Abstract
Understanding changes in the mood and mental health of large populations is a challenge, with the need for large numbers of samples to uncover any regular patterns within the data. The use of data generated by online activities of healthy individuals offers the opportunity to perform such observations on the large scales and for the long periods that are required. Various studies have previously examined circadian fluctuations of mood in this way. In this study, we investigate seasonal fluctuations in mood and mental health by analyzing the access logs of Wikipedia pages and the content of Twitter in the UK over a period of four years. By using standard methods of Natural Language Processing, we extract daily indicators of negative affect, anxiety, anger and sadness from Twitter and compare this with the overall daily traffic to Wikipedia pages about mental health disorders. We show that both negative affect on Twitter and access to mental health pages on Wikipedia follow an annual cycle, both peaking during the winter months. Breaking this down into specific moods and pages, we find that peak access to the Wikipedia page for Seasonal Affective Disorder coincides with the peak period for the sadness indicator in Twitter content, with both most over-expressed in November and December. A period of heightened anger and anxiety on Twitter partly overlaps with increased information seeking about stress, panic and eating disorders on Wikipedia in the late winter and early spring. Finally, we compare Twitter mood indicators with various weather time series, finding that negative affect and anger can be partially explained in terms of the climatic temperature and photoperiod, sadness can be partially explained by the photoperiod and the perceived change in the photoperiod, while anxiety is partially explained by the level of precipitation. Using these multiple sources of data allows us to have access to inexpensive, although indirect, information about collective variations in mood over long periods of time, in turn helping us to begin to separate out the various possible causes of these fluctuations.
| Original language | English |
|---|---|
| Title of host publication | 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW 2016) |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| Number of pages | 7 |
| ISBN (Electronic) | 9781509059102 |
| ISBN (Print) | 9781509059119 |
| DOIs | |
| Publication status | E-pub ahead of print - 2 Feb 2017 |
| Event | ICDM 2016: IEEE International Conference on Data Mining - Barcelona, Spain Duration: 12 Dec 2016 → 15 Dec 2016 |
Conference
| Conference | ICDM 2016 |
|---|---|
| Country/Territory | Spain |
| City | Barcelona |
| Period | 12/12/16 → 15/12/16 |
Keywords
- Collective Mood
- Mental Health
- Social Media
- Wikipedia
- Discrete Fourier Transform
- Multiple Tests
- Seasonal Cycles
- Information Fusion
- Big Data
Access to Document
- Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/document/7836767/. Please refer to any applicable terms of use of the publisher.