Semantic Assistants 2.0 - spring boot application

This spring boot application uses several NLP services for Named Entity Recognition tasks with two micro services.

The first micro service provides three GATE pipelines:

As a second microservice, a Python based application offers access to language models:

Huggingface Bert-base-NER extracting people, locations and organizations
SciSpacy with the en_core_sci_sm model recognizing biomedical data.

Demo

A demonstration is available: https://semsearch.fmi.uni-jena.de/SA2.0/swagger-ui.html

Prerequisites

JDK 11

Install Adopt Open JDK 11 (https://adoptopenjdk.net/)

Maven

Install Maven from Maven

Python

Python version 3.9

GATE pipelines

Please download the BiodivTagger and the OrganismTagger from Zenodo: .

The GATE ANNIE pipeline is default pipeline available in the GATE framework.

Installation and running

run below commands in order:

cd SA2.0/nlp

mvn spring-boot:run

Get started

It takes a little time until the application starts.

This project contains a Swagger API platform.

http://localhost:8080/swagger-ui.html#/

Settings

go to the:

src/main/resources/application.properties

set the Annie, BioTagger and OrganismTagger location on your system

PythonAPI

For using the Huggingface service, you need to run this app.

Prerequisites

Anaconda

Install Anaconda from Anaconda

Installation and running

run blow commands in order:

cd SA2.0/pythonAPI

python -m venv myenv

.\myenv\Scripts\activate

Linux: source myenv/bin/activate

pip install -r requirements.txt

python main.py