scholarly journals A modified correlation in principal component analysis for torrential rainfall patterns identification

2020 ◽  
Vol 9 (4) ◽  
pp. 655
Author(s):  
Shazlyn Milleana Shaharudin ◽  
Norhaiza Ahmad ◽  
Siti Mariana Che Mat Nor
Keyword(s):  
Principal Component Analysis ◽  
High Dimension ◽  
Pearson Correlation ◽  
Principal Component ◽  
Breakdown Point ◽  
Distribution Patterns ◽  
Component Analysis ◽  
Peninsular Malaysia ◽  
Torrential Rainfall ◽  
Rainfall Patterns

This paper presents a modified correlation in principal component analysis (PCA) for selection number of clusters in identifying rainfall patterns. The approach of a clustering as guided by PCA is extensively employed in data with high dimension especially in identifying the spatial distribution patterns of daily torrential rainfall. Typically, a common method of identifying rainfall patterns for climatological investigation employed T mode-based Pearson correlation matrix to extract the relative variance retained. However, the data of rainfall in Peninsular Malaysia involved skewed observations in the direction of higher values with pure tendencies of values that are positive. Therefore, using Pearson correlation which was basing on PCA on rainfall set of data has the potentioal to influence the partitions of cluster as well as producing exceptionally clusters that are eneven in a space with high dimension. For current research, to resolve the unbalanced clusters challenge regarding the patterns of rainfall caused by the skewed character of the data, a robust dimension reduction method in PCA was employed. Thus, it led to the introduction of a robust measure in PCA with Tukey’s biweight correlation to downweigh observations along with the optimal breakdown point to obtain PCA’s quantity of components. Outcomes of this study displayed a highly substantial progress for the robust PCA, contrasting with the PCA-based Pearson correlation in respects to the average amount of acquired clusters and indicated 70% variance cumulative percentage at the breakdown point of 0.4.

scholarly journals Identification of Rainfall Patterns on Hydrological Simulation Using Robust Principal Component Analysis

2018 ◽  
Vol 11 (3) ◽  
pp. 1162 ◽  
Author(s):  
S.M. Shaharudin ◽  
N. Ahmad ◽  
N.H. Zainuddin ◽  
N.S. Mohamed
Keyword(s):  
Principal Component Analysis ◽  
Simulated Data ◽  
Principal Component ◽  
Breakdown Point ◽  
Component Analysis ◽  
Data Matrix ◽  
Robust Pca ◽  
Data Set ◽  
Number Of Components ◽  
Rainfall Patterns

A robust dimension reduction method in Principal Component Analysis (PCA) was used to rectify the issue of unbalanced clusters in rainfall patterns due to the skewed nature of rainfall data. A robust measure in PCA using Tukey’s biweight correlation to downweigh observations was introduced and the optimum breakdown point to extract the number of components in PCA using this approach is proposed. A set of simulated data matrix that mimicked the real data set was used to determine an appropriate breakdown point for robust PCA and  compare the performance of the both approaches. The simulated data indicated a breakdown point of 70% cumulative percentage of variance gave a good balance in extracting the number of components .The results showed a  more significant and substantial improvement with the robust PCA than the PCA based Pearson correlation in terms of the average number of clusters obtained and its cluster quality.

scholarly journals An Efficient Method to Improve the Clustering Performance using Hybrid Robust Principal Component Analysis-Spectral biclustering in Rainfall Patterns Identification

2019 ◽  
Vol 8 (3) ◽  
pp. 237
Author(s):  
Shazlyn Milleana Shaharudin ◽  
Shuhaida Ismail ◽  
Siti Mariana Che Mat Nor ◽  
Norhaiza Ahmad
Keyword(s):  
Principal Component Analysis ◽  
Data Center ◽  
Principal Component ◽  
Small Variation ◽  
Component Analysis ◽  
Peninsular Malaysia ◽  
Robust Pca ◽  
Robust Principal Component Analysis ◽  
Proposed Model ◽  
Rainfall Patterns

<p>In this study, hybrid RPCA-spectral biclustering model is proposed in identifying the Peninsular Malaysia rainfall pattern. This model is a combination between Robust Principal Component Analysis (RPCA) and bi-clustering in order to overcome the skewness problem that existed in the Peninsular Malaysia rainfall data. The ability of Robust PCA is more resilient to outlier given that it assesses every observation and downweights the ones which deviate from the data center compared to classical PCA. Meanwhile, two way-clustering able to simultaneously cluster along two variables and exhibit a high correlation compared to one-way cluster analysis. The experimental results showed that the best cumulative percentage of variation in between 65% - 70% for both Robust and classical PCA. Meanwhile, the number of clusters has improved from six disjointed cluster in Robust PCA-kMeans to eight disjointed cluster for the proposed model. Further analysis shows that the proposed model has smaller variation with the values of 0.0034 compared to 0.030 in Robust PCA-kMeans model. Evident from this analysis, it is proven that the proposed RPCA-spectral biclustering model is predominantly acclimatized to the identifying rainfall patterns in Peninsular Malaysia due to the small variation of the clustering result.</p>

PSXIII-26 Supplementation with a polyphenolic pine bark extract reduces ammonia nitrogen production in a RUSITEC system as determined by principal component analysis

2021 ◽  
Vol 99 (Supplement_3) ◽  
pp. 435-436
Author(s):  
Nelson Vera ◽  
Constanza Gutierrez ◽  
Pamela Williams ◽  
Cecilia Fuentealba ◽  
Rodrigo Allende ◽  
...  
Keyword(s):  
Principal Component Analysis ◽  
Repeated Measures ◽  
Pearson Correlation ◽  
Principal Component ◽  
Component Analysis ◽  
Ammonia Nitrogen ◽  
Pine Bark ◽  
Bark Extract ◽  
Ruminal Fermentation ◽  
Fermentation Parameters

Abstract The aim of the study was to correlate the effects of supplementation with a polyphenolic pine bark extract (PBE) in diets with different forage to concentrate (F:C) ratio on methane (CH4), ammonia nitrogen (NH3–N) production and ruminal fermentation parameters using the Rumen Simulation Technique (RUSITEC). The experimental diets were F:C 70:30 (HF) or F:C 30:70 (HC) with or without 2% PBE on a DM basis. The four diets were isoproteic (15% CP), with similar OM (HF 94% and HC 96%), but different NDF (HF 40% and HC 25%). The treatments, in duplicate, were assigned in an 8 fermenter RUSITEC apparatus. Incubations were run twice, with 5 days of sampling after 10 days adaptation. The experimental design was a 2x2 factorial arrangement in a randomized complete block with repeated measures. Pearson correlation and principal component analysis (PCA) were conducted to elucidate relationships among PBE total polyphenols (TP) and the variables evaluated. The TP was highly correlated with NH3–N (r = –0.98; P &lt; 0.001) and butyrate (r = –0.85; P &lt; 0.001), and had a high correlation with propionate (r = 0.75; P &lt; 0.001) and acetate (r = 0.68; P = 0.001). Correlation with total VFA was moderate (r = –0.59; P = 0.006), and CH4 yield and IVDMD there were not correlated (r ≤ –0.07; P ≥ 0.188). The PCA (KMO = 0.655; BTS &lt; 0.001) shows that 75.2% of the total variation is explained by the first two principal components (PC1 = 46.5% and PC2 = 28.7%). In the score plot, PC1 discriminated between diets with and without PBE, while the PC2 separated based on NDF. The loading plot showed that TP and propionate were clustered, and had inverse directions to NH3–N. In conclusion, the PBE supplementation reduces NH3–N production in a RUSITEC system without decreasing CH4 yield or negatively affecting ruminal fermentation parameters.

scholarly journals No-tillage system participatory quality index

2020 ◽  
Vol 24 (2) ◽  
pp. 128-133 ◽  
Author(s):  
Tiago S. Telles ◽  
Ana J. Righetto ◽  
Marco A. P. Lourenço ◽  
Graziela M. C. Barbosa
Keyword(s):  
Principal Component Analysis ◽  
Cluster Analysis ◽  
Crop Rotation ◽  
Quality Index ◽  
Pearson Correlation ◽  
Principal Component ◽  
Component Analysis ◽  
Soil Tillage ◽  
No Tillage ◽  
Tillage System

ABSTRACT The no-tillage system participatory quality index aims to evaluate the quality and efficiency of soil management under no-tillage systems and consists of a weighted sum of eight indicators: intensity of crop rotation, diversity of crop rotation, persistence of crop residues in the soil surface, frequency of soil tillage, use of agricultural terraces, evaluation of soil conservation, balance of soil fertilization and time of adoption of the no-tillage system. The aim of this study was to assess the extent to which these indicators correlate with the no-tillage system participatory quality index and to characterize the farmers who participated in the research. The data used were provided by ITAIPU Binacional for the indicators of the no-tillage system participatory quality index II. Descriptive analyses were performed, and the Pearson correlation coefficient between the index and each indicator was calculated. To assess the relationship between the indicators and the farmers’ behavior toward the indicators, principal component analysis and cluster analysis were performed. Although all correlations are significant at p-value ≤ 0.05, some correlations are weak, indicating a need for improvement of the index. The principal component analysis identified three principal components, which explained 66% of the variability of the data, and the cluster analysis separated the 121 farmers into five groups. It was verified that the no-tillage system participatory quality index II has some limitations and should therefore be reevaluated to increase its efficiency as an indicator of the quality of the no-tillage system.

scholarly journals Inferência Exploratória de Dados Espaço-Temporal da Precipitação Pluviométrica no Nordeste Brasileiro

2020 ◽  
Vol 13 (5) ◽  
pp. 2019
Author(s):  
Jhon Lennon Bezerra da Silva ◽  
Geber Barbosa De Albuquerque Moura ◽  
Marcos Vinícios Da Silva ◽  
Roni Valter De Souza Guedes ◽  
Pabrício Marcos Oliveira Lopes ◽  
...  
Keyword(s):  
Principal Component Analysis ◽  
Multivariate Analysis ◽  
Temporal Variability ◽  
Principal Component ◽  
Component Analysis ◽  
Northeastern Brazil ◽  
Annual Rainfall ◽  
Intermittent Rivers ◽  
Rainfall Patterns ◽  
Brazilian Northeast

A gestão eficiente dos recursos hídricos no Nordeste brasileiro torna-se fundamental diante do regime hidrológico dos rios intermitentes, dos quais muitos são extremamente críticos. Todavia estes dependem de um regime pluviométrico irregular, tanto em escala de tempo mensal quanto anual. Objetivou-se determinar a variabilidade espaço-temporal da precipitação pluviométrica total anual, averiguando-se, também, as regiões com padrões de precipitação semelhantes por técnicas de análise multivariada (clusters e componentes principais) no Nordeste do Brasil. Foram analisados dados de precipitação pluviométrica total anual, entre os anos de 1995 e 2016, de 37 diferentes estações meteorológicas do INMET, estas situadas nos limites territoriais dos nove estados do Nordeste brasileiro. A análise de clusters verificou a formação de quatro grupos distintos, com padrões semelhantes de precipitação nas regiões dentro dos grupos, conforme também observado na análise de componentes principais. A padronização e/ou variabilidade espaço-temporal da precipitação pluviométrica dos municípios analisados mostrou-se está intimamente associada as condições das estações do ano e anomalias climatológicas, aos fatores de uso e ocupação do solo, condições de altitude e relevo, tais quais favorecem na formação e estabilidade de chuvas menores ou maiores no Nordeste brasileiro. A análise multivariada de cluster e componentes principal identificaram padrões e semelhanças pluviométricas de grupos, nos diferentes estados do Nordeste do Brasil entre os anos de 1995 e 2016. Exploratory Inference of Spatial-Temporal Data of Rainfall in the Brazilian Northeast ABSTRACTThe efficient management of water resources in the Northeast of Brazil is essential in view of the hydrological regime of intermittent rivers, of which many are extremely critical, as they depend on an irregular rainfall regime, both on a monthly and annual time scale. The objective of this study was to determine the spatial and temporal variability of the annual total rainfall, also investigating the regions with similar rainfall patterns by multivariate analysis techniques (clusters and principal components) in Brazilian Northeast. Data from total annual rainfall between the years 1995 and 2016, of 37 different INMET weather stations were analyzed, located within the territorial limit of the nine states of Brazilian Northeast. Cluster analysis verified the formation of four distinct groups, with similar precipitation patterns in the regions within the groups as also observed in the principal component analysis. The pattern and/or spatial-temporal variability of rainfall in the municipalities analyzed was shown to be intimately associated with the conditions of the year and climatic anomalies stations, and the factors of land use and occupation, altitude and relief conditions, such as favoring the formation and stability of minor or major rain in the Brazilian Northeast. Multivariate cluster and principal component analysis identified rainfall patterns and similarities of groups, in the different states of Northeastern Brazil between the years 1995 and 2016.Keywords: multivariate analysis, climate change, semiarid, regional climate patterns.

scholarly journals Modal Principal Component Analysis

2020 ◽  
Vol 32 (10) ◽  
pp. 1901-1935
Author(s):  
Keishi Sando ◽  
Hideitsu Hino
Keyword(s):  
Principal Component Analysis ◽  
Minor Component ◽  
Principal Component ◽  
Breakdown Point ◽  
Component Analysis ◽  
Finite Sample ◽  
Robust Pca ◽  
Theoretical Contribution ◽  
Mode Estimation ◽  
Finite Sample Breakdown Point

Principal component analysis (PCA) is a widely used method for data processing, such as for dimension reduction and visualization. Standard PCA is known to be sensitive to outliers, and various robust PCA methods have been proposed. It has been shown that the robustness of many statistical methods can be improved using mode estimation instead of mean estimation, because mode estimation is not significantly affected by the presence of outliers. Thus, this study proposes a modal principal component analysis (MPCA), which is a robust PCA method based on mode estimation. The proposed method finds the minor component by estimating the mode of the projected data points. As a theoretical contribution, probabilistic convergence property, influence function, finite-sample breakdown point, and its lower bound for the proposed MPCA are derived. The experimental results show that the proposed method has advantages over conventional methods.

Sign in / Sign up

Export Citation Format

Share Document