All task-related data stored on the LibCrowds platform is made publically available for further research or integration into other systems. There are several ways of accessing this data, which are described further below.
In the majority of cases, the final result data is likely to be the most useful. This is the data produced once a task has been completed and all contributions analysed. In their raw form, these final results are serialised according to the Web Annotatations data model and wrapped with additional information about the tasks that produced them. See the Data Model section for more details.
All task and contribution data is made available for direct download in CSV and JSON formats. The structure of each dataset varies according to its purpose. For example, the task datasets contain information required to configure each task, whereas the contribution datasets contain the information submitted as answers to those tasks. See the Downloads section for more details.
Python examples in Jupyter Notebooks¶
These Notebooks provide another way to explore and create simple visualisations of LibCrowds data.
A range of API endpoints are provided to further transform these datasets. For instance, the final results data can be retrived as Annotation Lists that can be consumed by IIIF compatible image viewers. See the API section for more details.