Skip to content

Data visualization of historical and predicted data made with React and AWS.

Notifications You must be signed in to change notification settings

haingo-raz/AWS-Visualisation-Website

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

40 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NBA Sentiment Analysis and Prediction

Overview

This project displays visualizations of historical data, predictions, and sentiment analysis data for five NBA teams: Lakers, Celtics, Warriors, Rockets, and Bulls. The historical data has been collected using the NEWS API (articles) and the balldontlie API (past NBA scores). AWS has been used for making predictions (SageMaker) and performing sentiment analysis (Comprehend). The score predictions were made using the historical scores, while the sentiment analysis was made based on the tone of the articles about each team.

Features

  • Visualization of historical data.
  • Sentiment analysis for each team's related news articles.
  • Real-time prediction of each team's future scores.

Data Collection

Score data of the five NBA teams was collected using the Balldontlie API (score) and the News API (articles).

Sentiment Analysis

Amazon Comprehend was used to perform sentiment analysis on the news articles related to each team. The results are displayed on a pie chart, with each slice representing the ratio of positive, negative, and neutral news articles.

Prediction Model

Amazon SageMaker was used to train a logistic regression model on historical data for each team. The model predicts the scores of each team in upcoming matches. The results are displayed on a line chart.

Technologies Used

Tool Usage
ReactJS Frontend framework
Scss Style the frontend
Recharts Chart library for line charts and pie charts
TypeScript Provides static types and interfaces to ensure type safety
Python Used to create a code that separates JSON files into training and test sets
News API Used with TypeScript to download text data for sentiment analysis
balldontlie API Used to download past NBA scores
Amazon DynamoDB Database service used to store historical score, score predictions, and sentiment analysis data
Amazon SageMaker Used to generate predictions
Amazon Comprehend Used to perform sentiment analysis on the data downloaded from News API
Amazon S3 Storage service used to
Amazon Lambda Connect event together for data processing. Triggers sentiment analysis and web sockets update
Amazon Cloudwatch Used to monitor other AWS services
API Gateway Assign a connection ID to website visitors

Screenshots

Homepage

Homepage

Lakers score visualisation against time

Lakers score

Lakers score prediction against time

Lakers score prediction

Visualisation board for Houston Rockets

Houston Rockets Visualisation

Credits

API Usage
Balldontlie API Get score data of past NBA matches
News API Get news articles for sentiment analysis purposes

About

Data visualization of historical and predicted data made with React and AWS.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published