Skip to content

ancamariac/Twinvest

Repository files navigation

Finance tracker based on sentiment analysis using NLP

The objective of the project is to develop a web application based on a machine learning model for analyzing news in the economic domain and making relevant predictions. Users will have the ability to create an account where they can add their income and expenses on a monthly basis, monitoring their budget with the help of statistics. Additionally, they will have access to a news page that will provide selected data based on user preferences, which will be used to make predictions about the price movement of financial assets.

Visualizing the Structure of the Neural Network

Project Structure

Technologies Used

The technologies employed in this project are:

  • MERN Stack: Utilizing MongoDB for the database, Express.js for the server, React for the front-end, and Node.js for server-side scripting.
  • Machine Learning Model: BERT for tokenization and Feed-Forward like neural network - creating and training a machine learning model to analyze news data and make predictions.
  • Web Scraping: Implementing web scraping techniques to collect relevant data for predictions.

BERT paper - https://arxiv.org/pdf/1810.04805.pdf

Technologies Used

Results

Image 1 Image 2
Image 3 Image 4

Registration, Authentication, and Password Reset

Home Page

Viewing the Dashboard

Dashboard

Adding Incomes and Payments to Account

Incomes

Expenses

Settings Panel

Settings

Viewing Twitter Posts

Tweets

Viewing Tagged Google News

News

Viewing Predictions for Tracked Financial Assets

Predictions

Real-Time Evolution of Tracked Financial Assets

Realtime

About

Web application for financial data analysis, based on a machine learning classification model for news and social media posts, enabling price predictions

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors