ALvisNLP/ML corpus processing engine
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
alvisnlp-bibliome
alvisnlp-core
alvisnlp-rest
alvisnlp-test
doc
docs
share
.gitattributes
.gitignore
LICENSE
NOTICE
README.md
WINDOWS.md
install.bat
install.ps1
install.sh
pom.xml
test-alvisnlp.sh

README.md

Description

AlvisNLP/ML is a configurable NLP batch processing pipeline. It annotates text documents for the semantic annotation of textual documents. It integrates Natural Language Processing (NLP) tools for sentence and word segmentation, named-entity recognition, term analysis, semantic typing and relation extraction. These tools rely on resources such as terminologies or ontologies for the adaptation to the application domain. Alvis NLP/ML contains several tools for (semi)-automatic acquisition of these resources, using Machine Learning (ML) techniques. New components can be easily integrated into the pipeline. Part of this work has been funded by the European project Alvis and the French project Quaero.

AlvisNLP/ML is held by the Bibliome group at Inra Jouy-en-Josas, France

See

Please contact Robert Bossy if you have any questions.

Foreword

This file contains instructions to download, compile AlvisNLP/ML, install the command-line interface, and deploy the web application.

The instructions assume you are running on a Unix system with a shell. If you are running on Windows, then check the instructions in the WINDOWS.md file.

Prerequisites

  • git
  • Java >= 8
  • Maven >= 3.0.5

Download

git clone https://github.com/Bibliome/alvisnlp

Compile and build

cd alvisnlp

mvn clean install

Command-line interface

Install

Host-specific parameter values

We recommend that you set default parameter values for your host. These parameter values avoid to set parameters in plans for external tools.

cp share/default-param-values.xml.template share/default-param-values.xml

This will create a standard default parameter file in share/default-param-values.xml. Edit this file and fill parameter values appropriate for your host.

Default command-line options

If you wish your installed AlvisNLP/ML to run with default command-line options, then you put them on the file named default-options.txt in the share directory. Once installed, the options will be automatically prepended to each invocation of alvisnlp.

In the share directory, there is an example file named default-options.txt.template.

Copy files to installation directory

./install.sh DIR

DIR is the base directory of your AlvisNLP/ML install. This directory must exist. Launch this script as root if necessary.

Running AlvisNLP/ML

DIR/bin/alvisnlp -help

DIR is the base directory of your AlvisNLP/ML install. You migh also add the DIR/bin sub-directory to your PATH environment variable.

export PATH=DIR/bin:$PATH

Web service

Deploy

Deploy the the alvisnlp-rest/target/alvisnlp-rest.war file in your favourite application container.

For instance, on glassfish, run:

asadmin deploy --contextroot CONTEXT --name NAME alvisnlp-rest/target/alvisnlp-rest.war

Set context parameters

Set the following context parameters:

Variable Description
alvisnlp.url-base Absolute URL of the deployed AlvisNLP/ML application.
It should usually be the URL of the container cocatenated with the application context root.
alvisnlp.processing-dir Directory where the data for each run will be stored.
alvisnlp.plan-dir Directory where exposed plans are found.
alvisnlp.resource-dir Directory where to find resources used in plans.
alvisnlp.executor-class Fully qualified name of the class that launches runs.
Default: fr.inra.maiage.bibliome.alvisnlp.web.executor.ThreadExecutor, executes each run in a separate thread on the same server.

Jetty Maven Plugin

You can quick-test the Web Service with the Jetty Maven Plugin:

mvn jetty:run-war

Use it

From a browser open the URL of the AlvisNLP/ML application.