Version Similarity-based Model for Volunteers' Reputation of Volunteered Geographic Information: A Case Study of Polygon

At present, it is difficult to evaluate the quality of volunteered geographic information(VGI), which have malicious, false, and poor quality data. Therefore, a version similarity-based reputation model for volunteers of VGI system is proposed. In the model, each editing to a geographic spatial object of each volunteer is defined as a version. When the object version is modified by other users, support degree of the version is computed through version similarity. Then, support degree of every object contributed by a volunteer is calculated according by others' modifications. The volunteer's reputation is obtained through weighted average of all his support degrees. The version similarity composites major factors of spatial similarity and attributes similarity between versions of a same object. Polygon objects are employed as an example to describe the computation process of volunteer reputation of our model. For verifying and analyzing the rationality of our reputation model, Berlin's historical data from OpenStreetMap are used for experiments. It shows that users' reputation calculated by our model has a positive correlation with quality of objects contributed by them as a whole