Methods
This study investigated the correlation between the number of new cases of COVID-19 and the search index from Google. The aim of this study was to create an effective and affordable model to predict new cases, which would enable prompt and correct decision- making regarding public policies to limit the spread of COVID-19.
Google trend is known as a publicly available web-based tool which analyse relative search for a particular query. Google trend can provides the number of relative searches for a particular query within a particular time and a particular region or worldwide. Terms such as Coronavirus, handwashing, and face masks in Bahasa were collected from GT during the period of January 1st 2020 to march 31st , 2020. These search terms represented the information search for COVID-19 and practice of personal precautions to prevent disease transmission. Relative search volume (RSV) data were filtered by geographic regions in Indonesia. The data collected from Google Trends is adjusted to the time and location, so the comparisons between queries can be easier. These queries were searched from 2019/12/12 to 2020/04/04 within the seven provinces in Indonesia such DKI Jakarta, East Java, Yogyakarta Special Province, Central Java, West Java, Bali and South Sulawesi. These countries were selected because there were positive COVID-19 cases in these provinces in the beginning of the outbreak. Besides, since the RSV in the Indonesia and some provinces is less than “1” for several days, we treated it as RSV equal to “1” in these days for better reflecting the Indonesian and people in the selected provinces response to COVID-19 as conducted in the previous study (Hu et al., 2020). The results then being downloaded in the format of Common Separated Values, which displayed on a scale from 0 to 100. GT data then were compared with daily data on COVID-19 cases that were obtained from the website of COVID centre Ministry of Health Indonesia.