Watch Wolf (वॉच वुल्फ 🐺)

Track 4: Sentiment Analysis of social media posts

Website: https://watch-wolf.vercel.app/

Problem Statement & Expectation

People are only a click away from obtaining vast amounts of data. With information comes people's opinions, and with that comes people's positive and negative perspectives on a subject. This can lead to bullying and the spread of hateful remarks about someone or something.

The solution must be able to classify the sentiments that are expressed in the text source which is associated with that user on platforms such as Facebook, Instagram, and Twitter. It should accurately identify when the tone of the posts becomes inflammatory and hateful.

Solution

Watch Wolf is the web platform that analyze the sentiment and provides the sentiment and entities report for the text, images and PDF attachment. This is possible with hindi and english language. In addition to this features, the platform also provides the tweet analysis wherein the user can provide the tweet URL and the user is provided with the information of tweet, sentiment and entity report. Along with this, watch wolf provides a chrome extension whose major features includes injecting sentiment scores and mail service feature on the tweets directly observed on twitter.

Tech Stack

Front-end

Framework: NextJS
Styling: CSS, React-bootstrap
API: Fetch, Axios
Deployment: Vercel

Back-end

Framework: Spring Boot, Java
API: AWS (Comprehend, Rekognition), Twitter API, JUnit, Apache PDFBox
Deployment: AWS EC2

Chrome Extension

Framework: Vanilla Javascript
API: Chrome Extension Manifest V3

Setup on local computer

Clone this particular repo - git clone https://github.com/AemieJ/watch-wolf-client.git

Client-side

Open a terminal and perform the following steps:

Move into the client directory: cd client
Change from production ➡️ development environment by moving to config directory and performing following changes:
```
// server.js
const dev = "development";
```
```
// client.js
const dev = "development";
```
Open the terminal again within the client directory and include the following commands:
```
$ yarn && yarn dev
```
OR
```
$ npm install && npm run dev
```
Client side will be up and running on http://localhost:3000

Chrome Extension

Open chrome and type in chrome://extensions
Switch on the developer mode and select on load unpacked
Select the extension folder
For demo, you can move to particular hashtag on the official twitter website and observe the features of the extension