Alternative Strategies for Mapping ACS Estimates and Error of Estimation

Francis, Joe; Tontisirin, Nij; Anantsuksomsri, Sutee; Vink, Jan; Zhong, Viktor

doi:10.1007/978-94-017-8990-5_16

Cited by 12 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is agreed and confirmed by Jebur et al [120], Bui et al [121], Bui et al [107] and Shirzadi et al [112]. Gully erosion susceptibility maps were prepared by all machine learning models used in this study and then classified using four known classification methods such as natural breaks, quantile, geometrical interval and equal interval [122]. All these methods for classifying gully erosion susceptibility maps and the results indicated that, for example natural breaks, geometrical interval and equal interval, had a low logical prediction as visualization.…”

Section: Discussionsupporting

confidence: 80%

“…Additionally, quantile method could assign more gully erosion location in the high and very high susceptibility classes of gully susceptibility maps in all machine learning algorithms. Some researchers have used the quantile classification method to divide the natural hazards susceptibility index such as Umar et al [123], while Farncis et al [122] used natural beaks, Pham et al [124] used geometrical interval classification methods in their study.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Ensemble Artificial Intelligence Approach for Gully Erosion Mapping in a Semi-Arid Watershed (Iran)

Bui

Shirzadi

Shahabi

et al. 2019

Sensors

View full text Add to dashboard Cite

In this study, we introduced a novel hybrid artificial intelligence approach of rotation forest (RF) as a Meta/ensemble classifier based on alternating decision tree (ADTree) as a base classifier called RF-ADTree in order to spatially predict gully erosion at Klocheh watershed of Kurdistan province, Iran. A total of 915 gully erosion locations along with 22 gully conditioning factors were used to construct a database. Some soft computing benchmark models (SCBM) including the ADTree, the Support Vector Machine by two kernel functions such as Polynomial and Radial Base Function (SVM-Polynomial and SVM-RBF), the Logistic Regression (LR), and the Naïve Bayes Multinomial Updatable (NBMU) models were used for comparison of the designed model. Results indicated that 19 conditioning factors were effective among which distance to river, geomorphology, land use, hydrological group, lithology and slope angle were the most remarkable factors for gully modeling process. Additionally, results of modeling concluded the RF-ADTree ensemble model could significantly improve (area under the curve (AUC) = 0.906) the prediction accuracy of the ADTree model (AUC = 0.882). The new proposed model had also the highest performance (AUC = 0.913) in comparison to the SVM-Polynomial model (AUC = 0.879), the SVM-RBF model (AUC = 0.867), the LR model (AUC = 0.75), the ADTree model (AUC = 0.861) and the NBMU model (AUC = 0.811).

show abstract

Section: Discussionsupporting

confidence: 80%

Section: Discussionmentioning

confidence: 99%

A Novel Ensemble Artificial Intelligence Approach for Gully Erosion Mapping in a Semi-Arid Watershed (Iran)

Bui

Shirzadi

Shahabi

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…The bivariate approach also has been popularly used for uncertainty visualization, because an attribute and its associated uncertainty information may be considered to be a pair of variables. Previous studies conclude that the bivariate approach is more effective than adjacent display in terms of cluster detection [9,22,29]. A major advantage of this approach is that readers do not need to switch their focus back and forth between two maps to build connections between the estimates and their uncertainties [11].…”

Section: Related Workmentioning

confidence: 99%

“…However, some studies [11,38] discuss that a visual representation of absolute error measures (e.g., the standard error of an estimate) may lead to an inappropriate conclusion and interpretation because large attribute values tend to have large standard errors. Furthermore, Francis et al [29] discuss the appropriateness of using relative measures of errors, such as the coefficient of variation (CV), for counts (e.g., the total number of households and of occupied houses) and median or mean values (e.g., median age and mean household income). Nevertheless, an absolute measure of error is still appropriate for proportional or ratio values (e.g., percentage of Asians), which are bounded by a specific value range: i.e., between 0 and 1 for proportional values and between 0 and 100 for percentages [29].…”

Section: An Applicationmentioning

confidence: 99%

Geovisualizing attribute uncertainty of interval and ratio variables: A framework and an implementation for vector data

Koo

Chun

Griffith

2018

Journal of Visual Languages & Computing

View full text Add to dashboard Cite

Geovisualization of attribute uncertainty helps users to recognize underlying processes of spatial data. However, it still lacks an availability of uncertainty visualization tools in a standard GIS environment. This paper proposes a framework for attribute uncertainty visualization by extending bivariate mapping techniques. Specifically, this framework utilizes two cartographic techniques, choropleth mapping and proportional symbol mapping based on the types of attributes. This framework is implemented as an extension of ArcGIS in which three types of visualization tools are available: overlaid symbols on a choropleth map, coloring properties to a proportional symbol map, and composite symbols.

show abstract

“…The cartographic styles of bivariate hachures followed the recommendations of Slocum et al (2009), and the classification of hachure intervals on choropleth maps followed the standard of Francis et al (2015).…”

Section: Multivariate Geovisualization Productsmentioning

confidence: 99%

Multivariate Geovisualization of Dengue, Zika and Chikungunya Cases in Brazil: A Didactic Experience

Vasconcelos

Pinho

2017

Hygeia

View full text Add to dashboard Cite

In recent decades, the Aedes aegypti mosquito, the vector of dengue, Zika, chikungunya and other diseases, has become an increasing concern in Brazil and other tropical countries. However, the spatial distribution of these diseases in Brazil is not homogeneous, implying that there are distinct risk levels and various guiding strategies for fighting these diseases. This paper presents a didactic exploratory multivariate spatial data analysis of dengue, Zika and chikungunya cases in Brazil in 2016 in the context of an undergraduate course. The students elaborated, interpreted and discussed multivariate maps using techniques such as proportional symbols, hachure density, bivariate and trivariate choropleth composition, contiguous cartograms, conditional maps and matrices of maps and graphs. The students investigated the various interpretative possibilities based on the intrinsic and extrinsic combinations of multivariate geovisualizations. Through their interpretations of the spatial representations, the students were able to identify that the states of Rio Grande do Norte, Paraíba and Alagoas are experiencing the most critical situations for these three diseases, while the state of Minas Gerais has a serious situation only for dengue. In contrast, the maps clearly revealed that the states of Rio Grande do Sul, Santa Catarina and the Northern Region of the country experienced fewer cases of the diseases. It can be concluded that the undergraduate course improved the students' abilities to build and interpret spatial representations using geotechnologies.Keywords: Aedes aegypti. Dengue. Zika. Chikungunya. Geovisualization. Exploratory Spatial Data Analysis. Geotechnology. Geographic Information Systems. Didactics. RESUMOA transmissão de Dengue, Zika, Chikungunya e outras enfermidades pelo mosquito Aedes aegypti tem se tornado, no decorrer das últimas décadas, uma preocupação cada vez mais relevante no Brasil e em outros países tropicais. Todavia, a distribuição espacial dos casos dessas doenças no Brasil não é homogênea, implicando diferentes graus de risco e estratégias de combate às doenças. Este artigo apresenta uma experiência didática de análise exploratória espacial multivariada dos dados de ocorrência de Dengue, Zika e Chikungunya no Brasil, em 2016, no contexto de uma disciplina de graduação. Foram elaborados, interpretados discutidos, em conjunto com os estudantes, mapas multivariados utilizando técnicas de símbolos proporcionais, densidade de hachuras, composição coroplética bivariada e trivariada, cartogramas contíguos, mapas condicionais e matrizes de mapas e gráficos. Investigaram-se as diversas possibilidades de interpretação a partir das combinações intrínsecas e extrínsecas de geovisualização multivariada.

show abstract

Alternative Strategies for Mapping ACS Estimates and Error of Estimation

Cited by 12 publications

References 12 publications

A Novel Ensemble Artificial Intelligence Approach for Gully Erosion Mapping in a Semi-Arid Watershed (Iran)

A Novel Ensemble Artificial Intelligence Approach for Gully Erosion Mapping in a Semi-Arid Watershed (Iran)

Geovisualizing attribute uncertainty of interval and ratio variables: A framework and an implementation for vector data

Multivariate Geovisualization of Dengue, Zika and Chikungunya Cases in Brazil: A Didactic Experience

Contact Info

Product

Resources

About