Identification of homogeneous regions for regional frequency analysis using the self-organizing map

Lin, Gwo‐Fong; Chen, Lu-Hsien

doi:10.1016/j.jhydrol.2005.09.009

Cited by 161 publications

(113 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One being the K-Means technique and the other is the probability technique. K-Means is one of the most commonly used non-hierarchical clustering techniques (Lin and Chen 2006). K-Means technique involves "d" number of variables of the dataset, "x", and N number of feature vectors which are able to cluster and classify into K number of clusters (Burn and Goel, 2000).…”

Section: Non-hierarchical Clustering (K-means) Methodsmentioning

confidence: 99%

Financial Differences and Similarities of Islamic Banks: A Study on Qismut Countries

Yıldırım¹

2015

Pressacademia

View full text Add to dashboard Cite

Six countries, namely Qatar, Indonesia, Saudi Arabia, Malaysia, UAE and Turkey (abbreviated as QISMUT) are considered to play an important role in the future international development of the Islamic finance. QISMUT countries have a majority Muslim population who prefers Islamic finance for their banking needs. QISMUT countries account for 38 million customers which is 67% of the global Islamic bank customer base. The focus of this study is to define similar (or different) international banks using the financial data collected from Islamic banks operating in QISMUT countries between 2012 and 2014. In this context, 56 Islamic banks were categorized in the light of their financial ratios for the period between 2012 and 2014 using cluster analysis. There was also an attempt to identify these clusters using multidimensional scaling analysis. Positions of QISMUT Islamic banks in the multidimensional space were defined using multidimensional scaling analysis. Results showed that Islamic banks operating in QISMUT countries are similar in terms of their financial structure.

show abstract

Section: Non-hierarchical Clustering (K-means) Methodsmentioning

confidence: 99%

Financial Differences and Similarities of Islamic Banks: A Study on Qismut Countries

Yıldırım¹

2015

Pressacademia

View full text Add to dashboard Cite

show abstract

“…After comparing the SOM and k-means methods, Bação et al (2005) proposed the use of SOMs as possible substitutes for the k-means clustering algorithms. Lin & Chen (2006) tested the cluster accuracy of the SOM, the k-means method and Ward's method based on experimental data sets that the amount of cluster dispersion and the cluster membership are controlled and known. They found that the SOM determined the cluster membership more accurately than the K-means method and Ward's method.…”

Section: Advantages Over Other Conventional Methodsmentioning

confidence: 99%

“…Pelletier et al (2009) applied the SOM in the characterization of 1-h rainfall temporal patterns in a Québec City case study. Lin & Chen (2006) and Lin & Wu (2007) used the SOM to analyze the rainfall data on Taiwan Island. Recently, Hsu & Li (2010) used the SOM and wavelet methods to explore spatio-temporal characteristics of the 22 years of precipitation data for Taiwan Island.…”

Section: Introductionmentioning

confidence: 99%

A Review of Self-Organizing Map Applications in Meteorology and Oceanography

Liu¹,

Weisberg²

2011

Self Organizing Maps - Applications and Novel Algorithm Design

150

108

View full text Add to dashboard Cite

“…Il est aujourd'hui communément accepté que la proximité géographique des bassins versants ne représente pas une garantie de leur similarité hydrologique. Cependant, plusieurs auteurs continuent à essayer de raffiner les approches de détermination des régions fixes basées, par exemple, sur l'analyse factorielle des correspondances, la classification ascendante hiérarchique ou l'approche de classification par carte auto-organisée (voir, par exemple, LIN et CHEN, 2006;RAO et SRINIVAS, 2006a. Les nouvelles approches proposées tentent souvent d'intégrer dans les procédures de détermination des régions homogènes de nouveaux éléments tels que l'utilisation des approches floues pour la modélisation de l'incertitude associée à l'information utilisée.…”

Section: Régionaleunclassified

Synthèse des développements récents en analyse régionale des extrêmes hydrologiques

Ouarda

St‐Hilaire

Bobée

2008

rseau

View full text Add to dashboard Cite

SYNTHÈSE DES DÉVELOPPEMENTS RÉCENTS EN ANALYSE RÉGIONALE DES EXTRÊMES HYDROLOGIQUES ABSTRACTAdequate estimation of extreme hydrological variables is essential for the rational design and operation of a variety of hydraulic structures, due to the significant risk that is associated with these activities. Local frequency analysis is commonly used for the estimation of extreme hydrological events at sites where an adequate amount of data is available. However, data are usually only collected at a relatively limited number of sites. In practice, it frequently happens that little or no streamflow data is available at a site of interest (where a dam is to be constructed for example). In such cases, hydrologists can utilize a regional flood frequency procedure, relying on data available from other basins with a similar hydrologic regime.Various methods have been developed over the last few years for the regional analysis of extreme hydrological events. These regionalization approaches aim to estimate different characteristics of the extreme hydrological phenomena of interest, make different assumptions and hypotheses concerning these hydrological phenomena, rely on various types of data, and often fall under completely different theories. The present paper aims to review and classify recent developments in regional frequency analysis of extreme hydrological variables. Analyse régionale des extrêmes hydrologiques 220The specific objectives of the paper are to: i) review the main recent developments in regional hydrologic modeling that have been proposed during the last few years; ii) classify these developments into different groups according to the theoretical background of the method, its specific objectives, and the characteristics of hydrological extreme phenomena it is intended to deal with; iii) propose a comprehensive discussion of these methods, and point out the hypotheses, limitations, data requirements, and potential of each one; iv) identify the new challenges facing engineers in terms of regional frequency analysis of hydrological extremes; and v) propose potential promising directions for future research work which aim to meet these new challenges.Recent developments reviewed in the present paper include improvements in classical approaches for regional delineation and for information transfer, methods combining the delineation and estimation steps, seasonality-based methods, multivariate models for regional frequency analysis, the QdF approach, non stationary models, and approaches for the combination of local and regional data. The paper provides also a discussion of the various hydrological variables treated with regional estimation methodologies, comparative studies of these methodologies, and practical tools that were developed for regional frequency analysis. It is hoped that this document will contribute towards closing the gap between theory and practice, by narrowing the wide body of literature that is available, and by providing comprehensive propositions for regional frequency analysis approache...

show abstract

Identification of homogeneous regions for regional frequency analysis using the self-organizing map

Cited by 161 publications

References 20 publications

Financial Differences and Similarities of Islamic Banks: A Study on Qismut Countries

Financial Differences and Similarities of Islamic Banks: A Study on Qismut Countries

A Review of Self-Organizing Map Applications in Meteorology and Oceanography

Synthèse des développements récents en analyse régionale des extrêmes hydrologiques

Contact Info

Product

Resources

About