Social Media Analysis for Situation Awareness during Crises (SMASAC) Tutorial

A script that finds all the bad people who don't follow you back

Webapp designed to analyze social media posts!

How to analyse social media data archives!

Real-time social media analytics

R scripts and example CSVs from my time as an intern and research assistant for the INSPIRE Project at the START Consortium in College Park, MD

Analysis of the Oct 2019 p-score dataset

This repository contains the source code of paper: "Semantic-based End-to-End Learning for Typhoon Intensity Prediction"

Exploration of BERT-BiLSTM models with Layer Aggregation (attention-based and capsule-routing-based) and Hidden-State Aggregation (attention-based and capsule-routing-based).

🍊 Find out how many bots follow any given Twitter acount

Implementation of Hadoop and Spark

#FakersGonnaFake: using simple statistical tools and machine learning to audit instagram accounts for authenticity

This is a case study of Reddit based on Big Data Analytics. To analyze the products review based on community site reddit to discover trends, sentiments, new opportunities and potential crises of such products to guide community users in decision-making.

To analyze the products review based on community site reddit to discover trends, sentiments, new opportunities and potential crises of such products to guide community users in decision-making.

A repo of my marketing analytics projects that cover digital attribution models and social media analytics.

Do a brand perception of a Twitter trend by extracting Twitter mentions for the brand and conduct text mining (Correlation, Frequency, Topic Modelling, Sentiment Analysis) on it.

A multiple linear model predicting lifetime engaged users on a cosmetic company's Facebook page. Data extracted from post metrics from over 500 posts in 2014. Achieved an accuracy of nearly 80%!

A support vector regression model predicting lifetime engaged users on a cosmetic company's Facebook page. Data extracted from post metrics from over 500 posts in 2014. Achieved an accuracy of over 80%!