Download the CSV template to get started

With the CSV import tool, it is possible to import papers in bulk to the ORKG. After uploading the CSV file, you will be able to see how to data will be imported (in step 5). The CSV file has to be in UTF-8 encoding for better compatibility. Before starting the actual import it is possible to first preview the data to ensure it is read and formatted correctly.

After importing a CSV file, it is possible to create a comparison using the imported papers.

Formatting rules

The extracted table should be formatted in such a way that it adheres to the following rules:

1. First row labels. Only the first row should contain the header labels (transpose the table if necessary, reformat the table if necessary)
2. One paper per row. Each row should contain a single paper. It is not possible to use multiple rows to describe the same contribution. Merge information in the same cells if necessary.
3. Use one row per contribution and repeat the paper title as needed If a paper has multiple contributions. It's not possible to split a single contribution across multiple rows.
4. The title or DOI for a paper are required fields.
   Tip: When a DOI is provided, the metadata of this row is fetched automatically. So it is not necessary to provide the paper metadata fields: title, authors, publication_month, publication_year, published_in, url.
5. Special cell values. It is possible to predefine a column data type or cell data type by appending the type in angular brackets. See the data types descriptions below for more details.
6. Special cell values. It is possible to prepend resource: to a cell value to create a resource instead. Also, it is possible to use an existing ORKG resource. Use as cell value orkg: followed by the resource ID to do this (e.g. orkg:R100). See the CSV example below for more details.

*In this example, Location property in the second paper will take two values (Hannover and Dresden)*

Step 1: CSV import

Upload the CSV file.

Step 2: Map properties

Map column headers to ORKG properties. An automated mapping will support you with that, reusing existing properties is preferred over creating new ones.

Step 3: Map column types

To maintain consistency across your dataset, each column should be mapped to a simple, standardized data type. An automated mapping process will assist you by suggesting the most suitable data type for each column based on its content.

If individual cells deviate from the automatic column type detection, you’ll have the opportunity to define their specific data types in the next step manually—unless this was already done before import. After mapping, an automatic validation will be performed to detect and report any inconsistencies (i.e., values that do not match the assigned column type). Data types defined in a cell take precedence over those defined in the header.

You can define cell-level data types by appending the type directly after the content in the same cell.

• In cell: 213<integer>
• In header: Number of participants<integer>
• Automatic: http://google.com (considered a <url> automatically)

List of data types (case-insensitive)

• <Text> or <String>
• <Decimal>
• <Integer> (whole numbers)
• <Boolean> (true or false, 0 or 1)
• <Date> (date values YYYY-MM-DD)
• <URL> (starts with a protocol like http)

(Optional) Step 3a: Export as CSV

When you want to save the current progress of the editing of your CSV, you can export the CSV again. This file will contain the headers that are mapped already to ORKG properties in form of their ID as well as their set data types in form of angle brackets appended.

Step 4: Check file

Check the data for validation errors and edit single cells if necessary. Here you can set individual cell data types by editing a single cell.

Step 5: Import papers

Once all validation errors are resolved and you are satisfied with the result, you can import the papers into the ORKG. A last preview will show you how the imported data will look like.