CIFAR-10: KNN-Based Ensemble of Classifiers

Abouelnaga, Yehya; Ali, Ola S.; Rady, Hager; Moustafa, Mohamed N.

doi:10.1109/csci.2016.0225

Cited by 54 publications

(23 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But according to the best-known results, K-Nearest neighbor can reach 35% accuracy if it's used with the right distance function and right value of K. and there are also other ways such as principal component analysis which could improve its performance furthermore. Moreover, KNN nearest neighbor could be used in combination with convolutional neural networks to increase its accuracy [2].…”

Section: B Experiments and Resultsmentioning

confidence: 99%

“…In this area, Y. Abouelnaga and al tried on their work to work on an assembled model that use several CNN models and combine it with a KNN approach optimized by PCA (principal component analysis), and they have achieved good results on classifying the CIFAR-10 ( Fig. 1) dataset [2]. In the same area, L. H. Thai et al have used SVM together with artificial Neural networks to construct their model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Performance Analysis of Machine Learning Techniques for Smart Agriculture: Comparison of Supervised Classification Approaches

Mouhssine¹,

Abdoun²,

Haimoudi³

2020

IJACSA

View full text Add to dashboard Cite

Agriculture form one of the most important aspects of life necessities, it is responsible to feed 7.7 billion person for the time being, and it is expected to supply more than 9.6 billion individual in 2050, the thing that made classical farming insufficient, and give birth to the notion of smart farming, and the race has begun toward using the latest technologies in the field. They integrate the Internet of Things (IoT), automation, Artificial Intelligence (AI), etc. And as researchers from a country that highly depends on agriculture, we have decided to also contribute to this evolution, and we chose Machine learning (ML) as our entrance to the field to satisfy the need for automated classification of the different products produced by a farm. In this work, we wanted to solve the problem of automatic classification of agricultural products, without the need of any human intervention, and we concentrate on the classification of red fruits, due to our proximity to a location that its product is red fruits. In other words, we are doing a comparative study among the well-known approaches that are used in image classification, and we are applying the best-found method to correctly classify the pictures of red fruits. And this empirically leads us to achieve great results as shown in the numerical result area.

show abstract

Section: B Experiments and Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Machine Learning Techniques for Smart Agriculture: Comparison of Supervised Classification Approaches

Mouhssine¹,

Abdoun²,

Haimoudi³

2020

IJACSA

View full text Add to dashboard Cite

show abstract

“…The prototype obtained 78.90% accuracy in classification of a said dataset on a GPU unit. Abouelnaga et al [33] constructed a collaborative classifier on K-nearest neighbor. They used a combination of KNN and CNN to reduce the overfitting by Principal Component Analysis (PCA).…”

Section: Related Workmentioning

confidence: 99%

An Effective and Improved CNN-ELM Classifier for Handwritten Digits Recognition and Classification

Ali

Pei

et al. 2020

Symmetry

View full text Add to dashboard Cite

Optical character recognition is gaining immense importance in the domain of deep learning. With each passing day, handwritten digits (0–9) data are increasing rapidly, and plenty of research has been conducted thus far. However, there is still a need to develop a robust model that can fetch useful information and investigate self-build handwritten digit data efficiently and effectively. The convolutional neural network (CNN) models incorporating a sigmoid activation function with a large number of derivatives have low efficiency in terms of feature extraction. Here, we designed a novel CNN model integrated with the extreme learning machine (ELM) algorithm. In this model, the sigmoid activation function is upgraded as the rectified linear unit (ReLU) activation function, and the CNN unit along with the ReLU activation function are used as a feature extractor. The ELM unit works as the image classifier, which makes the perfect symmetry for handwritten digit recognition. A deeplearning4j (DL4J) framework-based CNN-ELM model was developed and trained using the Modified National Institute of Standards and Technology (MNIST) database. Validation of the model was performed through self-build handwritten digits and USPS test datasets. Furthermore, we observed the variation of accuracies by adding various hidden layers in the architecture. Results reveal that the CNN-ELM-DL4J approach outperforms the conventional CNN models in terms of accuracy and computational time.

show abstract

“…It provides a good insight on how some networks can even exceed the human accuracy of 93.91%. [2] presents the accuracy achieved by various conventional machine learning algorithms such as Logistic Regression, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and a combination of these algorithms with Principal Component Analysis (PCA). It also aims to study CNNs and has achieved an accuracy of 94.03% using an ensemble ABSTRACT of four CNN classifiers and one KNN classifier along with data augmentation.…”

Section: Related Workmentioning

confidence: 99%

Comparative Study of Various Convolutional Neural Networks on Cifar-10

Goyal¹

2020

IJMTST

View full text Add to dashboard Cite

Image recognition plays a foundational role in the field of computer vision and there has been extensive research to develop state-of-the-art techniques especially using Convolutional Neural Network (CNN). This paper aims to study some CNNs, heavily inspired by highly popular state-of-the-art CNNs, designed from scratch specifically for the Cifar-10 dataset and present a fair comparison between them.

show abstract

CIFAR-10: KNN-Based Ensemble of Classifiers

Cited by 54 publications

References 21 publications

Performance Analysis of Machine Learning Techniques for Smart Agriculture: Comparison of Supervised Classification Approaches

Performance Analysis of Machine Learning Techniques for Smart Agriculture: Comparison of Supervised Classification Approaches

An Effective and Improved CNN-ELM Classifier for Handwritten Digits Recognition and Classification

Comparative Study of Various Convolutional Neural Networks on Cifar-10

Contact Info

Product

Resources

About