Installation

Download

Installing Neji is simple:

• Download the latest distribution package
• Extract the compressed file
• Use ./neji.sh to annotate documents
• Use ./nejiTrain.sh to train new NER machine learning models
• Use ./nejiWeb.sh to create a new Web server

The distribution package includes:

• Shell scripts for using Neji
• GDep and OpenNLP tools and models
• NER model for Gene and Protein recognition
• Example dictionaries
• Example corpora to annotate and train

Please visit the documentation pages to use Neji as: CLI and SDK.

Maven

Neji can be included in your project as a Maven dependency.

Add the following repository:

<repository>
    <id>bioinformatics-all</id>
    <name>Bioinformatics Nexus (All)</name>
    <url>http://bioinformatics.ua.pt/maven/content/groups/public</url>
</repository>

Add the following dependency:

<dependency>
  <groupId>pt.ua.tm</groupId>
  <artifactId>neji</artifactId>
  <version>2.0</version>
</dependency>

Requirements

We recommend using the latest version of Linux or MacOS with Java 7 or later.

Neji runs on Windows but it performs slower, due to a C++ library dependency of GDep. If you do not use GDep, the performance of Neji on Windows is the same as on Linux or MacOS.