Skip to content

mohdtalal3/ORB-Visualization-NLP

Repository files navigation

ORB: Visualization Platform

Description

ORB is a versatile visualization platform designed to facilitate dynamic exploration of data through interactive visual representations. Offering a range of visualization layouts such as icicle, chord, Tillford, Sugiyama, grid layout, and radial Sugiyama layout, it provides users with a user-friendly interface to interact with different visualization styles.

Additionally, ORB now integrates advanced natural language processing capabilities to introduce two innovative features: the Sentiment Network Graph and the Sentiment Dashboard.

The Sentiment Network Graph enables users to explore email interactions visually. It represents the connections between top senders and recipients, allowing users to dive into sentiment and emotion analysis within conversations.

The Sentiment Dashboard offers a streamlined approach to analyzing email conversations. Users can input sender and recipient email addresses to access sentiment analysis results. The dashboard presents extracted sentiments and associated keywords, enhancing users' understanding of the underlying emotions in their communications.

Video:

video.mp4

Dataset link

https://www.kaggle.com/datasets/wcukierski/enron-email-dataset/data

Installation

To install the dependencies, use the following command:

Usage

Run app.py to open the web view of the ORB platform. Upload your data files containing the required information for visualization. Interact with the visualizations to analyze and understand your data more effectively. [See the attached video of how the code works]

Project Structure

  • app.py: Contains the main code for the ORB platform.
  • templates/: Contains HTML templates for visualization.
  • preprocessing 1&2: Contains data cleaning and extensive NLP.
  • static/: Contains images and CSS files for styling.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published