An ANN Model to Classify Multinomial Datasets with Optimized Target Using Particle Swarm Optimization Technique

Dash, Nilamadhab; Priyadarshini, Rojalina; Misra, Rajiv

doi:10.1007/978-81-322-2205-7_34

“…The proposed work incorporates, modification and improvement in the previous work (Dash et al, 2015) where a PSO-BPNN approach is being used. Unlike a PSO-BPNN, which consumes a considerable amount of time in the process of computing input layer output and hidden layer output and iterating the process till both the input weights and hidden weights are not optimised to map to the output, mainly in case of large datasets.…”

Section: Proposed Approachmentioning

confidence: 99%

An improved extreme learning machine to classify multinomial datasets using particle swarm optimisation

Dash

¹

,

Priyadarshini

²

,

Misra

³

2017

IJISDC

Self Cite

0

View full text Add to dashboard Cite

Abstract:In this paper, we propose a particle swarm-based extreme learning machine (ELM) to classify datasets with varying number of classes. This work emphasises on a couple of important parameters, like maximisation of classification accuracy and minimisation of training time. As a machine classifier, an ELM has been chosen, which is an improvement over back propagation network. For each of the input dataset an optimised target was determined by using particle swarm optimisation (PSO) technique. Those specific targets are used with the input data to train the ELM during classification process. For this, some of the bench mark classification datasets are used. To compare the proposed method and some of the existing methods an extensive experimental study has been carried out; a comparative analysis is done by taking parameters like percentage of classification accuracy, training time and complexity of the computing algorithm.Keywords: multinomial classification; extreme learning machine; ELM; normalisation; particle swarm optimisation; PSO; back propagation neural network; classification; classification accuracy; input; target; complexity of algorithm.Reference to this paper should be made as follows: Dash, N., Priyadarshini, R. and Misra, R. (2017) 'An improved extreme learning machine to classify multinomial datasets using particle swarm optimisation', Int. J. Intelligent Systems Design and Computing, Vol. 1, Nos. 1/2, pp.127-144. Rachita Misra received her Doctoral degree from IIT Kharagpur, and is currently working as a Professor in the Department of IT, in C.V. Raman College of Engineering, Bhubaneswar. With more than 15 years of IT industry experience and 15 years of teaching and research experience, she has more than 30 publications in various reputed journals and conferences. Her areas of interest are digital image processing, data mining, soft computing, parallel computing, software engineering and project management. This paper is a revised and expanded version of a paper entitled 'An ANN Model to classify multinomial datasets with optimized target using particle swarm optimization technique' presented at

show abstract

“…Multi-class image classification is typically a difficult and more challenging problem than binary classification; it aims at building a model that discriminates between instances of more than two classes. Some methods, such as k-Nearest Neighbour (k-NN), naturally permit handling the multi-class classification task [63]. However, extending tree-based GP to perform multi-class image classification needs to be handled carefully [184,281,274,168].…”

Section: (Ii) How Can Gp Be Used To Perform Multi-class Image Classification Using Only a Small Number Of Training Instances?mentioning

confidence: 99%

Genetic Programming for Automatically Synthesising Robust Image Descriptors with A Small Number of Instances

Al-Sahaf¹

View full text Add to dashboard Cite

<p>Image classification is a core task in many applications of computer vision, including object detection and recognition. It aims at analysing the visual content and automatically categorising a set of images into different groups. Performing image classification can largely be affected by the features used to perform this task. Extracting features from images is a challenging task due to the large search space size and practical requirements such as domain knowledge and human intervention. Human intervention is usually needed to identify a good set of keypoints (regions of interest), design a set of features to be extracted from those keypoints such as lines and corners, and develop a way to extract those features. Automating these tasks has great potential to dramatically decrease the time and cost, and may potentially improve the performance of the classification task. There are two well-recognised approaches in the literature to automate the processes of identifying keypoints and extracting image features. Designing a set of domain-independent features is the first approach, where the focus is on dividing the image into a number of predefined regions and extracting features from those regions. The second approach is synthesising a function or a set of functions to form an image descriptor that aims at automatically detecting a set of keypoints such as lines and corners, and performing feature extraction. Although employing image descriptors is more effective and very popular in the literature, designing those descriptors is a difficult task that in most cases requires domain-expert intervention. The overall goal of this thesis is to develop a new domain independent Genetic Programming (GP) approach to image classification by utilising GP to evolve programs that are capable of automatically detecting diverse and informative keypoints, designing a set of features, and performing feature extraction using only a small number of training instances to facilitate image classification, and are robust to different image changes such as illumination and rotation. This thesis focuses on incorporating a variety of simple arithmetic operators and first-order statistics (mid-level features) into the evolutionary process and on representation of GP to evolve programs that are robust to image changes for image classification. This thesis proposes methods for domain-independent binary classification in images using GP to automatically identify regions within an image that have the potential to improve classification while considering the limitation of having a small training set. Experimental results show that in over 67% of cases the new methods significantly outperform the use of existing hand-crafted features and features automatically detected by other methods. This thesis proposes the first GP approach for automatically evolving an illumination-invariant dense image descriptor that detects automatically designed keypoints, and performs feature extraction using only a few instances of each class. The experimental results show improvement of 86% on average compared to two GP-based methods, and can significantly outperform domain-expert hand-crafted descriptors in more than 89% of the cases. This thesis also considers rotation variation of images and proposes a method for automatically evolving rotation-invariant image descriptors through integrating a set of first-order statistics as terminals. Compared to hand-crafted descriptors, the experimental results reveal that the proposed method has significantly better performance in more than 83% of the cases. This thesis proposes a new GP representation that allows the system to automatically choose the length of the feature vector side-by-side with evolving an image descriptor. Automatically determining the length of the feature vector helps to reduce the number of the parameters to be set. The results show that this method has evolved descriptors with a very small feature vector which yet still significantly outperform the competitive methods in more than 91% of the cases. This thesis proposes a method for transfer learning by model in GP, where an image descriptor evolved on instances of a related problem (source domain) is applied directly to solve a problem being tackled (target domain). The results show that the new method evolves image descriptors that have better generalisability compared to hand-crafted image descriptors. Those automatically evolved descriptors show positive influence on classifying the target domain datasets in more than 56% of the cases.</p>

show abstract

“…Multi-class image classification is typically a difficult and more challenging problem than binary classification; it aims at building a model that discriminates between instances of more than two classes. Some methods, such as k-Nearest Neighbour (k-NN), naturally permit handling the multi-class classification task [63]. However, extending tree-based GP to perform multi-class image classification needs to be handled carefully [184,281,274,168].…”

Section: (Ii) How Can Gp Be Used To Perform Multi-class Image Classification Using Only a Small Number Of Training Instances?mentioning

confidence: 99%

Genetic Programming for Automatically Synthesising Robust Image Descriptors with A Small Number of Instances

Al-Sahaf¹

0

View full text Add to dashboard Cite

<p>Image classification is a core task in many applications of computer vision, including object detection and recognition. It aims at analysing the visual content and automatically categorising a set of images into different groups. Performing image classification can largely be affected by the features used to perform this task. Extracting features from images is a challenging task due to the large search space size and practical requirements such as domain knowledge and human intervention. Human intervention is usually needed to identify a good set of keypoints (regions of interest), design a set of features to be extracted from those keypoints such as lines and corners, and develop a way to extract those features. Automating these tasks has great potential to dramatically decrease the time and cost, and may potentially improve the performance of the classification task. There are two well-recognised approaches in the literature to automate the processes of identifying keypoints and extracting image features. Designing a set of domain-independent features is the first approach, where the focus is on dividing the image into a number of predefined regions and extracting features from those regions. The second approach is synthesising a function or a set of functions to form an image descriptor that aims at automatically detecting a set of keypoints such as lines and corners, and performing feature extraction. Although employing image descriptors is more effective and very popular in the literature, designing those descriptors is a difficult task that in most cases requires domain-expert intervention. The overall goal of this thesis is to develop a new domain independent Genetic Programming (GP) approach to image classification by utilising GP to evolve programs that are capable of automatically detecting diverse and informative keypoints, designing a set of features, and performing feature extraction using only a small number of training instances to facilitate image classification, and are robust to different image changes such as illumination and rotation. This thesis focuses on incorporating a variety of simple arithmetic operators and first-order statistics (mid-level features) into the evolutionary process and on representation of GP to evolve programs that are robust to image changes for image classification. This thesis proposes methods for domain-independent binary classification in images using GP to automatically identify regions within an image that have the potential to improve classification while considering the limitation of having a small training set. Experimental results show that in over 67% of cases the new methods significantly outperform the use of existing hand-crafted features and features automatically detected by other methods. This thesis proposes the first GP approach for automatically evolving an illumination-invariant dense image descriptor that detects automatically designed keypoints, and performs feature extraction using only a few instances of each class. The experimental results show improvement of 86% on average compared to two GP-based methods, and can significantly outperform domain-expert hand-crafted descriptors in more than 89% of the cases. This thesis also considers rotation variation of images and proposes a method for automatically evolving rotation-invariant image descriptors through integrating a set of first-order statistics as terminals. Compared to hand-crafted descriptors, the experimental results reveal that the proposed method has significantly better performance in more than 83% of the cases. This thesis proposes a new GP representation that allows the system to automatically choose the length of the feature vector side-by-side with evolving an image descriptor. Automatically determining the length of the feature vector helps to reduce the number of the parameters to be set. The results show that this method has evolved descriptors with a very small feature vector which yet still significantly outperform the competitive methods in more than 91% of the cases. This thesis proposes a method for transfer learning by model in GP, where an image descriptor evolved on instances of a related problem (source domain) is applied directly to solve a problem being tackled (target domain). The results show that the new method evolves image descriptors that have better generalisability compared to hand-crafted image descriptors. Those automatically evolved descriptors show positive influence on classifying the target domain datasets in more than 56% of the cases.</p>

show abstract

An ANN Model to Classify Multinomial Datasets with Optimized Target Using Particle Swarm Optimization Technique

Cited by 3 publications

References 6 publications

An improved extreme learning machine to classify multinomial datasets using particle swarm optimisation

An improved extreme learning machine to classify multinomial datasets using particle swarm optimisation

Genetic Programming for Automatically Synthesising Robust Image Descriptors with A Small Number of Instances

Genetic Programming for Automatically Synthesising Robust Image Descriptors with A Small Number of Instances

Contact Info

Product

Resources

About