Sample CSV file
If you're trying Keatext for the first time and don't have any data to experiment with, you are welcome to use this sample CSV file. It contains reviews of rooms and apartments for rent and was built using data available on Inside Airbnb.
This section will guide you through the process of creating a dataset with that CSV file.
- Read Import CSV data and follow steps 1 to 7.
- Click No CSV file? to download the CSV file to your computer and make sure you save it in a location where it will be easy to find.
- Click Choose file and find the CSV file.
- Under Data Structure, make sure
Commais selected, because values are delimited by a comma.
Deciding which fields to check under the Analyze and Filter columns can seem tricky, so we'll guide you through it. The CSV file has 11 fields. Field names are fairly self-explanatory.
|Review ID||✖||✖||Should only be checked under Filter if you need the ID to retrieve the original document.|
|Review Date||✖||✔||Required to use a time series visualization.|
|Review Content||✔||✔||This field contains the unstructured text that needs to be analyzed. Can also be checked under Filter if you want to filter records based on the content of this field.|
|Listing Title||✖||✔||Even though this field contains text, we don't want to analyze it because the content is set by the listing owner, not by guests. It can still be used as a filter to find all reviews for a specific listing.|
|Neighbourhood||✖||✔||Can be used as a filter to find reviews for a specific neighbourhood.|
|City||✖||✔||Can be used as a filter to find reviews for a specific city.|
|State||✖||✔||Can be used as a filter to find reviews for a specific state.|
|Country||✖||✔||Can be used as a filter to find reviews for a specific country.|
|Room Type||✖||✔||Can be used as a filter to find reviews for a specific room type.|
|Room Price||✖||✔||Can be used as a filter to find reviews for listings in a specific price range.|
|Room Availability||✖||✔||Can be used to find reviews for listings that are available for a certain number of days throughout the year.|
In this case, we only wish to analyze one field containing unstructured text. We exclude the Listing Title field because a title like Clean, beautiful apartment is set by the listing owner and may not reflect the opinion of guests. A title picked by the reviewer, however, might be useful to you and should be analyzed when you import your own data.
Keatext can usually detect a field's type automatically, but you should always check the Type column for mistakes. If a date is detected as a string, for example, dates may be stored in an invalid format (see Supported date formats). Don't forget that dates must be valid in order to use time visualizations.
Once you're satisfied with the dataset configuration, click—easy peasy!