Coreference

This Gem provides coreference resolution for various languages (en, es, it, nl, de, fr).

The CorefGraph module provides an implementation of the Multi-Sieve Pass system for for Coreference Resolution system originally proposed by the Stanford NLP Group (Lee et al., 2013). This system proposes a number of deterministic passes, ranging from high precision to higher recall, each dealing with a different manner in which coreference manifests itself in running text.

Although more sieves are available, in order to facilitate the integration of the coreference system for the 6 languages of OpeNER we have included here 4 sieves: Exact String Matching, Precise Constructs, Strict Head Match and Pronoun Match (the sieve nomenclature follows Lee et al (2013). Furthermore, as it has been reported, this sieves are responsible for most of the performance in the Stanford system.

The implementation is a result of a collaboration between the IXA NLP http://ixa.si.ehu.es and LinguaMedia Groups http://linguamedia.deusto.es.

Confused by some terminology?

This software is part of a larger collection of natural language processing tools known as “the OpeNER project”. You can find more information about the project at the OpeNER portal. There you can also find references to terms like KAF (an XML standard to represent linguistic annotations in texts), component, cores, scenario’s and pipelines.

Quick Use Example

Installing the coreference can be done by executing:

gem install opener-coreference

Please bare in mind that all components in OpeNER take KAF as an input and output KAF by default.

Command line interface

You should now be able to call the coreference as a regular shell command: by its name. Once installed the gem normalyl sits in your path so you can call it directly from anywhere.

This aplication reads a text from standard input in order to identify the language.

cat some_kind_of_kaf_file.kaf | coreference

This will output:

<coref coid="co5">
  <!--Her father-->
  <span>
    <target id="t45"/>
    <target id="t46"/>
  </span>
  <!--him-->
  <span>
    <target id="t58"/>
  </span>
  <!--his-->
  <span>
    <target id="t54"/>
  </span>
</coref>

Webservices

You can launch a language identification webservice by executing:

coreference-server

This will launch a mini webserver with the webservice. It defaults to port 9292, so you can access it at http://localhost:9292.

To launch it on a different port provide the -p [port-number] option like this:

coreference-server -p 1234

It then launches at http://localhost:1234

Documentation on the Webservice is provided by surfing to the urls provided above. For more information on how to launch a webservice run the command with the -h option.

Daemon

Last but not least the coreference comes shipped with a daemon that can read jobs (and write) jobs to and from Amazon SQS queues. For more information type:

coreference-daemon -h

Description of dependencies

This component runs best if you run it in an environment suited for OpeNER components. You can find an installation guide and helper tools in the OpeNER installer and an installation guide on the Opener Website

At least you need the following system setup:

Depenencies for normal use:

  • Ruby (MRI) 1.9.3 or newer
  • Python 2.6

Adapting CorefGraph-en to your language

There are a number of changes needed to be made to make CorefGraph works for other languages. Although we have try to keep the language dependent features to a minimum, you will still need to create some dictionaries for your own language and make some very minor changes in the code. Here is the list of very file in the Corefgraph module that needs to be changed. Every change except one (see below) to be done in the $project/core/corefgraph/resources directory:

  • dictionaries/$lang_determiners.py
  • dictionaries/$lang_pronouns.py
  • dictionaries/$lang_verbs.py
  • dictionaries/$lang_stopwords.py
  • dictionaries/$lang_temporals.py

  • tagset/$TAGSETNAME_pos.py
  • tagset/$TAGSETNAME_constituent.py

  • files/animate/$lang.animate.txt
  • files/animate/$lang.inanimate.txt

  • files/demonym/$lang.txt

  • files/gender/$lang.male.unigrams.txt
  • files/gender/$lang.female.unigrams.txt
  • files/gender/$lang.neutral.unigrams.txt
  • files/gender/$lang.namegender.combine.txt
  • files/gender/$lang.gender.data

  • files/number/$lang.plural.unigrams.txt
  • files/number/$lang.singular.unigrams.txt

The Core

The component is a fat wrapper around the actual language technology core. You can find the core technolies in the following repositories:

Where to go from here

Report problem/Get help

If you encounter problems, please email support@opener-project.eu or leave an issue in the issue tracker.

Contributing

  1. Fork it http://github.com/opener-project/coreference/fork
  2. Create your feature branch (git checkout -b my-new-feature)
  3. Commit your changes (git commit -am 'Add some feature')
  4. Push to the branch (git push origin my-new-feature)
  5. Create new Pull Request

Reference

Command Line Interface

cat some_kind_of_kaf_file.kaf | coreference
<coref coid="co5">
  <!--Her father-->
  <span>
    <target id="t45"/>
    <target id="t46"/>
  </span>
  <!--him-->
  <span>
    <target id="t58"/>
  </span>
  <!--his-->
  <span>
    <target id="t54"/>
  </span>
</coref>

Webservice

You can launch a webservice by executing:

coreference-server

After launching the server, you can reach the webservice at http://localhost:9292.

The webservice takes several options that get passed along to Puma, the webserver used by the component. The options are:

    -b, --bind URI                   URI to bind to (tcp://, unix://, ssl://)
    -C, --config PATH                Load PATH as a config file
        --control URL                The bind url to use for the control server
                                     Use 'auto' to use temp unix server
        --control-token TOKEN        The token to use as authentication for the control server
    -d, --daemon                     Daemonize the server into the background
        --debug                      Log lowlevel debugging information
        --dir DIR                    Change to DIR before starting
    -e, --environment ENVIRONMENT    The environment to run the Rack app on (default development)
    -I, --include PATH               Specify $LOAD_PATH directories
    -p, --port PORT                  Define the TCP port to bind to
                                     Use -b for more advanced options
        --pidfile PATH               Use PATH as a pidfile
        --preload                    Preload the app. Cluster mode only
        --prune-bundler              Prune out the bundler env if possible
    -q, --quiet                      Quiet down the output
    -R, --restart-cmd CMD            The puma command to run during a hot restart
                                     Default: inferred
    -S, --state PATH                 Where to store the state details
    -t, --threads INT                min:max threads to use (default 0:16)
        --tcp-mode                   Run the app in raw TCP mode instead of HTTP mode
    -V, --version                    Print the version information
    -w, --workers COUNT              Activate cluster mode: How many worker processes to create
        --tag NAME                   Additional text to display in process listing
    -h, --help                       Show help

Daemon

The daemon has the default OpeNER daemon options. Being:

Usage: coreference-daemon <start|stop|restart> [options]

When calling coreference without <start|stop|restart> the daemon will start as a foreground process

Daemon options:
    -i, --input QUEUE_NAME           Input queue name
    -o, --output QUEUE_NAME          Output queue name
        --batch-size COUNT           Request x messages at once where x is between 1 and 10
        --buffer-size COUNT          Size of input and output buffer. Defaults to 4 * batch-size
        --sleep-interval SECONDS     The interval to sleep when the queue is empty (seconds)
    -r, --readers COUNT              number of reader threads
    -w, --workers COUNT              number of worker thread
    -p, --writers COUNT              number of writer / pusher threads
    -l, --logfile, --log FILENAME    Filename and path of logfile. Defaults to STDOUT
    -P, --pidfile, --pid FILENAME    Filename and path of pidfile. Defaults to /var/run/tokenizer.pid
        --pidpath DIRNAME            Directory where to put the PID file. Is Overwritten by --pid if that option is present
        --debug                      Turn on debug log level
        --relentless                 Be relentless, fail fast, fail hard, do not continue processing when encountering component errors

Environment Variables

These daemons make use of Amazon SQS queues and other Amazon services. The access to these services and other environment variables can be configured using a .opener-daemons-env file in the home directory of the current user.

It is also possible to provide the environment variables directly to the deamon.

For example:

AWS_REGION='eu-west-1' coreference start [other options]

We advise to have the following environment variables available:

  • AWS_ACCESS_KEY_ID
  • AWS_SECRET_ACCESS_KEY
  • AWS_REGION

Languages

  • Dutch (nl)
  • English (en)
  • French (fr)
  • German (de)
  • Italian (it)
  • Spanish (es)
This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 261712.