TIMME: Twitter Ideology-detection via Multi-task Multi-relational Embedding (code & data)

Node Embeddings in Dynamic Graphs

Collect and process real time twitter data plotting various metrics like volume , proportion, sentiment. Analyze tweet node networks and map them geographically.

Temporal Walk Based Centrality Metric for Graph Streams

A gathering of digital methods recipes for research, teaching and collaborations from across the Public Data Lab.

Build a classifier to classify anti-national tweets from normal tweets based on the Khalistan movement.

Finding similarity index of a community in the twitter network using multiple views through inter-graph random walk kernel (IRWK).

🔍 This project is about searching the twitter for Soccer Prediction using popular keywords and applying sentiment analysis on the data.

twitter scrapper to scarpe the tweets of a particular word and save them in a csv file

Codes and data necessary to reproduce the results of the article K. Shyrokykh, M. Girnyk and L. Dellmuth, "Short Text Classification with Machine Learning in the Social Sciences: The Case of Climate Change on Twitter," submitted.

Utility python package for RG17 and UO17 Twitter tennis data sets.

A repository with Twitter data and tweet classifiers using several machine learning approaches to measure the accuracy and performance of the approaches on classifying tweets.🐧

Analysing Twitter data to identify urban issues

Twitter Sentiment Analysis

Under the broad research question: What are the impact of COVID-19 in ones life? Our team collected Twitter data to answer research hypothesis, producing models, and have a better understanding of the factors that influence one's reaction to the global pandemic.

R code, markdown document, and Gephi files used to create a social network analysis project examining three information operations removed from Twitter in 2021.

Twitter Semantic Data Mining for Harvey Hurricane Credible Event Detection

In this project, I have explored the world of tweet analysis in the case of two European universities: University of Tartu and Lund University.

A Bidirectional LSTM model to classify whether a given tweet talks about a real disaster or not. This was my project in "CSC 522: Automated Learning and Data Analysis" course at NC State University.