Analyzing the Trend and Forecasting of Covid-19 Outbreak Using Machine Learning Technique

Benie Bebela, Suboh Alkhushayni, CIS, MNSU, Mnakato, MN, 56001

This global pandemic has impacted the world and most particularly the healthcare system in an unprecedented way. Relying on old prevention methods is not an option. This project can revolutionize how prevention is made for the covid-19 by making accurate predictions. Its main goal is analyzing large set of data and through computerized ways making accurate predictions that will help advancing into mitigating the impact of the pandemic. Through Machine learning technique, intelligent systems will be able to identify the patterns in the evolution of the virus and help keep track of the numbers of active cases, deaths, and recoveries. It is not only limited to the US or states but it will be able to analyze worldwide data and give real time accuracy in information and prediction by calculating such numbers as the mortality and recovery rates. This will help make predictions for periods of as long as 90 days and through patterns determines common factors that can make certain individuals more susceptible than others. 

• Gathering worldwide data related to the pandemic
 • Analysis phase:
a. Designing charts, graphs for visualization of results.
b. Using polynomial regression(used to find relationship between data in order to draw the most accurate result line on     graphs) and SVM learning model in the python programming language for predictions which basic purposes are to analyze data for used classification and regression analysis. This will also help in comparing data. 
c. Having our covid-19 dataset, analyzing factors such as mortality and recovery rate that globally and individually for  certain countries states if needed.

Anticipated outcomes
The main outcome is to have a clear view of the present situation in term of deaths, case and recoveries through the data visualization models that we will build. We will compare different countries and states result.

Additional Abstract Information

Presenter: Benie Bebela

Institution: Minnesota State University, Mankato

Type: Poster

Subject: Computer Science

Status: Approved

Time and Location

Session: Poster 5
Date/Time: Tue 12:30pm-1:30pm
Session Number: 4051