Twitter Sentiment Dynamics and Network Structure Analysis

A collection of Twitter data used to study the relationship between user sentiment and the evolving network structure created by @-mentions. It applies three sentiment scoring algorithms, including SentiStrength, to analyze collective emotions. The research was authored by Nathaniel Charlton and last updated in June 2020.

Use Cases

• Analyze the relationship between dynamic centrality measures and the usage of positive and negative sentiment scores among users.
• Track sentiment level stability within structurally stable Twitter communities over time and correlate sudden changes with external events.
• Calibrate an agent-based model using empirical sentiment measures to simulate collective emotive responses.
• Study the dynamics of collective emotions by examining sentiment scores applied to a large corpus of tweets.

Strengths

• Applies three distinct sentiment scoring algorithms, including the open-source SentiStrength program.
• Research findings are based on a large dataset of tweets, providing a substantial empirical basis.
• Dataset is associated with published research making three specific contributions regarding sentiment and network dynamics.

Limitations

• Specific dataset size, row count, column structure, and file formats are unknown, limiting reproducibility.
• Data is from 2020 or earlier, which may not reflect current Twitter platform dynamics or user behavior.
• Lack of sample data or column definitions prevents detailed understanding of the data's granularity and variables.

Provenance

Source

Dryad digital repository.

Collection Method

Data gathered from Twitter and analyzed using sentiment scoring algorithms.

Freshness

Last updated on 2020-06-24.