Categorization AI

Per project users can train categorization AIs. One document is assigned to one category of the project.

Users can activate one categorization AI per project to assign a document to a category automatically.

Categorization AI Detail


The project used to train the categorization AI.


Statuses of an AI training.

  • “Queuing for training…”: The categorization AI is waiting in the queue for its training to be started.

  • “Data loading in progress…”: The training process has started and the Konfuzio server loads the training data into memory.

  • “AI training in progress…”: The training data is loaded into memory and the actual training takes place.

  • “AI evaluation in progress…”: The categorization AI is trained and the evaluation of the trained categorization AI is beeing conducted.”

  • “Training finished.”: The categorization AI is evaluated and can be used.

In case the categorization AI could not be trained it will have the status “Contact support”.


The description to document the reason for training.


Incremented version per training

AI Parameter

Saved status of when training started.

Created At

Date and time when training was started.

Train categorization AI

The training process is 100 % automated, so the only setup users need add a short description. The short description will help to relate the intention behind any change in the project to the quality of the categorization AI.

In order to train a Categorization AI, documents from at least two different categories need to be in the training set.

To improve the quality of a categorization AI make sure to use only documents which relate to one category. If you have several categories in one file split those files before you upload them.

Retrain categorization AI

If you have new documents uploaded to your project you can train a new version of your categorization AI.

  1. Add those to the Status: Training documents

  2. Train categorization AI, see above

  3. As you use the same documents with Status: Test documents but increased the number of documents with Status: Training documents the AI quality should improve.

  4. Read more about how to Improve Extraction AI to improve your categorization AI even further.

Splitting a pdf containing a batch of scanned documents

We provide the functionality to split stacked scans on request.

Categorization AI actions

Evaluate categorization AIs

If you change the documents assigned to the status test dataset you can evaluate old categorization AIs. This is helpful to evaluate different categorization AIs on the current test dataset.