Wyniki wyszukiwania - BazTech

1

Detection of various lung diseases including COVID-19 using extreme learning machine algorithm based on the features extracted from a lightweight CNN architecture

Nahiduzzaman Md., Faruq Goni Md Omaer, Robiul Islam Md., Sayeed Abu, Shamim Anower Md., Ahsan Mominul, Haider Julfikar, Kowalski Marcin

Biocybernetics and Biomedical Engineering

|

2023

|

Vol. 43, no. 3

528--550

EN

Around the world, several lung diseases such as pneumonia, cardiomegaly, and tuberculosis (TB) contribute to severe illness, hospitalization or even death, particularly for elderly and medically vulnerable patients. In the last few decades, several new types of lungrelated diseases have taken the lives of millions of people, and COVID-19 has taken almost 6.27 million lives. To fight against lung diseases, timely and correct diagnosis with appropriate treatment is crucial in the current COVID-19 pandemic. In this study, an intelligent recognition system for seven lung diseases has been proposed based on machine learning (ML) techniques to aid the medical experts. Chest X-ray (CXR) images of lung diseases were collected from several publicly available databases. A lightweight convolutional neural network (CNN) has been used to extract characteristic features from the raw pixel values of the CXR images. The best feature subset has been identified using the Pearson Correlation Coefficient (PCC). Finally, the extreme learning machine (ELM) has been used to perform the classification task to assist faster learning and reduced computational complexity. The proposed CNN-PCC-ELM model achieved an accuracy of 96.22% with an Area Under Curve (AUC) of 99.48% for eight class classification. The outcomes from the proposed model demonstrated better performance than the existing state-of-the-art (SOTA) models in the case of COVID-19, pneumonia, and tuberculosis detection in both binary and multiclass classifications. For eight class classification, the proposed model achieved precision, recall and fi-score and ROC are 100%, 99%, 100% and 99.99% respectively for COVID-19 detection demonstrating its robustness. Therefore, the proposed model has overshadowed the existing pioneering models to accurately differentiate COVID-19 from the other lung diseases that can assist the medical physicians in treating the patient effectively.

2

Colour Difference Classification for Dyed Fabrics Based on Differential Evolution with Dynamic Parameter Selection to Optimise the Output Regularisation Extreme Learning Machine

Zhou Zhiyu, Liu Dexin, Zhang Jianxin, Zhu Zefei, Yang Donghe, Jiang Likai

Fibres & Textiles in Eastern Europe

|

2021

|

Vol. 29, no. 3 (147)

97--102

EN

A novel optimisation technique based on the differential evolution (DE) algorithm with dynamic parameter selection (DPS-DE) is proposed to develop a colour difference classification model for dyed fabrics, improve the classification accuracy, and optimise the output regularisation extreme learning machine (RELM). The technique proposed is known as DPS-DE-RELM and has three major differences compared with DE-ELM: (1) Considering that the traditional ELM provides an illness solution based on the output weights, DE is proposed to optimise the output of the RELM. (2) Considering the simple parameter setting of the traditional algorithm, the DE algorithm with DPS is adopted. (3) For DPS, an optimal range of parameters is chosen, and the efficiency of the algorithm is significantly improved. This study analyses the colour difference classification of fabric images captured under standard lighting based on the DPS-DE-RELM algorithm. First, the colour difference of the fabric images is calculated and six color-difference-related features extracted, and second the features are classified into five different levels based on the perception of humans. Finally, a colour difference classification model is built based on the DPS-DERELM algorithm, and then the optimal classification model suitable for this study is selected. The experimental results show that the output method with regularisation parameters can achieve a maximum classification accuracy of 98.87%, which is higher compared with the aforementioned optimised original ELM algorithm, which can achieve a maximum accuracy of 84.67%. Therefore, the method proposed has the advantages of greater convergence speed, high classification accuracy, and robustness.

PL

W pracy zaproponowano nowatorską technikę optymalizacji opartą na algorytmie ewolucji różnicowej (DE) z doborem parametrów (DPS-DE) w celu opracowania modelu klasyfikacji różnicy kolorów dla tkanin barwionych, poprawy dokładności klasyfikacji i optymalizacji regularyzacji wyjściowej maszyny do uczącej się (RELM). Zaproponowana technika jest znana jako DPS-DE-RELM i cechuje się trzema głównymi różnicami w porównaniu do DE-ELM: (1) Biorąc pod uwagę, że tradycyjny ELM zapewnia rozwiązanie w oparciu o wagi wyjściowe, proponuje się DE w celu optymalizacji wydajności RELM. (2) Biorąc pod uwagę proste ustawienie parametrów tradycyjnego algorytmu, przyjęto algorytm DE z DPS. (3) W przypadku DPS wybierany jest optymalny zakres parametrów, a wydajność algorytmu znacznie się poprawia. Podczas badania przeanalizowano klasyfikację różnic kolorów obrazów tkanin zarejestrowanych w standardowym oświetleniu w oparciu o algorytm DPS-DE-RELM. Po pierwsze, obliczono różnicę kolorów obrazów tkanin i wyodrębniono sześć cech związanych z różnicą kolorów, a po drugie cechy te zaklasyfikowano na pięciu różnych poziomach w oparciu o percepcję ludzi. Na koniec zbudowano model klasyfikacji różnicy kolorów w oparciu o algorytm DPS-DE-RELM, a następnie wybrano optymalny model klasyfikacji odpowiedni do tego badania. Wyniki eksperymentalne pokazały, że metoda wyjściowa z parametrami regularyzacji może osiągnąć maksymalną dokładność klasyfikacji wynoszącą 98,87%, czyli wyższą w porównaniu z zoptymalizowanym oryginalnym algorytmem ELM, który może osiągnąć maksymalną dokładność na poziomie 84,67%. Stwierdzono, że zaproponowana metoda niesie ze sobą korzyści w postaci większej szybkości zbieżności, wysokiej dokładności klasyfikacji i odporności.

3

Clothing Image Classification with a Dragonfly Algorithm Optimised Online Sequential Extreme Learning Machine

Li Jianqiang, Shi Weimin, Yang Donghe

Fibres & Textiles in Eastern Europe

|

2021

|

Vol. 29, no. 3 (147)

91--96

EN

This study proposes a solution for the issue of the low classification accuracy of clothing images. Using Fashion-MNIST as the clothing image dataset, we propose a clothing image classification technology based on an online sequential extreme learning machine (OSELM) optimised by the dragonfly algorithm (DA). First, we transform the Fashion-MNIST dataset into a data set that we extract from the corresponding grey image. Then, considering that the input weight and hidden layer bias of an OSELM are generated randomly, a DA is proposed to optimise the input weight and hidden layer bias of the OSELM to reduce the influence of random generation on the classification results. Finally, the optimised OSELM is applied to the clothing image classification. Compared to the other seven types of classification algorithms, the proposed clothing image classification model with the DA-optimised OSELM reached 93.98% accuracy when it contained 350 hidden nodes. Its performance was superior to other algorithms that were configured with the same number of hidden nodes. From a stability analysis of the box-plot, it was found that there were no outliers exhibited by the DA-OSELM model, whereas some other models had outliers or had lower stability compared to the model proposed, thereby validating the efficacy of the solution proposed.

PL

W pracy zaproponowano rozwiązanie problemu niskiej dokładności klasyfikacyjnej obrazów odzieży. Wykorzystując Fashion-MNIST jako zbiór danych obrazu odzieży, zaproponowano technologię klasyfikacji obrazów odzieży w oparciu o sekwencyjną maszynę uczącą się (OSELM) zoptymalizowaną przez algorytm Dragonfly (DA). Najpierw przekształcono zbiór danych Fashion-MNIST w zestaw danych, który wyodrębniono z obrazu. Następnie, biorąc pod uwagę, że waga wejściowa i odchylenie warstwy ukrytej OSELM były generowane losowo, w celu zmniejszenia wpływu generowania losowego na wyniki klasyfikacji zaproponowano DA w celu optymalizacji wagi wejściowej i obciążenia warstwy ukrytej OSELM. Następnie, zoptymalizowany OSELM zastosowano do klasyfikacji obrazu odzieży. W porównaniu z pozostałymi siedmioma typami algorytmów klasyfikacji, proponowany model klasyfikacji obrazu odzieży ze zoptymalizowanym przez DA OSELM osiągnął dokładność 93,98%. Jego wydajność przewyższyła inne algorytmy. Na podstawie analizy stabilności wykresu stwierdzono, że nie było wartości odstających wykazywanych przez model DA-OSELM, podczas gdy niektóre inne modele miały wartości odstające lub miały niższą stabilność w porównaniu z proponowanym modelem, potwierdzono w ten sposób skuteczność proponowanego rozwiązania.

4

Fabric Defect Detection and Classifier via Multi-Scale Dictionary Learning and an Adaptive Differential Evolution Optimized Regularization Extreme Learning Machine

Zhou Zhiyu, Wang Chao, Gao Xu, Zhu Zefei, Hu Xudong, Zheng Xiao, Jiang Likai

Fibres & Textiles in Eastern Europe

|

2019

|

Vol. 27, no. 1 (133)

67--77

EN

To develop an automatic detection and classifier model for fabric defects, a novel detection and classifier technique based on multi-scale dictionary learning and the adaptive differential evolution algorithm optimised regularisation extreme learning machine (ADE-RELM) is proposed. Firstly in order to speed up dictionary updating under the condition of guaranteeing dictionary sparseness, k-means singular value decomposition (KSVD) dictionary learning is used. Then multi-scale KSVD dictionary learning is presented to extract texture features of textile images more accurately. Finally a unique ADE-RELM is designed to build a defect classifier model. In the training ADE-RELM classifier stage, a self-adaptive mutation operator is used to solve the parameter setting problem of the original differential evolution algorithm, then the adaptive differential evolution algorithm is utilised to calculate the optimal input weights and hidden bias of RELM. The method proposed is committed to detecting common defects like broken warp, broken weft, oil, and the declining warp of grey-level and pure colour fabrics. Experimental results show that compared with the traditional Gabor filter method, morphological operation and local binary pattern, the method proposed in this paper can locate defects precisely and achieve high detection efficiency.

PL

W celu opracowania automatycznego modelu wykrywania i klasyfikowania defektów tkanin, zaproponowano nowatorską technikę wykrywania i klasyfikowania opartą na zastosowaniu maszyny uczącej się (ADE-RELM). Proponowana metoda ma na celu wykrywanie powszechnych defektów, takich jak przerwana osnowa i wątek oraz zabrudzenia po oleju. Wyniki eksperymentalne pokazują, że w porównaniu z tradycyjną metodą filtrów Gabora, operacją morfologiczną i lokalnym wzorcem binarnym, proponowana w artykule metoda pozwala na precyzyjne zlokalizowanie defektów i osiąga wysoką skuteczność ich wykrywania.

5

Nonparametric statistical analysis for multiple comparison of machine learning regression algorithms

Trawiński B., Smętek M., Telec Z., Lasota T.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 4

867-881

EN

In the paper we present some guidelines for the application of nonparametric statistical tests and post-hoc procedures devised to perform multiple comparisons of machine learning algorithms. We emphasize that it is necessary to distinguish between pairwise and multiple comparison tests. We show that the pairwise Wilcoxon test, when employed to multiple comparisons, will lead to overoptimistic conclusions. We carry out intensive normality examination employing ten different tests showing that the output of machine learning algorithms for regression problems does not satisfy normality requirements. We conduct experiments on nonparametric statistical tests and post-hoc procedures designed for multiple 1 x N and N x N comparisons with six different neural regression algorithms over 29 benchmark regression data sets. Our investigation proves the usefulness and strength of multiple comparison statistical procedures to analyse and select machine learning algorithms.

6

Multi-label classification using error correcting output codes

Kajdanowicz T., Kazienko P.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 4

829-840

EN

A framework for multi-label classification extended by Error Correcting Output Codes (ECOCs) is introduced and empirically examined in the article. The solution assumes the base multi-label classifiers to be a noisy channel and applies ECOCs in order to recover the classification errors made by individual classifiers. The framework was examined through exhaustive studies over combinations of three distinct classification algorithms and four ECOC methods employed in the multi-label classification problem. The experimental results revealed that (i) the Bode-Chaudhuri-Hocquenghem (BCH) code matched with any multi-label classifier results in better classification quality; (ii) the accuracy of the binary relevance classification method strongly depends on the coding scheme; (iii) the label power-set and the RAkEL classifier consume the same time for computation irrespective of the coding utilized; (iv) in general, they are not suitable for ECOCs because they are not capable to benefit from ECOC correcting abilities; (v) the all-pairs code combined with binary relevance is not suitable for datasets with larger label sets.

7

Graph-based generation of a meta-learning search space

Jankowski N.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 3

647-667

EN

Meta-learning is becoming more and more important in current and future research concentrated around broadly defined data mining or computational intelligence. It can solve problems that cannot be solved by any single, specialized algorithm. The overall characteristic of each meta-learning algorithm mainly depends on two elements: the learning machine space and the supervisory procedure. The former restricts the space of all possible learning machines to a subspace to be browsed by a meta-learning algorithm. The latter determines the order of selected learning machines with a module responsible for machine complexity evaluation, organizes tests and performs analysis of results. In this article we present a framework for meta-learning search that can be seen as a method of sophisticated description and evaluation of functional search spaces of learning machine configurations used in meta-learning. Machine spaces will be defined by specially defined graphs where vertices are specialized machine configuration generators. By using such graphs the learning machine space may be modeled in a much more flexible way, depending on the characteristics of the problem considered and a priori knowledge. The presented method of search space description is used together with an advanced algorithm which orders test tasks according to their complexities.

8

Data-driven models for fault detection using kernel PCA: A water distribution system case study

Nowicki A., Grochowski M., Duzinkiewicz K.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 4

939-949

EN

Kernel Principal Component Analysis (KPCA), an example of machine learning, can be considered a non-linear extension of the PCA method. While various applications of KPCA are known, this paper explores the possibility to use it for building a data-driven model of a non-linear system-the water distribution system of the Chojnice town (Poland). This model is utilised for fault detection with the emphasis on water leakage detection. A systematic description of the system's framework is followed by evaluation of its performance. Simulations prove that the presented approach is both flexible and efficient.

9

A rainfall forecasting method using machine learning models and its application to the Fukuoka city case

Sumi S. M., Zaman M. F., Hirose H.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 4

841-854

EN

In the present article, an attempt is made to derive optimal data-driven machine learning methods for forecasting an average daily and monthly rainfall of the Fukuoka city in Japan. This comparative study is conducted concentrating on three aspects: modelling inputs, modelling methods and pre-processing techniques. A comparison between linear correlation analysis and average mutual information is made to find an optimal input technique. For the modelling of the rainfall, a novel hybrid multi-model method is proposed and compared with its constituent models. The models include the artificial neural network, multivariate adaptive regression splines, the k-nearest neighbour, and radial basis support vector regression. Each of these methods is applied to model the daily and monthly rainfall, coupled with a pre-processing technique including moving average and principal component analysis. In the first stage of the hybrid method, sub-models from each of the above methods are constructed with different parameter settings. In the second stage, the sub-models are ranked with a variable selection technique and the higher ranked models are selected based on the leave-one-out cross-validation error. The forecasting of the hybrid model is performed by the weighted combination of the finally selected models.