Your project collected a lot of input? Great! But now the real work only starts: analyzing all that text-based input and transforming it into insights worth sharing.

Our autotagging tool helps you do just that. Follow the steps below to make sure the time spent on processing is minimal and the value you get out maximal.

Important: Due to the nature of the data we work with, our automated predictions will never be 100% accurate. Therefore, consider the autotagging as a very good starting point, after which you can manually correct and complement. This already creates a significant efficiency gain.

Important1: For now, our tool only covers the analysis of inputs. We plan to expand the scope to surveys and comments soon. Do you have survey results or comments to analyze in the meantime? Feel free to reach out to karel@citizenlab.co, and we'll see if we can help you out.

Steps to take - Quick overview

  1. Autotag

  2. Complement & correct

  3. Export

  4. Visualize

1. Autotag

To understand what the input is about without having to read everything in detail first, autotagging helps you discover the topics and trends within the input.

First, you need to choose the topics and trends you want to match all input with. Luckily, you don't have to do that on your own. We detect keyphrases in the input and suggests those as potential topics or trends for further analysis. Pick the ones that seem relevant, add some that you've used before or fill in new ones if you have already an idea what the input is about.

Don't worry to make 'wrong' guesses, you can redo this analysis at any time in a matter of seconds.

Once you've decided which topics and trends to use as tags for further analysis, hit 'Auto-tag'. A second algorithm will now compare each tag with each idea and score them based on how related they seem. When the score reaches a certain threshold, that tag gets added to that idea. The end result? All or most of the selected ideas have now one or multiple tags added, which eases further analysis.

2. Complement & correct

Once the autotag process is completed, you can now manually go through the inputs and easily approve, reject and complement the machine's predictions.

With our keyboard shortcuts and an optimized layout, this manual process is as efficient as it could be.

3. Export

Once done, you can now export the inputs to Excel.

The export file will contain a column for each tag, with a 1 indicating that the tag has been assigned to the input of that row and a 0 indicating that the tag hasn't been assigned. Autotags that haven't been approved have a score between 0 and 1, indicating our confidence that that tag belongs to that input.

The export can be used as a starting point to create visuals for your report, or it can serve as an internal tool to browse your inputs in a more structured way.

  • Are you passing on the input to different teams or departments? They can now easily filter on those tags that are relevant to them.

  • Is the input a starting point for further discussions or participation? Prepare those sessions by looking at one (group of) tag(s) at a time.

4. Visualize

From the data in your export file, you can now create any type of data visualization that brings your core message across.

Some examples below:

An overview of the identified themes

A map with the spotted locations, by topic

Did this answer your question?