Public Surveillance of Social Media for Suicide Using Advanced Deep Learning Models in Japan: Time Series Study From 2012 to 2022.

BACKGROUND: Social media platforms have been increasingly used to express suicidal thoughts, feelings, and acts, raising public concerns over time. A large body of literature has explored the suicide risks identified by people's expressions on social media. However, there is not enough evidence to conclude that social media provides public surveillance for suicide without aligning suicide risks detected on social media with actual suicidal behaviors. Corroborating this alignment is a crucial foundation for suicide prevention and intervention through social media and for estimating and predicting suicide in countries with no reliable suicide statistics. OBJECTIVE: This study aimed to corroborate whether the suicide risks identified on social media align with actual suicidal behaviors. This aim was achieved by tracking suicide risks detected by 62 million tweets posted in Japan over a 10-year period and assessing the locational and temporal alignment of such suicide risks with actual suicide behaviors recorded in national suicide statistics. METHODS: This study used a human-in-the-loop approach to identify suicide-risk tweets posted in Japan from January 2013 to December 2022. This approach involved keyword-filtered data mining, data scanning by human efforts, and data refinement via an advanced natural language processing model termed Bidirectional Encoder Representations from Transformers. The tweet-identified suicide risks were then compared with actual suicide records in both temporal and spatial dimensions to validate if they were statistically correlated. RESULTS: Twitter-identified suicide risks and actual suicide records were temporally correlated by month in the 10 years from 2013 to 2022 (correlation coefficient=0.533; P<.001); this correlation coefficient is higher at 0.652 when we advanced the Twitter-identified suicide risks 1 month earlier to compare with the actual suicide records. These 2 indicators were also spatially correlated by city with a correlation coefficient of 0.699 (P<.001) for the 10-year period. Among the 267 cities with the top quintile of suicide risks identified from both tweets and actual suicide records, 73.5% (n=196) of cities overlapped. In addition, Twitter-identified suicide risks were at a relatively lower level after midnight compared to a higher level in the afternoon, as well as a higher level on Sundays and Saturdays compared to weekdays. CONCLUSIONS: Social media platforms provide an anonymous space where people express their suicidal thoughts, ideation, and acts. Such expressions can serve as an alternative source to estimating and predicting suicide in countries without reliable suicide statistics. It can also provide real-time tracking of suicide risks, serving as an early warning for suicide. The identification of areas where suicide risks are highly concentrated is crucial for location-based mental health planning, enabling suicide prevention and intervention through social media in a spatially and temporally explicit manner.

Implementing spatial segregation measures in R.

Reliable and accurate estimation of residential segregation between population groups is important for understanding the extent of social cohesion and integration in our society. Although there have been considerable methodological advances in the measurement of segregation over the last several decades, the recently developed measures have not been widely used in the literature, in part due to their complex calculation. To address this problem, we have implemented several newly proposed segregation indices in R, an open source software environment for statistical computing and graphics, as a package called seg. Although there are already a few standalone applications and add-on packages that provide access to similar methods, our implementation has a number of advantages over the existing tools. First, our implementation is flexible in the sense that it provides detailed control over the calculation process with a wide range of input parameters. Most of the parameters have carefully chosen defaults, which perform acceptably in many situations, so less experienced users can also use the implemented functions without too much difficulty. Second, there is no need to export results to other software programs for further analysis. We provide coercion methods that enable the transformation of our output classes into general R classes, so the user can use thousands of standard and modern statistical techniques, which are already available in R, for the post-processing of the results. Third, our implementation does not require commercial software to operate, so it is accessible to a wider group of people.