tf-idf, short for term frequency–inverse document frequency, is a numerical statistic that is intended to reflect how important a word is to a document in a collection or corpus.

The purpose of this project is to implement tf-idf, input given a set of files with a specific relationship, and output the tf-idf weight value of each file. Specifically, the "word" with the highest k is displayed and its weight value, as shown in the figure above. Alternatively, you can enter a word and output a weight value for that word in all files.

English can be segmented by blanks, but Chinese cannot. So we used Jieba Chinese text segmentation to collect the corpus of word. The word weighting value is then obtained using the tf-idf algorithm.

In fact, jieba also has built-in "keyword extraction based on td-idf algorithm", but according to its source code, jieba actually only reads one file to calculate TF. The IDF part reads their own custom corpus, so the result is not accurate (not based on the set of related files to calculate the inverse frequency). Specifically, you can try simple tf-idf jieba version here.

• Python 3
• jieba

git clone https://github.com/Jasonnor/tf-idf-python.git
cd tf-idf-python/src/
python -u tf_idf.py

python -u main_gui.py

Please feel free to open issues or submit pull requests.

• Tf-idf. (n.d.). in Wikipedia - https://zh.wikipedia.org/wiki/Tf-idf
• jieba - https://github.com/fxsjy/jieba

tf–idf-python is released under the MIT License. See the LICENSE file for details.