2016-02-10 15:35:26 +01:00
|
|
|
CitationExtractor
|
|
|
|
|
=================
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This is a simple web interface above
|
|
|
|
|
[libbmc](https://github.com/Phyks/libbmc/) and especially
|
|
|
|
|
[CERMINE](https://github.com/CeON/CERMINE) to extract references from a given
|
|
|
|
|
PDF file.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
## Installation
|
|
|
|
|
|
|
|
|
|
* Clone this repository.
|
|
|
|
|
* Install requirements: `pip install -r requirements.txt`
|
|
|
|
|
* Create a config file: `cp config.py.example config.py` and edit it according
|
|
|
|
|
to your needs.
|
|
|
|
|
* Run it: `python3 ./main.py`. Head to `http://localhost:8080` to see it live.
|
|
|
|
|
|
2016-02-11 15:44:26 +01:00
|
|
|
With the default configuration, it will forward uploaded papers to CERMINE
|
|
|
|
|
API, which is really slow and puts a lot of charge on these servers. You
|
|
|
|
|
should download an executable JAR from [their
|
|
|
|
|
repo](http://maven.icm.edu.pl/artifactory/simple/kdd-releases/pl/edu/icm/cermine/cermine-impl/)
|
|
|
|
|
to your server and indicate the path to this JAR file in the `config.py` file.
|
|
|
|
|
|
2016-02-10 15:35:26 +01:00
|
|
|
|
|
|
|
|
## LICENSE
|
|
|
|
|
|
|
|
|
|
This repo is under an MIT license.
|
