is decreased by using the corresponding Aadhaar amount and multimodal biometrics. This is authentication will be achieved by inspecting whether the registered Aadhaar amount and multimodal biometrics match or not without linking it with the entire biometric database to boost authentication pace. The polling of ballots will be performed automatically so that space is decreased and results can be announced in less moment. The improvisations strive to increase the system's safety, efficiency, efficiency, scalability. This suggested a safe internet e-voting system (EVS) that utilizes as its backend UIDAI or Aadhaar database. In this job, the entry pictures are originally preprocessed and the removal of features is achieved using Improved Gabor filters. Enhanced Support Vector Machine Algorithm (eSVM) is used to match and classify features development of this scheme will render the polling method more comfortable and can, therefore, contribute to enhanced turn-out. Using this multimodal biometric system for voting purposes, election rigging was easily avoided.

A new fruit fly optimization algorithm enhanced support vector machine for diagnosis of breast cancer based on high-level features

BackgroundIt is of great clinical significance to develop an accurate computer aided system to accurately diagnose the breast cancer. In this study, an enhanced machine learning framework is established to diagnose the breast cancer. The core of this framework is to adopt fruit fly optimization algorithm (FOA) enhanced by Levy flight (LF) strategy (LFOA) to optimize two key parameters of support vector machine (SVM) and build LFOA-based SVM (LFOA-SVM) for diagnosing the breast cancer. The high-level features abstracted from the volunteers are utilized to diagnose the breast cancer for the first time.ResultsIn order to verify the effectiveness of the proposed method, 10-fold cross-validation method is used to make comparison among the proposed method, FOA-SVM (model based on original FOA), PSO-SVM (model based on original particle swarm optimization), GA-SVM (model based on genetic algorithm), random forest, back propagation neural network and SVM. The main novelty of LFOA-SVM lies in the combination of FOA with LF strategy that enhances the quality for FOA, thus improving the convergence rate of the FOA optimization process as well as the probability of escaping from local optimal solution.ConclusionsThe experimental results demonstrate that the proposed LFOA-SVM method can beat other counterparts in terms of various performance metrics. It can very well distinguish malignant breast cancer from benign ones and assist the doctor with clinical diagnosis.

Epileptic detection based on whale optimization enhanced support vector machine

A hybrid novel meta-heuristic optimization algorithm termed Whale Optimization Algorithm (WOA) and Support Vector Machines (SVMs) to obtain an Electroencephalogram (EEG) classification approach was proposed here in this paper for automatic seizure detection termed WOA-SVM. In the proposed approach, the Discrete Wavelet Transform (DWT) was applied to extract the main features and then decompose it into four level of decomposition tree. Furthermore, WOA was utilized to find the more significant feature subset of EEG from a larger feature pool as well as to enhance the parameters of SVM classifier. In order to detect epileptic, SVM with a Radial Basis kernel Function (RBF) was applied. Eventually, the proposed WOA-SVM approach is able to enhance the diagnosis of epilepsy as revealed from the statistical results with accuracy 100% for normal subject data versus epileptic data.

Machine Learning Algorithms and Fault Detection for Improved Belief Function Based Decision Fusion in Wireless Sensor Networks.

Decision fusion is used to fuse classification results and improve the classification accuracy in order to reduce the consumption of energy and bandwidth demand for data transmission. The decentralized classification fusion problem was the reason to use the belief function-based decision fusion approach in Wireless Sensor Networks (WSNs). With the consideration of improving the belief function fusion approach, we have proposed four classification techniques, namely Enhanced K-Nearest Neighbor (EKNN), Enhanced Extreme Learning Machine (EELM), Enhanced Support Vector Machine (ESVM), and Enhanced Recurrent Extreme Learning Machine (ERELM). In addition, WSNs are prone to errors and faults because of their different software, hardware failures, and their deployment in diverse fields. Because of these challenges, efficient fault detection methods must be used to detect faults in a WSN in a timely manner. We have induced four types of faults: offset fault, gain fault, stuck-at fault, and out of bounds fault, and used enhanced classification methods to solve the sensor failure issues. Experimental results show that ERELM gave the first best result for the improvement of the belief function fusion approach. The other three proposed techniques ESVM, EELM, and EKNN provided the second, third, and fourth best results, respectively. The proposed enhanced classifiers are used for fault detection and are evaluated using three performance metrics, i.e., Detection Accuracy (DA), True Positive Rate (TPR), and Error Rate (ER). Simulations show that the proposed methods outperform the existing techniques and give better results for the belief function and fault detection in WSNs.

An enhanced support vector machine model for urban expansion prediction

Given the dynamic and convoluted nature of urban expansion process and the necessity of handling continuous and categorical variables, non-normal distributed data, and non-linear relationships, urban expansion modeling is challenging. To handle these issues effectively and enhance the quality of urban expansion prediction, the capabilities of support vector machine (SVM) technique are explored in this study. A binary SVM model is developed using three different data sampling methods and nineteen predictor variables, four of which are first introduced in this study. The model is configured by regulating the penalty parameter, selecting the most appropriate kernel function, and setting the best value for the kernel function's parameter. A novel combination of goodness-of-fit metrics is used to more realistically evaluate the model accuracy to predict built and unbuilt land cells as well as changed and unchanged land cells in the whole study area. The implementation of the developed model in Guilford County, NC, over the period of 2001–2011, as a case study, demonstrated highly accurate and reliable results. The best performance of the model with the training accuracy of 98% and the testing accuracy of 85% was achieved using a balanced sampling method, fourteen predictor variables, the penalty parameter equal to 1, the radial basis function (RBF) kernel, and the value of 2 for the kernel's parameter. The urban expansion model based on SVM method can substantially improve the prediction accuracy and would be helpful for making appropriate plans and policies to mitigate the adverse impacts of urban expansion.

Parameter optimisation of support vector machine using mutant particle swarm optimisation for diagnosis of metal-oxide surge arrester conditions

ABSTRACTThis paper proposes an enhanced support vector machine (SVM), whose parameters are optimised by a novel mutant particle swarm optimisation (mutant PSO) algorithm to identify metal-oxide surge arrester conditions. The total leakage current and its resistive component under different arrester conditions are obtained and then are inputted into a multilayer SVM for the purpose of fault identification. Then, a mutant PSO-based technique is investigated to increase the classification accuracy as well as the training speed of the SVM classifier. The proposed technique has been tested on an actual data set obtained from Taipower Company to monitor five arrester operating conditions, including normal (N), pre-fault (A), tracking (T), abnormal (U) and degradation (D). Furthermore, to demonstrate the effectiveness of the proposed mutant PSO, the obtained results are compared to those obtained by using cross-validation method, genetic algorithm and particle swarm optimisation.

Efficient time series data classification using sliding window technique based improved association rule mining with enhanced support vector machine

Time series analysis is an important and complex problem in machine learning and statistics. In the existing system, Support Vector Machine (SVM) and Association Rule Mining (ARM) is introduced to implement the time series data. However it has issues with lower accuracy and higher time complexity. Also it has issue with optimal rules discovery and segmentation on time series data. To avoid the above mentioned issues, in the proposed research Sliding Window Technique based Improved ARM with Enhanced SVM (SWT-IARM with ESVM) is proposed. In the proposed system, the preprocessing is performed using Modified K-Means Clustering (MKMC). The indexing process is done by using R-tree which is used to provide faster results. Segmentation is performed by using SWT and it reduces the cost complexity by optimal segments. Then IARM is applied on efficient rule discovery process by generating the most frequent rules. By using ESVM classification approach, the rules are classified more accurately.

A novel hybrid prediction model for aggregated loads of buildings by considering the electric vehicles

In this paper, a new prediction model for aggregated loads of buildings has been propose. Due to high correlation of prediction performance with related horizons and aggregated more customers, a new strategy is developed to provide a forecasting model based on high accuracy. While, consumption shape of a single users normally has low structure to be exploited and vice versa. So, combining different users develops the relative prediction performance up to special point. Outside this point, no more enhancement in relative performance can be found. This model is useful in optimal power system operation and planning in micro-grids. In this paper, besides the aggregated loads of building, the electric vehicle (EV) impact on network has been considered. While, by considering the load growth prediction and impact of EV adaption on load curves this problem can be an important issue for the power grid operation and management. So, an accurate prediction model is presented in this work which is composed of new feature selection and enhanced support vector machine (ESVM) based forecast engine. Effectiveness of the proposed model is carried out to the part of Budapest city through comparing with other prediction models. Obtained results demonstrate the validity of proposed method.

Design and Development of Collaborative Detection and Taxonomy of DDoS Attacks Using ESVM

Distributed Denial of Service (DDoS) assault is a ceaseless basic risk to the web. Application layer DDoS Attack is gotten from the lower layers. Application layer based DDoS assaults utilize honest to goodness HTTP asks for after foundation of TCP three-way handshaking and overpowers the casualty assets, for example, attachments, CPU, memory, circle, database transfer speed. Arrange layer based DDoS assaults sends the SYN, UDP and ICMP solicitations to the server and debilitates the transfer speed. An oddity discovery system is proposed in this paper to identify DDoS assaults utilizing Enhanced Support Vector Machine (ESVM). The Application layer DDoS Attack, for example, HTTP Flooding, DNS Spoofing and Network layer DDoS Attack, for example, Port Scanning, TCP Flooding, UDP Flooding, ICMP Flooding, Land Flooding. Session Flooding is taken as test tests for ESVM. The Normal client gets to conduct characteristics is taken as preparing tests for ESVM. The movement from the testing tests and preparing tests are Cross Validated and the better arrangement exactness is acquired. Application and Network layer DDoS assaults are arranged with order exactness of 99 % with ESVM.

A learning-based hybrid framework for detection and defence of DDoS attacks

Distributed denial of service (DDoS) attacks are those which deplete the valuable resource available for the legitimate user and reduces the business value of any web service provided. This sort of cyber-attacks has to be detected and respective actions have to be taken on them. An integrated detection and defensive mechanism is proposed in this paper to generate and detect DDoS attacks using machine learning algorithms such as back propagation neural network (BPNN), self-organising map (SOM) and enhanced support vector machine (ESVM) and to identify the real IP address of the spoofed attack source using the entropy-based defensive mechanism. The detection and defence mechanism are found to be effective in identifying the attack source with 99% accuracy using ESVM and response time of less than two seconds using the entropy-based tracing scheme. The real source of attacks is filtered using the IP tables to defend the DDoS attacks.