Many users want to include their own labels, concepts, or tags alongside their data entries when uploading CSV files for analysis. Beehive AI fully supports this, making it easy to incorporate a coding system directly during the upload process for more customized analysis. Here’s how it works:
Key Features
Direct Label Upload: Labels, tags, or concepts can be added directly within CSV files.
Multiple Labels: Multiple labels can be included for a single data entry.
Steps to Include Labels in CSV Files
1️⃣ Prepare the CSV File
To include labels with data entries:
Add a New Column: Create an additional column in the CSV file specifically for storing the labels.
While not strictly required, it is strongly recommended to place this new
- Labelscolumn immediately after the Parent column it is linked to.
ID | Date | What would convince you to switch to a plant based burger? |
|
502cjl5… | 2020-02-28 | I would switch if there are healthier ingredients. |
|
6a1upc… | 2020-03-01 | Taste |
|
Mark the Column: The new column’s header must be exactly the same as the Parent column header, followed by the addition of
- Labels
For example:Parent column header:
What would convince you to switch to a plant based burger?Labels column header:
What would convince you to switch to a plant based burger? - Labels
ID | Date | What would convince you to switch to a plant based burger? | What would convince you to switch to a plant based burger? - Labels |
502cjl5… | 2020-02-28 | I would switch if there are healthier ingredients. |
|
6a1upc… | 2020-03-01 | Taste |
|
2️⃣ Specify the Pre-defined Labels
In the new labels column, add the relevant labels for each data entry, separated by commas.
One or multiple labels can be added as needed.
⚠️ Important: All labels must be in lower case to ensure accurate processing by the platform.
Supported ✅:
health,ingredients– lower caseNot Supported ❌:
Health,Ingredients– capitalized, which does not meet the required format.Not Supported ❌:
HEALTH,INGREDIENTS– capitalized, which does not meet the required format.
Example row in the CSV:
ID | Date | What would convince you to switch to a plant based burger? | What would convince you to switch to a plant based burger? - Labels |
502cjl5… | 2020-02-28 | I would switch if there are healthier ingredients. | health,ingredients |
6a1upc… | 2020-03-01 | Taste | taste |
3️⃣ Add a Configuration Row
In the CSV file, a new empty row must be added in the second row to accommodate a configuration line that informs the platform about which Parent column the labels in the - Labels column relate to.
ID | Date | What would convince you to switch to a plant based burger? | What would convince you to switch to a plant based burger? - Labels |
|
|
|
|
502cjl5… | 2020-02-28 | I am concerned that they are not that healthy for you, nor are they cheap. | health,price |
6a1upc… | 2020-03-01 | Taste | taste |
In this new second row, insert the following configuration line in the - Labels column using the exact format as shown below:
Example configuration row:
{"col":"What would convince you to switch to a plant based burger?"}ID | Date | What would convince you to switch to a plant based burger? | What would convince you to switch to a plant based burger? - Labels |
|
|
| {"col": "What would convince you to switch to a plant based burger?"} |
502cjl5… | 2020-02-28 | I am concerned that they are not that healthy for you, nor are they cheap. | health,price |
6a1upc… | 2020-03-01 | Taste | taste |
⚠️ Important:
The formatting of the configuration line must be matched exactly—including the spacing, symbols, and overall structure.
Only the text within the quotes (in this example:
"What would convince you to switch to a plant based burger?") should be swapped out for the Parent column header relevant to your dataset.The text you swap in must match your actual Parent column header exactly—character for character, including any punctuation and spacing.
Do not change any of the surrounding structure
{"col":"..."}, as the platform uses this precise format to recognize the labels column’s association with the parent data column.
This step is crucial for ensuring the platform accurately links the uploaded labels to the correct data entries.
4️⃣ Repeat for Multiple Columns
Users can repeat Steps 1-3 for each Parent column that has pre-defined labels. Simply ensure that:
Each new labels column follows the same naming convention (
Parent Column Header - Labels).Each new labels column has its own configuration row entry, matched to the corresponding parent column exactly.
By repeating these steps, users can easily manage and label multiple columns of data within the same upload.
5️⃣ Save & Upload
Save the file in CSV format.
Upload the CSV file to the platform as usual.
The system will automatically detect and apply the labels during analysis.
Example CSV Format
Here’s a snippet of how the final CSV might look:
The Parent column appears in blue
The new
- Labelscolumn appears in green
ID | Date | What would convince you to switch to a plant based burger? | What would convince you to switch to a plant based burger? - Labels |
|
|
| {"col": "What would convince you to switch to a plant based burger?"} |
502cjl5… | 2020-02-28 | I am concerned that they are not that healthy for you, nor are they cheap. | health,price |
6a1upc… | 2020-03-01 | Taste | taste |
Need help?
For any questions or help with file setup, our support team is available to assist!