You may already have ways to gather social media data — for example, a subscription to an external scraper like Apify, your own scrapers that you run locally, or access to a platform API. The manual upload function allows you to bring your own social media data into Phoenix in a format that works for the other components of Phoenix (classify and explore).

This is useful if you prefer to continue to gather data in a way you are used to and / or you have access to platforms we have not yet added to Phoenix Gather.

In order to use the manual upload function, you will need to put your data in a specific CSV format using this standard template.

• In the column names, message_ means either post or comment.
• The columns comment_replied_to_id_pi and comment_parent_post_id_pi are only relevant if you are uploading comments. If you are uploading only posts, leave these columns blank.
• The columns message_id_pi, message_author_id_pi, message_author_id_pi and comment_replied_to_id_pi and comment_parent_post_id_pi all refer to unique IDs for these elements. If you have downloaded data from a scraper or API, it will probably have these ids already. If not, then you will need to assign unique IDs.

Once your CSV is correctly formatted, save it locally so it is ready to be uploaded. Note that your file cannot be bigger than 1GB; if it is then split your dataset to make several smaller files.

Then go to the Phoenix platform, click on Create in the Gather tab, and scroll to the bottom to find “Manually upload gathered data”. You can give your Gather a name so that you remember what it is, and then upload your CSV file.

All uploaded data will be classified in the same way as data gathered in other ways, and you will be able to explore it alongside other data you have gathered.