A framework for classification of volunteered geographic data based on user's need. Issue 11 (3rd July 2021)

• Record Type: Journal Article
• Authors: Mohammadi, Nazila; Sedaghat, Amin
• Is Part Of: Geocarto international. Volume 36:Issue 11(2021)
• Page Start: 1276
• Page End: 1291
• Publication Date: 2021-07-03
• Subjects: Volunteered geographic data -- quality -- ANNs -- PCA -- indicator; Remote sensing -- Periodicals; Geographic information systems -- Periodicals; Geology -- Periodicals; Cartography -- Periodicals; 621.3678
• DOI: 10.1080/10106049.2019.1641562
• Languages: English
• ISSNs: 1010-6049

Abstract

Abstract: VGI is an attractive source of data, but the quality assurance limits its usages. This study proposes a framework to estimate the quality of the VGI and to classify them based on the user's need. For this purpose, a set of properties is defined to describe the data in various aspects. The principal component analysis (PCA) method is applied to reach a new set of uncorrelated indicators (UI). Volunteered data is classified based on the user's need and takes a quality index (QI). UI and QI values are used to train the ANN. Finally, the trained ANN determines the output of the network in a way that returns QI using the UI as inputs. The proposed method was applied to estimate the quality classes of VGI in a part of an urban area. According to the results of the confusion matrix, the total accuracy of the proposed framework was 81.6%.