Using Google Search Data to assess the risks during pandemics
Floor Location : M 066 N
Around the end of December 2019, a new type of coronavirus (SARS-CoV-2), caused the COVID-19 pandemic. As a result, the dramatic changes in lifestyles have posed a threat to our mental health, exacerbating the suicide rates to new heights. Current systems implemented by google in the status quo are lacking in efficiency and effectiveness. By using autoregressive integrated moving average (ARIMA) models, I found that Google searches are a leading indicator of numbers of completed suicides and identified terms related to suicide. With the data gathered, I designed a program using python that takes a list of search records or a CSV file. The data is then filtered through three steps: data cleaning, data analysis, data distribution. In data cleaning, any invalid google searches that are missing a keyword or time is filtered out of the calculations. In data analysis, each suicide-related search is given a weight based of the time of the search. The weighted average was used to assess the risk of suicide. In data distribution, the result are displayed using pie charts. Finally, the program was tested using theoretical cases of search histories, each representing a different level of risk. The purpose is to assist with suicide prevention by alerting local suicide prevention helplines if a high risk of suicide is detected.