Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Implemented TextRank #54

Merged
merged 6 commits into from Oct 17, 2018
Merged

Implemented TextRank #54

merged 6 commits into from Oct 17, 2018

Commits on Oct 11, 2018

  1. Implemented TextRank 

    textrank.py is the method described in the paper, along with its post-processing. the paper does not describe a candidate selection happening before the TextRank algorithm.
pketextrank.py is an implementation of the core method described in the paper. The candidate selection is the one used in PKE to allow comparison between the weighting of the candidates.
    @ygorg
    ygorg committed Oct 11, 2018
    Copy the full SHA
    82fee0a View commit details
    Browse the repository at this point in the history

Commits on Oct 15, 2018

  1. Removed paper implementation 

    Keeping only the comparable version
    @ygorg
    ygorg committed Oct 15, 2018
    Copy the full SHA
    d05511d View commit details
    Browse the repository at this point in the history

  2. Renamed "pketextrank.py" to "textrank.py"

    @ygorg
    ygorg committed Oct 15, 2018
    Copy the full SHA
    27605bb View commit details
    Browse the repository at this point in the history

  3. Checked parameters used in paper

    @ygorg
    ygorg committed Oct 15, 2018
    Copy the full SHA
    c9807a3 View commit details
    Browse the repository at this point in the history

Commits on Oct 16, 2018

  1. Implementation is more accurate 

    The candidate generation detailed in the paper is used when `top` parameter is used in candidate_weighting
The graph creation is more accurate according to the paper
    @ygorg
    ygorg committed Oct 16, 2018
    Copy the full SHA
    bb74fc1 View commit details
    Browse the repository at this point in the history

  2. Added a warning if less than n best candidates are returned

    @ygorg
    ygorg committed Oct 16, 2018
    Copy the full SHA
    31f8a42 View commit details
    Browse the repository at this point in the history