sumpy

SUMPY: a python automatic text summarization library

We currently have several baseline summarizers implemented:

[x] lede

[x] TextRank

[x] LexRank

[x] Centroid

[x] ROUGE ngram evaluation

and have plans to implement many more.

TODO:

[ ] ROUGE skip-gram and LCS evaluation

[ ] FreqSum/SumBasic

[ ] Submodular optimization based summarizers

[ ] lda/distributes sentence representation based summarizer

[ ] DEMS

[ ] ILP based summarizers

[ ] collect topic signatures/important word lists

SUMPY contains several ready to use summarizers with sensible defaults. Here is a simple example to get you started:

import sumpy

doc1 = "This is the text for document1. It is for explanatory purposes..."
doc2 = "This is another document text..."
doc3 = "And yet another document..."

inputs = [doc1, doc2, doc3]

print "lede summarizer:"
print sumpy.lede(inputs)

print "\ntextrank summarizer:"
print sumpy.textrank(inputs)

print "\ncentroid summarizer:"
print sumpy.centroid(inputs)