Skip to main content

Text summarization (sentence extraction) module with simple HTTP API. (currently, supports Japanese only)

Project description

Text summarization (sentence extraction) module with simple HTTP API. (Currently supports Japanese only)


MIT License


Python 2.7.*

  • numpy
  • scipy
  • scikit-learn
  • networkx
  • cherrypy
  • MeCab or janome
  • pulp (if you use ILP-based method)

Quick start

pip install summpy
python -m summpy.server -h -p 8080

Input Parameters

  • text: text (utf-8)
  • algo: (optional)
    • lexrank: LexRank, a graph-based summarization (default)
    • clexrank: Continuous LexRank
    • divrank: (experimental) DivRank (Diverse Rank, graph-based method). Since DivRank aims to provide non-redundant and high coverage information, it is suitable for multi-document summarization.
    • mcp: ILP-based method. Extracts sentences in terms of Maximum Coverage Problem.

Hyper parameters for how many sentences are shown (optional)

  • sent_limit: number of sentences (only {lex,clex,div}rank)
  • char_limit: number of characters
  • imp_require: cumulative scores [0.0-1.0] (only {lex,clex,div}rank)


from (



Response (JSON format)

  summary: [
  debug_info: {}

Try with browser


Python API

Example (Continuous LexRank)

from summpy.lexrank import summarize

# ensure type(text) is unicode
sentences, debug_info = summarize(
    text, sent_limit=5, continuous=True, debug=True

for sent in sentences:
    print sent.strip().encode(encoding)

For further details, see main part of summpy/ or


  • G. Erkan and D. Radev. LexRank: graph-based lexical centrality as salience in text summarization. J. Artif. Int. Res. 22(1), pages 457-479, 2004. (link)
  • Q. Mei, J. Guo, and D. Radev. DivRank: the Interplay of Prestige and Diversity in Information Networks. In Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘10), pages 1009-1018, 2010. (link)
  • H. Takamura and M. Okumura. Text Summarization Model Based on Maximum Coverage Problem and its Variant. In Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics (EACL ‘09), pages 781-789, 2009. (link)

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for summpy, version 0.2.1
Filename, size File type Python version Upload date Hashes
Filename, size summpy-0.2.1.tar.gz (13.5 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page