API Sentiment Analysis

A RESTFul API for Sentiment Analysis in Spanish.

Documentation

https://twittersentimentanalysis.github.io/API-Sentiment-Analysis/javadoc/index.html

Test

http://gessi-sw.essi.upc.edu:8080/sentiment-analysis-api/swagger-ui.html

How to run

Local

1. Clone this project to a local folder and go to root folder

   git clone https://github.com/twittersentimentanalysis/API-Sentiment-Analysis.git

2. Build the Spring Boot project with Maven

   mvn clean install

3. Run the project

   mvn spring-boot:run

4. Go to http://localhost:8080/swagger-ui.html and test the API

Tomcat

1. Clone this project to a local folder and go to root folder

   git clone https://github.com/twittersentimentanalysis/API-Sentiment-Analysis.git

2. Build the Spring Boot project with Maven to generate a WAR file

   mvn clean package

3. Run Tomcat server (you can download it here)

4. Go to http://localhost:8080/manager/html

5. Load generated WAR file into section Deploy > WAR file to deploy and click on Deploy

   By default generated WAR file is saved on target folder in the root project folder

6. Go to http://localhost:8080/sentiment-analysis-api/swagger-ui.html and test the API

**Note: File src/main/resources/config.properties can be modified at anytime to point to localhost or gessi endpoint.

Endpoints

Base URLs

Localhost

http://localhost:8080/api/

GESSI Server

http://gessi-sw.essi.upc.edu:8080/sentiment-analysis-api/

Path

Translator

translator?translator=Microsoft

Sentiment analysis

• BERT: emotion?tool=BERT
• BETO: emotion?tool=BETO
• SVC: emotion?tool=SVC
• ParallelDots: emotion?tool=ParallelDots

How to deploy

1. docker build -t api_sentiment_analyisis:latest .
2. docker run -d --name API_Sentiment_Analyisis -p 3005:3005 api_sentiment_analyisis:latest