How do I format my data for a sentiment analysis task on Metaranx?

Sentiment analysis is a language task that requires text-based data. You will train your AI on text-based data and test/use it on text-based data. When you have completed this list, you will have a dataset prepared for building a sentiment analysis AI on Metaranx.

Open your preferred spreadsheet in preparation for entering your data (you can use Excel, Google Sheets, or any spreadsheet program that can export to a .CSV). Your dataset should be a maximum of 5MB.

Set up three columns

‍
Column 1

• Column 1 (or "A") needs to exist but remain blank
• Column 1 (or "A") header/title remains blank
  ‍

Column 2

• Column 2 (or "B) needs to hold all your text-based data
• Column 2 (or "B") does not require a header. Remove blank headers.
• Add your data into rows, divided by their "label"

‍
Column 3

• Column 3 (or "C") needs to hold your labels
• Column 3 (or "C") does not require a header. Remove blank headers.
• Labels are between 1 and 10: for example all positive statements will be labeled as "0" and all negative statements will be labeled as "1" for a specific use case of positive versus negative sentiment detection. Your labels need to be consistent to their corresponding text, but can follow any pattern you choose.

• Add at least 1,000 rows per label with the goal to have 10,000 rows of data minimum
• Export to a .CSV file

• Upload to Metaranx to build your new AI!