Tweets_Classification_Webpage

The goal of this project is to be able to predict what rating customers on social media platforms would give to products. This enables businesses to better understand what customers think of their products as social media platforms such a Twitter and Youtube do not have rating systems.

This web application can search through Twitter and extract tweets which relate to a given keyword and classify the tweets into 5 categories. These categories represent ratings (out of 5) where 1 is bad and 5 is excellent. Ideally, the keywords should be products but, the webpage can also take in just about anything so long as people are talking about it on Twitter.

This web application utilizes a neural network and BERT (Bidirectional Encoder Representations for Transformers) to make the classifications of the tweets. The machine learning models are based on the Is Bigger Better? Text Classification using state-of-the-art BERT with limited Compute research paper by: Ayaz Nakhuda, David Ferris and Jastejpal Soora. This paper can be visted using this link: https://github.com/AyazNakhudaGitHub/BERT_Customer_Reviews_Classification/blob/main/Report_Group_24.pdf

Python, Django, Flask, HTML5 and CSS3 were mainly used.

To run this project locally one will need to:

Download all the files and maintain the directory structure.
Ensure that all neccessary modules are installed.
Download the machine learning models from this Google drive link: https://drive.google.com/drive/folders/1S1WCRVz2fr8oZd7EysAFK36Lhwab0Eb6?usp=sharing
Place the machine learning models into this directory for the API: BERT_API/test/ml_models

Get the credentials for access to the Twitter API and input them into the file sentiment_BERT_Web_Project/sentiment_BERT_Web_Project/views.py

Run the API as seen in the image below:

Type this command to get the wepage running: python manage.py runserver

Future plans to host this web application and the API on the Google Cloud Platform is currently in the works.

While a GIF is included, a video is provided to give a live demo:

BERT_Webpage.Demonstration.mp4

A webpage that utilizes machine learning to extract sentiments from tweets.

Related tags

Overview

Tweets_Classification_Webpage

Owner

Ayaz Nakhuda

Machine learning algorithms implementation

Management of exclusive GPU access for distributed machine learning workloads

Book Recommender System Using Sci-kit learn N-neighbours

Python Machine Learning Jupyter Notebooks (ML website)

Penguins species predictor app is used to classify penguins species created using python's scikit-learn, fastapi, numpy and joblib packages.

This repository contains the code to predict house price using Linear Regression Method

Adversarial Framework for (non-) Parametric Image Stylisation Mosaics

Dragonfly is an open source python library for scalable Bayesian optimisation.

This handbook accompanies the course: Machine Learning with Hung-Yi Lee

Distributed Tensorflow, Keras and PyTorch on Apache Spark/Flink & Ray

customer churn prediction prevention in telecom industry using machine learning and survival analysis

Model Validation Toolkit is a collection of tools to assist with validating machine learning models prior to deploying them to production and monitoring them after deployment to production.

Machine Learning Algorithms

Automated Machine Learning Pipeline for tabular data. Designed for predictive maintenance applications, failure identification, failure prediction, condition monitoring, etc.

A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models.

High performance implementation of Extreme Learning Machines (fast randomized neural networks).

A library of extension and helper modules for Python's data analysis and machine learning libraries.

Machine-care - A simple python script to take care of simple maintenance tasks

A pure-python implementation of the UpSet suite of visualisation methods by Lex, Gehlenborg et al.

DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions. DoWhy is based on a unified language for causal inference, combining causal graphical models and potential outcomes frameworks.