Prototype implementation of a keyword labeler used to perform the user study in my master thesis:
Text Classification with Limited Training Data
Author: Petr Laitoch
Advisor: RNDr. Jiří Hana, Ph.D.
Charles University
2021
An updated text of the thesis is available in ./thesis.pdf
.
User instructions are present in the appendix of the thesis.
We assume Python 3. Tested on Python 3.7.4 on both Mac and Linux.
# Setup
python -m venv keyword-labeler-venv
./keyword-labeler-venv/bin/pip install svd2vec pydantic nltk
# Run
./keyword-labeler-venv/bin/python run_keyword_labeler.py example-project
When the prompt loads, try to run some of the following example commands to get started:
print_keywords
add_keywords waiter
print_keywords_and_evaluate
bulk_all_keywords <Enter> helpful <CTRL_C>
bulk_all_keywords <Enter> helpful personable joked considerate