scholarly journals Classification of Autistic Spectrum Disorder Using Deep Neural Network With Particle Swarm Optimization

2022 ◽  
Vol 12 (1) ◽  
pp. 1-11
Author(s):  
Sanat Kumar Sahu ◽  
Pratibha Verma

In this paper, Feature Selection Technique (FST) namely Particle Swarm Optimization (PSO) has been used. The filter based PSO is a search method with Correlation-based Feature Selection (CBFS) as a fitness function. The FST has two key goals of improving classification efficiency and reducing feature counts. Artificial Neural Network (ANN) Based Multilayer Perceptron Network (MLP) and Deep Learning (DL) have been considered the classification methods on 2 benchmark Autistic Spectrum Disorder (ASD) dataset. The experimental result was compared to the non-reduced features and reduced feature of ASD datasets. The reduced feature give up enhanced results in both classifiers MLP and DL. In addition, an experimental study on the exhibitions of these methodologies has been conducted. Finally, a new trend of PSO-MLP and PSO-DL based classification model is proposed.

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Bingsheng Chen ◽  
Huijie Chen ◽  
Mengshan Li

Feature selection can classify the data with irrelevant features and improve the accuracy of data classification in pattern classification. At present, back propagation (BP) neural network and particle swarm optimization algorithm can be well combined with feature selection. On this basis, this paper adds interference factors to BP neural network and particle swarm optimization algorithm to improve the accuracy and practicability of feature selection. This paper summarizes the basic methods and requirements for feature selection and combines the benefits of global optimization with the feedback mechanism of BP neural networks to feature based on backpropagation and particle swarm optimization (BP-PSO). Firstly, a chaotic model is introduced to increase the diversity of particles in the initial process of particle swarm optimization, and an adaptive factor is introduced to enhance the global search ability of the algorithm. Then, the number of features is optimized to reduce the number of features on the basis of ensuring the accuracy of feature selection. Finally, different data sets are introduced to test the accuracy of feature selection, and the evaluation mechanisms of encapsulation mode and filtering mode are used to verify the practicability of the model. The results show that the average accuracy of BP-PSO is 8.65% higher than the suboptimal NDFs model in different data sets, and the performance of BP-PSO is 2.31% to 18.62% higher than the benchmark method in all data sets. It shows that BP-PSO can select more distinguishing feature subsets, which verifies the accuracy and practicability of this model.


2021 ◽  
Vol 18 (4) ◽  
pp. 1233-1238
Author(s):  
R. Sathya ◽  
L. R. Aravind Babu

Big data defines the state where the size, speed and kind of data go beyond a memory or executing capabilities for precise and timely decision-making. Big data analytics is integrated with ML and statistical methods for processing big data and recognizes the important data. At present times, the generation of online product reviews has exponentially increased at each and every second. These applications have resulted in developing the volumes of data which can be used for prediction and classification for decision making process. Compared with other models, various techniques are applied in solving the big data problem, feature selection (FS) is known to be an efficient method. FS operations could be exploring with the application of a subset of features which is related to the topic of précised definition of the existing datasets. Deplorably, search using this type of sub-sets results in the problems of combinatorial as well as maximum time consuming. The meta-heuristic approaches are typically employed to facilitate the choice of features. This paper presents an optimal extreme learning machine (ELM) based binary particle swarm optimization to precede the FS process. The proposed method develops a Fitness Function (FF) by applying ELM. And the best solution of the FF has been explored under the application of BPSO technique. For instance, the dataset of product review which are derived from Amazon including synthetic data, which is comprised with total of 235,000 positive and 147,000 negative review records is used. The experimental result implied that the ELM-BPSO technique is comparably best


2021 ◽  
Vol 11 (3) ◽  
pp. 803-809
Author(s):  
J. Jayanthi ◽  
T. Jayasankar ◽  
N. Krishnaraj ◽  
N. B. Prakash ◽  
A. Sagai Francis Britto ◽  
...  

Diabetic retinopathy (DR), a major cause of vision loss and it raises a major issue among diabetes people. DR considerably affect the financial condition of the society specially in medicinal sector. Once proper treatment is given to the DR patients, roughly 90% of patients can be saved from vision loss. So, it is needed to develop a DR classification model for classifying the stages and severity level of DR to offer better treatment. This article develops a novel Particle Swarm Optimization (PSO) algorithm based Convolutional Neural Network (CNN) Model called PSO-CNN model to detect and classify DR from the color fundus images. The proposed PSO-CNN model comprises three stages namely preprocessing, feature extraction and classification. Initially, preprocessing is carried out as a noise removal process to discard the noise present in the input image. Then, feature extraction process using PSO-CNN model is applied to extract the useful subset of features. Finally, the filtered features are given as input to the decision tree (DT) model for classifying the set of DR images. The simulation of the PSO-CNN model takes place using a benchmark DR database and the experimental outcome stated that the PSO-CNN model has outperformed all the compared methods in a significant way. The outcome of the simulation process indicated that the PSO-CNN model has offered maximum results.


2010 ◽  
Vol 20-23 ◽  
pp. 1378-1384 ◽  
Author(s):  
Ru Hui Ma ◽  
Yuan Liu

Neural network (NN) employed to settle network anomaly has become prevalent. However, traditional training algorithm for NN is not optimum, that is, often suboptimum, and encountering complicated network anomaly, an adaptive yet efficient NN or hybrid NN model should be better considered. Therefore, this paper proposes a novel network anomaly detection method employing wavelet fuzzy neural network (WFNN) to use modified Quantum-Behaved Particle Swarm Optimization (QPSO). In this paper, wavelet transform is applied to extract fault characteristics from the anomaly state. Fuzzy theory and neural network are employed to fuzzify the extracted information. Wavelet is then integrated with fuzzy neural network to form the wavelet fuzzy neural network (WFNN). The Quantum-Behaved Particle Swarm Optimization, which outperforms other optimization algorithm considerably on its simple architecture and fast convergence, has previously applied to solve optimum problem. However, the QPSO also has its own shortcomings. So, there exists a modified QPSO which is used to train WFNN in this paper. Experimental result on KDD99 intrusion detection datasets shows that this WFNN using the novel training algorithm has high detection rate while maintaining a low false positive rate.


2018 ◽  
Vol 4 (10) ◽  
pp. 6
Author(s):  
Shivangi Bhargava ◽  
Dr. Shivnath Ghosh

News popularity is the maximum growth of attention given for particular news article. The popularity of online news depends on various factors such as the number of social media, the number of visitor comments, the number of Likes, etc. It is therefore necessary to build an automatic decision support system to predict the popularity of the news as it will help in business intelligence too. The work presented in this study aims to find the best model to predict the popularity of online news using machine learning methods. In this work, the result analysis is performed by applying Co-relation algorithm, particle swarm optimization and principal component analysis. For performance evaluation support vector machine, naïve bayes, k-nearest neighbor and neural network classifiers are used to classify the popular and unpopular data. From the experimental results, it is observed that support vector machine and naïve bayes outperforms better with co-relation algorithm as well as k-NN and neural network outperforms better with particle swarm optimization.


Sign in / Sign up

Export Citation Format

Share Document