General correlation coefficient based agglomerative clustering

Pandove, Divya; Goel, Shivani; Rani, Rinkle

doi:10.1007/s10586-018-2863-y

Cited by 6 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In many works such as Kumar et al (2018), Kumar (2019), Elumalai et al (2020) and Ewaid et al (2021) neglect the process of evaluating the best clustering method, directly applying Ward's method. However, this procedure may not be the most suitable for defining the composition of the dataset grouping structure (Pandove et al, 2019). For this reason, it was decided in this work to carry out the validation through the agglomeration coefficient (CA) (Equation ).

CA = \frac{1}{n} \sum_{i = 1}^{n} (1 - d (normali))

where, n is the total number of observations in the data group; and d (i): is the dissimilarity of object i to the first grouping.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Sustainability assessment in cities in the eastern amazon

Lima,

Progênio,

Costa

et al. 2023

Sustainable Development

View full text Add to dashboard Cite

Indicators are important tools that help monitor and assess sustainability, providing crucial information for public management. This study aimed to analyze the behavior of municipalities in eastern amazon (State of Pará), through sustainability indicators that cover social, economic, environmental and basic sanitation dimensions, applying the statistical method of multivariate analysis. For the selection of indicators, a checklist of academic papers on sustainability was carried out. For the composition of the clusters, the Ward method was applied. To validate the number of groups, the Pakhira‐Bandyopadhyay‐Maulik (PBM) index was used, obtaining 2 groups for the sanitation and general dimensions, 3 for environmental and social and 7 groups for economic. In this study, it was possible to demonstrate that several municipalities have deficiencies related to access to water supply and mainly sewage collection and treatment, causing various consequences for the population, as they do not comply with the commitments of the Sustainable Development Goals.

show abstract

CA = \frac{1}{n} \sum_{i = 1}^{n} (1 - d (normali))

where, n is the total number of observations in the data group; and d (i): is the dissimilarity of object i to the first grouping.…”

Section: Methodsmentioning

confidence: 99%

“…The agglomeration coefficient (AC) is a technique that helps in choosing the best grouping method based on the analysis of the quality of the dendrogram and the structure of the clusters in the observed data. The CA varies between 0 and 1, and the closer to 1, the better the grouping method will be (Crispim et al, 2020; Pandove et al, 2019).…”

Section: Methodsmentioning

confidence: 99%

Sustainability assessment in cities in the eastern amazon

Lima,

Progênio,

Costa

et al. 2023

Sustainable Development

View full text Add to dashboard Cite

show abstract

“…Third, the following linking methods were chosen and tested: average, single, complete and ward (Hair et al, 2006). Finally, the optimized group was found using the agglomerative coefficient (AC), defined as (Pandove et al, 2018;Kaufman & Rousseeuw, 2005):…”

Section: Methodsmentioning

confidence: 99%

A Survey on Gravity Waves in the Brazilian Sector Based on Radiosonde Measurements from 32 Aerodromes

Brhian

Ridenti

Roberto³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this paper, we applied a variety of statistical methods to study gravity waves in the troposphere and lower stratosphere in the Brazilian sector, using an unprecedented large database from Instituto de Controle do Espaço Aéreo (ICEA) of radiosonde measurements carried out in 2014 on 32 locations in the Brazilian territory totaling 49,652 wind profiles. The average wind profiles were computed and classified by means of a hierarchical cluster analysis. The kinetic and potential energy densities of the gravity waves were determined using a detrending technique based on the least squares method and the Fast Fourier Transform. The time series of the energy densities were analyzed in detail and some persistent and seasonal behavior was found in some cases. A systematic search for quasi monochromatic waves was carried out and the main characteristics of such waves propagating in the troposphere and the lower stratosphere were found. The correlation analysis between the troposphere and the lower ionosphere based on parameters observed on both layers was used to investigate the wave coupling between the two layers. The results we found have implications in the so-called seeding problem of the equatorial ionospheric irregularities.

show abstract

“…After that, the clusters were computed using the following hierarchical linking methods: average, single, complete and Ward's method (Hair et al, 2006). In order to choose the best linking method, they were ranked according to their agglomerative coefficient (AC), defined as (Kaufman & Rousseeuw, 2005;Pandove et al, 2018):…”

Section: Cluster Analysismentioning

confidence: 99%

A Survey on Gravity Waves in the Brazilian Sector Based on Radiosonde Measurements From 32 Aerodromes

Brhian,

Ridenti,

Roberto

et al. 2024

JGR Atmospheres

View full text Add to dashboard Cite

In this paper, we applied a variety of statistical methods to study gravity waves in the troposphere and lower stratosphere in the Brazilian sector, using a large database from Instituto de Controle do Espaço Aéreo (ICEA) of radiosonde measurements carried out in 2014 at 32 locations in the Brazilian territory totaling 49,652 wind and temperature profiles. The average wind profiles were computed and classified by means of a hierarchical cluster analysis. The kinetic and potential energy densities of gravity waves were determined using a detrending technique based on the Least Squares Method and the Fast Fourier Transform. By analyzing the energy density time series it was found that tropospheric average values are consistently larger in the months of winter, late autumn and early spring. Stratospheric average values of variability and kinetic energy density are also consistently larger in this period. A systematic search for quasi monochromatic waves was carried out and their main characteristics such as horizontal/vertical wavelengths and velocities were determined both in the troposphere and lower stratosphere. A correlation analysis between the troposphere and the lower stratosphere based on the measured parameters was used to investigate the wave coupling between the two layers, and no significant correlation was found. Finally, a spatial correlation analysis between energy densities measured at different aerodromes in the same atmospheric layer was carried out, showing that energy densities are spatially correlated for distances less than 3,000–4,000 km.

show abstract

General correlation coefficient based agglomerative clustering

Cited by 6 publications

References 44 publications

Sustainability assessment in cities in the eastern amazon

Sustainability assessment in cities in the eastern amazon

A Survey on Gravity Waves in the Brazilian Sector Based on Radiosonde Measurements from 32 Aerodromes

A Survey on Gravity Waves in the Brazilian Sector Based on Radiosonde Measurements From 32 Aerodromes

Contact Info

Product

Resources

About