Miguel Vallejo Orti's 2024 research dataset includes code and data for integrating volunteer-digitized gully lines. The experiments compare three approaches: Kalman filtering with varying input lines, Kalman filtering with self-learning, and a cross-training strategy. This dataset, harvested from heiDATA Dataverse, focuses on the role of basemaps and the number of volunteer contributions in mapping accuracy.
Use Cases
- Evaluating the impact of basemap quality on volunteer geographic information (VGI) integration based on the described Kalman filter experiments.
- Comparing self-learning and cross-training strategies for improving gully line accuracy based on the described methodology.
- Assessing the relationship between the number of volunteer contributions and final map quality based on the described experimental setup.
Strengths
- Includes code for three distinct methodological approaches (Kalman filter, self-learning, cross-training) as stated in the description.
- Explicitly investigates the role of basemaps and the quantity of volunteer contributions, providing a structured experimental framework.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
Provenance
- Source
- heiDATA Harvested Dataverse
- Collection Method
- Likely contains lines digitized by volunteers and integrated using computational methods.
- Freshness
- Last updated 2024-03-10 10:04:23; freshness should be verified.