Weather forecasting based on hybrid neural model

Saba, Tanzila; Rehman, Amjad; AlGhamdi, Jarallah

doi:10.1007/s13201-017-0538-0

Cited by 52 publications

(27 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to challenges of the complexity of the blood cell, several studies deal with segmentation and morphological of the blood cells (Rehman et al, ). To increase the performance of segmentation and classification of the malaria parasite, the author use different types of features such as gray level co‐occurrence matrix based texture features and intensity (Ebrahim, Kolivand, Rehman, Rahim, & Saba, ; Fahad, Ghani Khan, Saba, Rehman, & Iqbal, ; Khan, Akram, et al, ; Khan, Lali, et al, ; Saba, ), which are propagating to the artificial neural network for malaria classification (Fadhil, Alkawaz, Rehman, & Saba, ; Iqbal, Ghani, Saba, & Rehman, ; Iqbal, Khan, Saba, & Rehman, ; Tahir et al, ). For segmentation, Rao's method and bounding box are utilized whereas for classification, back propagation neural network using several textures and shape features.…”

Section: Related Studiesmentioning

confidence: 99%

Plasmodium species aware based quantification of malaria parasitemia in light microscopy thin blood smear

Abbas

Saba

Rehman

et al. 2019

Microscopy Res & Technique

Self Cite

View full text Add to dashboard Cite

Malaria is a serious worldwide disease, caused by a bite of a female Anopheles mosquito. The parasite transferred into complex life round in which it is grown and reproduces into the human body. The detection and recognition of Plasmodium species are possible and efficient through a process called staining (Giemsa). The staining process slightly colorizes the red blood cells (RBCs) but highlights Plasmodium parasites, white blood cells and artifacts. Giemsa stains nuclei, chromatin in blue tone and RBCs in pink color. It has been reported in numerous studies that manual microscopy is not a trustworthy screening technique when performed by nonexperts. Malaria parasites host in RBCs when it enters the bloodstream. This paper presents segmentation of Plasmodium parasite from the thin blood smear points on region growing and dynamic convolution based filtering algorithm. After segmentation, malaria parasite classified into four Plasmodium species: Plasmodium falciparum, Plasmodium ovale, Plasmodium vivax, and Plasmodium malaria. The random forest and K-nearest neighbor are used for classification base on local binary pattern and hue saturation value features. The sensitivity for malaria parasitemia (MP) is 96.75% on training and testing of the proposed approach while specificity is 94.59%. Beside these, the comparisons of the two features are added to the proposed work for classification having sensitivity is 83.60% while having specificity is 94.90% through random forest classifier based on local binary pattern feature. K E Y W O R D S KNN and random forest, LBP-HSV features, Plasmodium parasite, region growing and dynamic convolution

show abstract

Section: Related Studiesmentioning

confidence: 99%

Plasmodium species aware based quantification of malaria parasitemia in light microscopy thin blood smear

Abbas

Saba

Rehman

et al. 2019

Microscopy Res & Technique

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly, the four remaining runs are carried out and compare the performance results of proposed and existing methods. The comparison results prove that the feature selection accuracy is considerably improved using PBCFS technique by 25% when compared to the existing hybrid neural model [11]. In addition, the PBCFS technique increases the feature selection accuracy by 44% when compared to SVR [12].…”

Section: Impact Feature Selection Accuracymentioning

confidence: 79%

“…Feature selection accuracy is defined as the number of features that are more relevant for weather prediction is selected correctly to the total number of features. The feature selection accuracy is mathematically calculated as follows, [11] and SVR [12]. As shown in figure 5, the numbers of features are taken as input varied from 4 to 20 for computing the feature selection accuracy.…”

Section: Impact Feature Selection Accuracymentioning

confidence: 99%

“…Let us consider 4 features in the first run. The feature selection accuracy of LPBC technique is 75% whereas the feature selection accuracy of the hybrid neural model [11] and SVR [12] are 50% and 50% respectively. Similarly, the four remaining runs are carried out and compare the performance results of proposed and existing methods.…”

Section: Impact Feature Selection Accuracymentioning

confidence: 99%

See 1 more Smart Citation

Point Biserial Correlated Feature Selection of Weather Data

S.B,

Balan

2019

IJEAT

View full text Add to dashboard Cite

Big data is said to be huge amount of data. These data’s may be either structured or unstructured data. It is used for performing prediction in many fields and one among them is weather forecasting. Many feature selection techniques has been introduced but all these techniques failed to get accurate result. In order to improve weather prediction with less complexity, a Point Biserial Correlated Feature Selection (PBCFS) technique is introduced. The big weather dataset comprises the ‘n’ numbers of features. Initially, the PBCFS technique uses a point biserial correlation coefficient to determine relevant feature or irrelevant features among the several features. These relevant features which is selected with the help of this feature selection method can be used for clustering, classification or any other method to perform prediction. The polytomous (i.e. different classes) regression function analyzes the input data with the selected features to provide the significant results as output. Experimental evaluation of proposed PBCFS technique and existing methods are carried out using a big weather dataset. The result shows that we get the output with high feature selection accuracy.

show abstract

“…Among these methods, fuzzy models of Takagi-Sugeno-Kang (TSK) as an adaptive neuro-fuzzy inference system (ANFIS) are known (Manu and Thalla 2017). To increase the speed and performance of ANFIS models, using optimization algorithms (Hosseini et al 2016;Gholami et al 2017b;Bonakdari and Zaji 2018;Karaboga and Kaya 2018) and evolutionary models (multi-objective optimization) (Dariane and Azimi 2016; Ahmadianfar et al 2017;Saba et al 2017;Karkevandi-Talkhooncheh et al 2017;Nouiri 2017) has been very common. Regarding the use of AI methods in prediction of the stable channel geometry dimensions (width, depth and slope), it can be noted to Madvar et al (2011), Taher-Shamsi et al (2013, Bonakdari and Gholami (2016), Gholami et al (2017a), Shaghaghi et al (2017Shaghaghi et al ( , 2018a.…”

Section: Introductionmentioning

confidence: 99%

Analyzing bank profile shape of alluvial stable channels using robust optimization and evolutionary ANFIS methods

et al. 2019

View full text Add to dashboard Cite

In natural rivers and artificial channels in addition to the channel dimensions (widening, reduction in slope and depth in the channel banks), formed shape profile in the case that the sediment on the banks with no movement (thresholds state) is of considerable importance for engineers. To determine the bank shape profiles, various theoretical, empirical and statistical relations have been provided based on physical and numerical models by different researchers. In this study, a simple model of adaptive neuro-fuzzy inference systems (ANFIS) is combined with two algorithms of differential evolution (DE) and singular value decomposition value (SVD) and the performance of these models to predict the stable shape profiles of the channels is evaluated and compared. In this paper, the main goal is to assess extensively the effect of hybrid models based on optimized algorithms (ANFIS-DE) and multi-objective evolutionary algorithm (ANFIS-DE/SVD) in improvement of performance of ANFIS and ANFIS-DE models, respectively. Accordingly, the results assessment show that all three ANFIS, ANFIS-DE and ANFIS-DE/SVD models are perfectly able to predict shape profiles in accordance with the observed profiles for the threshold channels. Using optimized and evolutionary algorithms has a positive impact on the performance of simple model of ANFIS. As compared to the simple ANFIS model, ANFIS-DE approximately 10.1% and ANFIS-DE/SVD model 7.2% is improved compared to the ANFIS-DE model. The accuracy of ANFIS-DE/SVD model showed better performance as well about 18.6% compared to the simple ANFIS model. Therefore, it can be said that not only DE optimization algorithms have a significant impact on increasing the performance of a simple ANFIS model but also using evolutionary algorithms (ANFIS-DE/SVD) reduce the ANFIS-DE model error accordingly. Polynomial equations of bank profiles proposed by hybrid ANFIS models in the present study can be used in design and implementation of cross section of stable channels.

show abstract

Weather forecasting based on hybrid neural model

Cited by 52 publications

References 22 publications

Plasmodium species aware based quantification of malaria parasitemia in light microscopy thin blood smear

Plasmodium species aware based quantification of malaria parasitemia in light microscopy thin blood smear

Point Biserial Correlated Feature Selection of Weather Data

Analyzing bank profile shape of alluvial stable channels using robust optimization and evolutionary ANFIS methods

Contact Info

Product

Resources

About