The following video provides more detail about the graph-based approaches used in TextRank, as well as trade-offs, benchmarks, comparisons with other algorithms, and so on:
"Single and Multiple Document Summarization with Graph-based Ranking Algorithms"
Microsoft Research on YouTube (2016-09-05)
For a general overview of this Python library and its usage, see:
"Graph-Based Data Science"
For related course materials and training, please check for calendar updates in the article "Natural Language Processing in Python".
In general, see listings on Google Scholar for citations of this work.
"Knowledge Graph-based Core Concept Identification in Learning Resources"
Rubén Manrique, Christian Grevisse, Olga Mariño, Steffen Rothkugel
8th Joint International Conference, JIST 2018 (2018-11-26)
"Use TextRank to Extract Most Important Sentences in Article"
The Artificial Impostor (2018-12-09)
"Automatic Keyword extraction using TextRank in Python"
Practical Data Analysis: Using Python & Open Source Technology
Dhiraj Bhuyan (2018-07)
"Beyond bag of words: Using PyTextRank to find Phrases and Summarize text"
"Text Summarization in Python: Extractive vs. Abstractive techniques revisited"
Pranay Mathur, Aman Gill, Aayush Yadav
RaRe Technologies (2017-04-05)
spaCy pipeline version of PyTextRank was originally written
very late one evening near the beach in A Coruña, Galica, while the
author was teaching an NLP seminar at the university during
Big Data Coruña 2019.
What a beautiful city, there within ancient Celtic lands, guarded by the Tower of Hercules. It's more than a bit reminiscent of Santa Cruz, California – or perhaps the other way around.