/Topic-Extraction

A Machine Learning NLP project to extract research topics based on publication titles using KeyBERT unsupervised keyword extraction approach.

Primary LanguageJupyter NotebookMIT LicenseMIT

Topic-Extraction

Overview

A Machine Learning NLP project to extract research topics based on publication titles using KeyBERT unsupervised keyword extraction approach.

Prerequisite

  • Anaconda 3
  • Python

Official Docs

Installation

Installing library to nlp preprocessing

pip install nltk 

Installing library to keyword extraction

pip install keybert
pip3 install keybert

Downloading nltk stopwords

If you don't have the nltk stopword data yet, you need to download the nltk stopword data as follows below.

import nltk
nltk.download()

Result

  • Raw Data

Screenshot 2023-01-22 at 04 47 00

  • Topic Result

Screenshot 2023-01-22 at 04 47 26