Human-centered Geographic Classification Study

First, thank you for your participation to help science. We are concerned with data quality of Volunteered Geographic Information (VGI) and particulary data classification. Nowadays, everyone have access to the internet can collect and publish geographic information. For example, OpenStreetMap (OSM) - the world free maps - have been evolved mostly by volunteers and amateurs. Thus, this study investigates human perception of gepgraphic objects. The study consists of 4 steps and would take about 25-30 minutes. THANK YOU.

Step 1

Participant's Information

You would fill in some anonymous information about your age, education, location, and your geographic expertise. We do not collect any identifiable personal information. Don't forget that protecting your private information is our priority.

Step 2

Task I: Landuse Classification

A set of geographic objects would be presented. For each object, your task is to identify the landuse classification of the given object with respect to various abstraction of classification.
You are provided with the CORINE schema of LU/LC classification. The schema divide into three levels of classification abstraction.

LU/LC = LandUse/LandCover

Step 3

Task II: Classification Plausibility

A geographic object and a set of plausible classes would be presented. Your task is to determine the plausibility level of each class for the given object with respect to your perception.
Afterwards, you are asked to provide opinion about object's characteristics and its geographic context.

Step 4


You would answer a questionnaire. We use NASA Task Load Index (NASA-TLX) to assess subjective workload. It has become the gold standard for measuring subjective workload across a wide range of applications.