Circadian Mood Variations in UK Twitter Content Over Four Years

A collection of over 800 million Twitter messages collected over 4 years in the United Kingdom, used to analyze daily and seasonal patterns in expressed emotions and fatigue. The study by Fabon Dzogang applied statistical and Fourier analysis to extract robust circadian signals for positive mood, negative mood, anger, sadness, and fatigue.

Use Cases

• Analyze circadian patterns of positive and negative mood signals extracted from over 800 million Twitter messages.
• Compare the stability of fatigue and anger cycles across seasons and weekend/weekday boundaries.
• Investigate the relationship between extracted anger and fatigue patterns and known circadian variations in plasma cortisol concentrations.
• Identify morning inflexion points in the time series of expressed emotions like sadness and positive mood.

Strengths

• Massive scale with over 800 million data points (Twitter messages).
• Longitudinal coverage spanning 4 years of data collection.
• Analysis focuses on specific, extracted emotional signals (positive mood, negative mood, anger, sadness, fatigue).
• Explicit geographic focus on the United Kingdom.

Limitations

• The underlying raw Twitter message data and extracted features are not directly described or available in the input.
• Analysis is based on a specific geographic region (UK), limiting generalizability.
• Data collection ended in 2020, making the findings potentially stale for current social media behavior.

Provenance

Source

Dryad digital repository.

Collection Method

Collection and analysis of Twitter messages (content not specified).

Time Range

4 years (specific dates not provided).

Freshness

Last updated 2020-06-24; no update frequency stated.

Geography

United Kingdom.