A research project implimentation for automatic text summarization. AllSummarizer is considered as an extractive method; Each sentence is scored based on some creteria, reorder the most scored ones then extract the first relevant ones.
You can find more about the method in the paper:
| Using clustering and a modified classification algorithm for automatic text summarization
------------ | -------------
Authors: | Abdelkrime Aries (A), Houda Oufaida (A) and Omar Nouali (B)
Affiliation: | A: Ecole Nationale Supérieue d'Informatique (ESI), Algeria;
B: Ctr. de recherche sur l'Information Scientifique et Technique (CERIST), Algeria
Booktitle: | Document Recognition and Retrieval XX. Proceedings of the SPIE, Volume 8658.
Date: | February 4, 2013
Address: | Burlingame, California, USA
Publisher: | SPIE
Link: | http://dx.doi.org/10.1117/12.2004001
Also, the participation of the system at MultiLing 2015 workshop:
| AllSummarizer system at MultiLing 2015: Multilingual single and multi-document summarization ------------ | ------------- Authors: | Abdelkrime Aries, Djamel Eddine Zegour and Khaled Walid Hidouci Affiliation: | Ecole Nationale Supérieue d'Informatique (ESI), Algeria Booktitle: | Proceedings of the SIGDIAL 2015 Conference, pages 237–244, , . 2015 Date: | September 2-4, 2015 Address: | Prague, Czech Republic Publisher: | Association for Computational Linguistics Link: | http://www.aclweb.org/anthology/W15-4634
TODO: add a brief description since there is a link to the paper
TODO: add a detailed description about the assemblance with the plugins (LangPi project)
The code is released under Apache 2.0 license. For more details about this license, check LICENSE file