PoolParty Extractor - Customize Configuration Files

11/09/2025

For stop word elimination, lemmatization of terms and rule-based person name recognition, the PoolParty Extractor (PPX) uses wordform lists, which you can customize.

PoolParty Extractor ships with defaults. If no customized configuration files are present, the defaults are used.

The word list for rule-based entity recognition contains first names.

Note

Defaults and custom configuration files are exclusive. If there is for instance a stop words language file for English (stopwords/en.txt), the stop word filter uses just this file ignoring the default.

As soon as a file gets loaded by an incoming request, the file is cached in memory. If the files are modified on disk, you need to restart the PoolParty server for the changes to take effect.

All configuration files will be read as UTF-8.

The format of stopword files is a simple list of words to be filtered:

stopwords/en.txt

...
the
their
theirs
them
themselves
then
thence
there
...

The format of wordform files is a tab separated list of two words, where the left word is an inflected form and the right is the lemma.

lemmatization/en.txt

...
biked     bike
bikers    biker
bikes     bike
bikeways  bikeway
...

The format of the person name file is a simple list of first names.

personNames.txt

...
Arbogast
Jayden
Jaromir
Leyla
Freia
Jolie
Marceau
Isaias
Isaiah
Francis
...

Customizing wordform maps in PoolParty allows you to modify how the extractor processes language by adding or removing specific wordform-lemma pairs. This is useful for correcting errors or tailoring the system to a specific taxonomy. For instance, you can prevent a word like "playing" from mapping to the lemma "play" if "play" is used exclusively for a different concept, such as a theatrical production.

To modify a wordform map, you must create a customization file named customization_xx.txt, where xx is the two-letter language code (e.g., customizaton_fr.txt for French). This file must contain commands to either add or delete wordform-lemma pairs.

The commands follow this format:

Delete <wordform> <lemma>: Removes an existing mapping.
Add <wordform> <lemma>: Adds a new mapping.

The following is an example for a possible customization_en.txt file:

# Delete a mapping to prevent an unwanted association
Delete apples Apple

# Add a new mapping to a specific term
Add governmental governments

# Delete an existing mapping that is causing ambiguity
Delete played play

Copy the customization file to the folder /opt/poolparty/config/extractor/lemmatization of the PoolParty server and restart PoolParty or call the api endpoint extractor/api/cache/clear to load the new configuration. To clear the cache via api, use the following curl command from any server:

curl -X POST --user <username>:<password> https://<server.poolparty.biz>/extractor/api/cache/clear

In this section: