Correcting Typos - Text Preparation Plugin?

Rhizom

Hi everyone,

Context: I have data from a survey. One question in the survey is multiple choice and answers predefined, but there is no data validation built into the survey. As a result, I have typos in the data. For example, a column "Genre" can include "Rcok, Clasisc, Jaz".

Question: Is there are smart/quick way to correct spelling mistakes in Dataiku?

Attempted solution: I tried using the Text Preparation Plugin but I get the following error: Error in Python process: At line 4: <class 'ModuleNotFoundError'>: No module named 'regex'. Could this be related to the Code environment (I'm using python 3.11 but the text preparation plugin seems to work with version 3.6 or 3.7.)?

Thanks!

Alexandru

Hi,

The plugin will work with the python mentioned here: https://github.com/dataiku/dss-plugin-nlp-preparation/blob/main/code-env/python/desc.json
Curently : "PYTHON36", "PYTHON37", "PYTHON38", "PYTHON39"
Could you try with one these python and see if it works for you?

Thanks