Self-harm is a heterogeneous term used to describe intentional self-injury and/or overdose using drugs or poisons, fatal or non-fatal. It is also linked with a variety of other wider social, economic and health care effects. The public health burden of self-harm is on the rise and is increasingly being recognized at a national level, as well as rising rates of self-harm which seem to be on the rise in developed countries just like they are on the rise in developing countries, and within both settings alongside modernization and rapid urbanization. Therefore, it is important for general policymakers and practitioners in the public health field to know about the prevalence of self-harm in a country along with timely information to do prevention of problems and to mitigate potential risk. The vast majority of recent self-harm studies rely on traditional statistical analysis of observational data to estimate the likelihood of self-harm in a population. A significant proportion of the countries do not have availability of statistical data, as mandated for the objective of prediction on a national level, or do not have them at the required granularity level or require an extensive time-lapse. FAST (free and large scale data approach to understanding and changing self-harm) is a new computational paradigm investigated this project, which exploits free social media data to harvest huge amounts of data to investigate its potential although they come from freely available resources. In this section, we present the Case Study of Thailand The model derived from the SIM method of FAST outperformed the traditional ARIA benchmark with an average improvement of 48% in MAPE for now casting and forecasting predictions using the framework proposed, as demonstrated in the experiments carried out in this case study.