key: cord-0698468-4xq9zucn authors: Chire Saire, J. E.; Lemus-Martin, R. title: Analysis of internet trends related to medications for COVID-19in ten countries with the highest number of cases date: 2020-09-09 journal: nan DOI: 10.1101/2020.09.08.20190785 sha: d462151843328fd8df1cb10630df1028aac08396 doc_id: 698468 cord_uid: 4xq9zucn During the last months, the pandemic generated by coronavirus SARS-CoV-2 moved governments, research groups and health organizations to plan, test and execute health policies. At the beginning, the treatment was not clear but in the next months, research groups have conducted studies to recommend or ban some medicines. At the same time, countries with most cases were changing from Asia, Europe to America. Different treatments and medications have been tested and recommended in many countries. such as Hydroxycloroquine ,Ivermectin, Azithromycin, Dexamethasone, Prednisone, Desivir . This paper is a preliminary study about what people are searching on Internet, considering ten countries with most cases in the world such as Chile, Spain, United Kingdom, Brazil, United States, India, Russia, South Africa, Peru, Mexico. Since first reported in Wuhan, China, in late December 2019, the outbreak of the novel coronavirus now known as SARS-CoV-2 has spread globally. As of the second week of August 2020, the novel coronavirus disease pandemic has now reached over 21 million confirmed cases worldwide and approximately 762,000 deaths worldwide according to the Johns Hopkins University Coronavirus Resource Center. COVID-19 case numbers are alarming in both their volume and widening geographic scope but there are also concerns about the accuracy of reported COVID-19 case numbers, particularly at earlier stages of the pandemic and in the last months within the LatinAmerican region , and whether underreporting may have blurred the true extent of the outbreak from the beginning and its health and economical impact. There are many studies using Google Trends to analyze, forecast topics related to dengue (Anggraeni and Aristiani 2016) , elections (Polykalas, Prezerakos, and Konidaris 2013), programming (Chen and Xing 2016) and recently, studying covid-19 topic in Ghana (Chire and Panford-Quainoo 2020) , France (Chire and Oblitas Cruz Copyright c 2021, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. 2020) and South America Region (Chire and Mahmood 2020) . The contribution of this paper: • Study the interest of citizens through Google Search tool around medication for Covid-19 • Analyze the differences between trends in countries with high incidence of cases The presented paper follows a methodology inspired on Cross Industry Standard Process for Data Mining(CRISP-DM) (Shearer 2000) . The steps for the conducted study are detailed in the next subsections. The first question to answer came from the announcements of new vacuum for coronavirus, therefore the question is: what treatment or medication is most popular in population? By consequence, to set the scope of the study, countries with more cases were selected. The analysis considers the ten countries (see Fig. 1 ) with more infections. Figure 1: Ten countries with the highest total cases 1 The proposed terms are related to medication used in many countries, the terms are: Clhorine Dioxide, Ivermectin, Azithromycin, Dexamethasone, Prednisone, Desivir. These terms were selected a review of news from different countries about treatment against covid19. Trends using Google search can be useful to analyze how was the interest of one country/city around one term. Period of study: 01-07-2020 to 22-07-2020 and the terms were mentioned previously. After the collection of data from Google Trends, this step helps to present data and support analysis about covid19 and medication in Top 10 countries. Series plot to show the evolution from 01 to 22 July for the selected terms. The analysis is considering the next countries: Usa, Brazil, India, Russia, South Africa, Peru, Mexico, Chile, Spain, Uk. Therefore, they are representing North America, South America, Asia, Europe. Table 1 is presenting the population of every country, number of covid19 total cases and proportion of total infections. It is important to consider that each country has different numbers of population then to observe the total number of cases is appropiate to analyze the proportion of total infections over the total population. If the analysis only considers data related to total cases of Table 1 , then Usa, Brazil and India are the three nations with more infections. But, considering column Percentage of Table 1 , Usa, Peru and Brazil are on the top. At the same time Usa is not the most populated country neither Brazil or Peru. Performing the query using the terms mentioned previously in the section of the proposal, results are in Figure 2 . Rusian, India presents a constant interest for Azythromycin. Chlorine dioxide in Peru, Mexico and Spain. Brazil, Mexico have a decay of interest about Ivermectine and Peru an increasing number during the last week. Prednisone in United States has a constant interest. Remdesivir has global decay from 01 July and a constant value for India. Dexamethasone for Russia, Brazil, Mexico, Peru has a constant interest. In conclusion, it is possible identify the interest of users using Google Trends which considers the searches made in one country to present the data. Besides, citizens from the countries with more infections are constantly searching information about medication showing the worrying about the 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 Using Google Trend data in forecasting number of dengue fever cases with ARI-MAX method case study: Surabaya, Indonesia Towards Correlating Search on Google and Asking on Stack Overflow Hope Amid of a Pandemic: Is Psychological Distress Alleviating in South America while Coronavirus is still on Surge? An algorithm based on Google Trends' data for future prediction. Case study: German elections The CRISP-DM Model: The New Blueprint for Data Mining