scholarly journals Classification of Earthquake-Induced Damage forR/CSlab Column Frames Using Multiclass SVM and Its Combination with MLP Neural Network

2014 ◽  
Vol 2014 ◽  
pp. 1-14 ◽  
Author(s):  
Ali Kia ◽  
Serhan Sensoy
Keyword(s):  
Neural Network ◽  
Seismic Vulnerability ◽  
Gaussian Function ◽  
Damage Index ◽  
Time History ◽  
Nonlinear Time History Analysis ◽  
Support Vector ◽  
Damage Classification ◽  
Ground Acceleration

Nonlinear time history analysis (NTHA) is an important engineering method in order to evaluate the seismic vulnerability of buildings under earthquake loads. However, it is time consuming and requires complex calculations and a high memory machine. In this study, two networks were used for damage classification: multiclass support vector machine (M-SVM) and combination of multilayer perceptron neural network with M-SVM (MM-SVM). In order to collect data, three frames ofR/Cslab column frame buildings with wide beams in slab were considered. For NTHA, twenty different ground motion records were selected and scaled to ten different levels of peak ground acceleration (PGA). Thus, 600 obtained data from the numerical simulations were applied to M-SVM and MM-SVM in order to predict the global damage classification of samples based on park and Ang damage index. Amongst the four different kernel tricks, the Gaussian function was determined as an efficient kernel trick using the maximum total accuracy method of test data. By comparing the obtained results from M-SVM and MM-SVM, the total classification accuracy of MM-SVM is more than M-SVM and it is accurate and reliable for global damage classification ofR/Cslab column frames. Furthermore, the proposed combined model is able to classify the classes with low members.

Convolutional neural network - Support vector machine based approach for classification of cyanobacteria and chlorophyta microalgae groups

2021 ◽  
pp. 102568
Author(s):  
Mesut Ersin Sonmez ◽  
Numan Eczacıoglu ◽  
Numan Emre Gumuş ◽  
Muhammet Fatih Aslan ◽  
Kadir Sabanci ◽  
...  
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Convolutional Neural Network ◽  
Support Vector ◽  
Network Support

scholarly journals Performance of an Existing Reinforced Concrete Building Designed in Accordance to Older Indonesian Seismic Code: A Case Study for a Hotel in Kupang, Indonesia

2018 ◽  
Vol 20 (1) ◽  
pp. 35
Author(s):  
Pamuda Pudjisuryadi ◽  
Benjamin Lumantarna ◽  
Ryan Setiawan ◽  
Christian Handoko
Keyword(s):  
Plastic Hinge ◽  
Time History ◽  
Nonlinear Time History Analysis ◽  
Seismic Load ◽  
Ground Acceleration ◽  
Seismic Code ◽  
Damage Level ◽  
History Analysis ◽  
Nonlinear Time History ◽  
Seismic Hazard Map

The recent seismic code SNI 1726-2012 is significantly different compared to the older code SNI 1726-2002. The seismic hazard map was significantly changed and the level of maximum considered earthquake was significantly increased. Therefore, buildings designed according to outdated code may not resist the higher demand required by newer code. In this study, seismic performance of Hotel X in Kupang, Indonesia which was designed based on SNI-1726-2002 is investigated. The structure was analyzed using Nonlinear Time History Analysis. The seismic load used was a spectrum consistent ground acceleration generated from El-Centro 18 May 1940 North-South component in accordance to SNI 1726-2012. The results show that Hotel X can resist maximum considered earthquake required by SNI 1726-2012. The maximum drift ratio is 0.81% which is lower than the limit set by FEMA 356-2000 (2%). Plastic hinge damage level is also lower than the allowance in ACMC 2001.

scholarly journals Malware Classification with Improved Convolutional Neural Network Model

2020 ◽  
Vol 12 (6) ◽  
pp. 30-43
Author(s):  
Sumit S. Lad ◽  
◽  
Amol C. Adamuthe
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Personal Information ◽  
Support Vector ◽  
Svm Classifier ◽  
Gray Scale ◽  
Malware Classification

Malware is a threat to people in the cyber world. It steals personal information and harms computer systems. Various developers and information security specialists around the globe continuously work on strategies for detecting malware. From the last few years, machine learning has been investigated by many researchers for malware classification. The existing solutions require more computing resources and are not efficient for datasets with large numbers of samples. Using existing feature extractors for extracting features of images consumes more resources. This paper presents a Convolutional Neural Network model with pre-processing and augmentation techniques for the classification of malware gray-scale images. An investigation is conducted on the Malimg dataset, which contains 9339 gray-scale images. The dataset created from binaries of malware belongs to 25 different families. To create a precise approach and considering the success of deep learning techniques for the classification of raising the volume of newly created malware, we proposed CNN and Hybrid CNN+SVM model. The CNN is used as an automatic feature extractor that uses less resource and time as compared to the existing methods. Proposed CNN model shows (98.03%) accuracy which is better than other existing CNN models namely VGG16 (96.96%), ResNet50 (97.11%) InceptionV3 (97.22%), Xception (97.56%). The execution time of the proposed CNN model is significantly reduced than other existing CNN models. The proposed CNN model is hybridized with a support vector machine. Instead of using Softmax as activation function, SVM performs the task of classifying the malware based on features extracted by the CNN model. The proposed fine-tuned model of CNN produces a well-selected features vector of 256 Neurons with the FC layer, which is input to SVM. Linear SVC kernel transforms the binary SVM classifier into multi-class SVM, which classifies the malware samples using the one-against-one method and delivers the accuracy of 99.59%.

scholarly journals SPECTRAL-SPATIAL CLASSIFICATION OF HYPERSPECTRAL IMAGERY USING NEURAL NETWORK ALGORITHM AND HIERARCHICAL SEGMENTATION

2019 ◽  
Vol XLII-2/W12 ◽  
pp. 1-5
Author(s):  
D. Akbari ◽  
M. Moradizadeh ◽  
M. Akbari
Keyword(s):  
Neural Network ◽  
Spatial Information ◽  
Hyperspectral Data ◽  
Support Vector ◽  
Washington Dc ◽  
Hierarchical Segmentation ◽  
Vector Machines ◽  
Neural Network Algorithm ◽  
New Framework

<p><strong>Abstract.</strong> This paper describes a new framework for classification of hyperspectral images, based on both spectral and spatial information. The spatial information is obtained by an enhanced Marker-based Hierarchical Segmentation (MHS) algorithm. The hyperspectral data is first fed into the Multi-Layer Perceptron (MLP) neural network classification algorithm. Then, the MHS algorithm is applied in order to increase the accuracy of less-accurately classified land-cover types. In the proposed approach, the markers are extracted from the classification maps obtained by MLP and Support Vector Machines (SVM) classifiers. Experimental results on Washington DC Mall hyperspectral dataset, demonstrate the superiority of proposed approach compared to the MLP and the original MHS algorithms.</p>

scholarly journals A Novel Method for Sea-Land Clutter Separation Using Regularized Randomized and Kernel Ridge Neural Networks

Sensors ◽  
2020 ◽  
Vol 20 (22) ◽  
pp. 6491
Author(s):  
Le Zhang ◽  
Jeyan Thiyagalingam ◽  
Anke Xue ◽  
Shuwen Xu
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Classification Accuracy ◽  
Signal Amplitude ◽  
Statistical Characteristics ◽  
Support Vector ◽  
Amplitude Change ◽  
Novel Method

Classification of clutter, especially in the context of shore based radars, plays a crucial role in several applications. However, the task of distinguishing and classifying the sea clutter from land clutter has been historically performed using clutter models and/or coastal maps. In this paper, we propose two machine learning, particularly neural network, based approaches for sea-land clutter separation, namely the regularized randomized neural network (RRNN) and the kernel ridge regression neural network (KRR). We use a number of features, such as energy variation, discrete signal amplitude change frequency, autocorrelation performance, and other statistical characteristics of the respective clutter distributions, to improve the performance of the classification. Our evaluation based on a unique mixed dataset, which is comprised of partially synthetic clutter data for land and real clutter data from sea, offers improved classification accuracy. More specifically, the RRNN and KRR methods offer 98.50% and 98.75% accuracy, outperforming the conventional support vector machine and extreme learning based solutions.

scholarly journals Research on Sentiment Classification Algorithms on Online Review

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-6
Author(s):  
Ruixia Yan ◽  
Zhijie Xia ◽  
Yanxi Xie ◽  
Xiaoli Wang ◽  
Zukang Song
Keyword(s):  
Neural Network ◽  
Sentiment Classification ◽  
Online Review ◽  
Support Vector ◽  
Classification Algorithms ◽  
Evaluation Indexes ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Reinforcement Learning Models

The product online review text contains a large number of opinions and emotions. In order to identify the public’s emotional and tendentious information, we present reinforcement learning models in which sentiment classification algorithms of product online review corpus are discussed in this paper. In order to explore the classification effect of different sentiment classification algorithms, we conducted a research on Naive Bayesian algorithm, support vector machine algorithm, and neural network algorithm and carried out some comparison using a concrete example. The evaluation indexes and the three algorithms are compared in different lengths of sentence and word vector dimensions. The results present that neural network algorithm is effective in the sentiment classification of product online review corpus.

Analysis of HCV’s Linearity Using Support Vector Machine (SVM)

2014 ◽  
Vol 548-549 ◽  
pp. 1265-1269
Author(s):  
Yun Sik Hwang ◽  
Byeong Joo Jun ◽  
Tae Seon Yoon
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Hepatitis C Virus ◽  
Data Analysis ◽  
Hepatitis C ◽  
Amino Acid ◽  
Support Vector ◽  
Svm Algorithm ◽  
Genetic Patterns

As the stage of bioinformatics has been upgraded, classification of certain pathogen has been improved into a new manner. The main topic of this research is genetic singularity of HCV (Hepatitis C Virus) and our objective is to assay features of the HCV's amino acid under usage of Support Vector Machine (SVM) algorithm. HCV data used in our experiment has 10 kinds of sequences and 257 kinds of data. According to data analysis, some peculiar genetic patterns of HCV’s linearity that discord pre-existing neural network and C5.0 were found.

Investigation of Damages in RC Frames Under Near Field Earthquakes Using a Damage Index

2016 ◽  
Vol 16 (02) ◽  
pp. 1450094 ◽  
Author(s):  
Seyed Morteza Zinati Yazdi ◽  
Mohammad Taghi Kazemi
Keyword(s):  
Near Field ◽  
Damage Index ◽  
Time History ◽  
General Rule ◽  
Nonlinear Time History Analysis ◽  
Far Field ◽  
Moment Frames ◽  
Moment Frame ◽  
Rc Frames ◽  
Nonlinear Time History

Heavy damages on structures caused by near field earthquakes in recent years has brought serious attention to this problem. An examination of previous records has shown significant differences for near field earthquakes, including a large energy pulse, unlike far field earthquakes. But as a general rule, the effects of near field earthquakes have been ignored in most building codes. The purpose of this paper is to investigate the effect of near field earthquakes on reinforced concrete (RC) moment frames. To achieve this goal, the Erduran damage index, an efficient way to calculate damage, was employed to analyze two 4- and 8-story RC moment frame buildings. The buildings with moderate and high ductility were designed by the strength criteria. Seven pairs of near field and far field earthquakes were scaled and used for dynamic nonlinear time history analysis. Using Erduran’s beam and column damage index, respectively, based on rotation and drift, the results from both near and far field earthquakes were compared. Moreover, for better assessment, 4-story buildings were evaluated from the performance based viewpoint of design. We observe from the results that most of the components of the structures under near field earthquakes sustained severe damages and in some cases even component failure. Components of the structures under near field earthquakes suffered from 30% more of damage, on average, than that under far field earthquakes.

Damage Classification of Composites Using Machine Learning

2019 ◽  
Author(s):  
Shweta Dabetwar ◽  
Stephen Ekwaro-Osire ◽  
João Paulo Dias
Keyword(s):  
Machine Learning ◽  
Composite Materials ◽  
Random Forest ◽  
Condition Monitoring ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Damage Classification ◽  
Combining Data ◽  
Ultrasonic Measurements

Abstract Composite materials have tremendous and ever-increasing applications in complex engineering systems; thus, it is important to develop non-destructive and efficient condition monitoring methods to improve damage prediction, thereby avoiding catastrophic failures and reducing standby time. Nondestructive condition monitoring techniques when combined with machine learning applications can contribute towards the stated improvements. Thus, the research question taken into consideration for this paper is “Can machine learning techniques provide efficient damage classification of composite materials to improve condition monitoring using features extracted from acousto-ultrasonic measurements?” In order to answer this question, acoustic-ultrasonic signals in Carbon Fiber Reinforced Polymer (CFRP) composites for distinct damage levels were taken from NASA Ames prognostics data repository. Statistical condition indicators of the signals were used as features to train and test four traditional machine learning algorithms such as K-nearest neighbors, support vector machine, Decision Tree and Random Forest, and their performance was compared and discussed. Results showed higher accuracy for Random Forest with a strong dependency on the feature extraction/selection techniques employed. By combining data analysis from acoustic-ultrasonic measurements in composite materials with machine learning tools, this work contributes to the development of intelligent damage classification algorithms that can be applied to advanced online diagnostics and health management strategies of composite materials, operating under more complex working conditions.

Sign in / Sign up

Export Citation Format

Share Document