Article ; Online: Exploiting the Potential of VGI Metadata to Develop A Data-Driven Framework for Predicting User’s Proficiency in OpenStreetMap Context
ISPRS International Journal of Geo-Information, Vol 8, Iss 11, p
2019 Volume 492
Abstract: Volunteered geographic information (VGI) encourages citizens to contribute geographic data voluntarily that helps to enhance geospatial databases. VGI’s significant limitations are trustworthiness and reliability concerning data quality due to the ... ...
| Abstract | Volunteered geographic information (VGI) encourages citizens to contribute geographic data voluntarily that helps to enhance geospatial databases. VGI’s significant limitations are trustworthiness and reliability concerning data quality due to the anonymity of data contributors. We propose a data-driven model to address these issues on OpenStreetMap (OSM), a particular case of VGI in recent times. This research examines the hypothesis of evaluating the proficiency of the contributor to assess the credibility of the data contributed. The proposed framework consists of two phases, namely, an exploratory data analysis phase and a learning phase. The former explores OSM data history to perform feature selection, resulting in “OSM Metadata” summarized using principal component analysis. The latter combines unsupervised and supervised learning through K-means for user-clustering and multi-class logistic regression for user classification. We identified five major classes representing user-proficiency levels based on contribution behavior in this study. We tested the framework with India OSM data history, where 17% of users are key contributors, and 27% are unexperienced local users. The results for classifying new users are satisfactory with 95.5% accuracy. Our conclusions recognize the potential of OSM metadata to illustrate the user’s contribution behavior without the knowledge of the user’s profile information. |
|---|---|
| Keywords | openstreetmap ; vgi ; metadata ; principal component analysis ; k-means ; multiclass logistic regression ; Geography (General) ; G1-922 |
| Subject code | 004 |
| Language | English |
| Publishing date | 2019-10-01T00:00:00Z |
| Publisher | MDPI AG |
| Document type | Article ; Online |
| Database | BASE - Bielefeld Academic Search Engine (life sciences selection) |
Full text online
More links
Kategorien
Inter-library loan at ZB MED
Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.