International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 07 | July 2022
p-ISSN: 2395-0072
www.irjet.net
Forecasting COVID-19 using Polynomial Regression and Support Vector Machine Javeriya Bano Altaf Hussain1, S.S.Hatkar2 1Dept.
of Computer Science and Engineering, Shri Guru Gobind Singhji Institute of Engineering and Technology, Nanded, Maharashtra, India 2Associate Professor, Dept. of Computer Science and Engineering, Shri Guru Gobind Singhji Institute of Engineering and Technology, Nanded, Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Many lives are getting affected by COVID-19
daily. Machine learning always plays an important role in health care sectors. Many researchers have used different machine learning models for prediction of COVID-19. This paper uses two supervised machine learning models i.e., polynomial regression and support vector machine. These models can forecast for the next 20 days COVID-19 cases. The efficiency of polynomial regression is more than support vector machine.
This paper includes four sections. Section 1 is about the introduction. In section 1 covid-19 has been explained. Its damaging effects have been discussed. Symptoms of COVID19 are listed. Preventive measures have been suggested. COVID-19 vaccines’ importance is discussed. Section 2 is all about training, testing and models used. Section 3 mentions information obtained from dataset and result. Section 4 ends the paper. At the end it was revealed that polynomial regression is better than support vector machine.
Key Words: Forecasting, Covid-19, Supervised machine
2. TRAINING, TESTING AND MODELS USED
learning, Support regression
2.1 Supervised Machine Learning Model
vector
machine,
Polynomial
In supervised machine learning, models or machines are trained firstly. For training of machines, “labelled” training data is used. When training a machine is done, the machine is now able to predict the output [3]. This paper uses two models for prediction of COVID-19. They are Support Vector Machine model and Polynomial Regression model.
1.INTRODUCTION COVID-19 is a disease which spreads from infected person to healthy person. In December 2019, the first case was found in Wuhan, China. Schools, colleges, markets, offices, parks, gyms, etc. had been shut down. The disease had spread in every corner of the world. Therefore, it has been declared a pandemic. Many people died. Many people migrated because they lost their jobs. To prevent spread of disease various prevention methods are implemented. They are face coverings( usually done by taking help of masks), hand washing, social distancing( keeping distance between two people ), Older people are at a higher risk of getting infected. Many complications have been observed in people post recovery i.e., kidney failure, pneumonia, etc. To understand long term effects many studies, need to be done [1].
2.1.1 Support Vector Machine (SVM) The full form of SVM is Support Vector Machine. SVM can put two different kinds of data into its respective category by taking help of a line. With help of this line, we can easily put data into its respective category. The best line is also known as a hyperplane. When SVM creates a hyperplane, it also selects nearest points. These nearest cases are called support vectors. Therefore, the algorithm's name is Support Vector Machine. There are two different categories of data that are classified using a hyperplane [4].
Many COVID-19 vaccines have been developed, tested, approved, and distributed all around the world. World’s largest population is believed to be vaccinated very soon. Vaccines do not confirm that vaccinated people will not get infected. There had been cases where people who had taken 2 doses of vaccines also got infected. Other measures like physical distancing, use of face masks is also necessary even if a person gets vaccinated.
To understand SVM, we can take help from the following case. Consider a situation, a strange horse looks very similar to a donkey. A model can be created that can tell whether it is a donkey or horse. Firstly, the model is trained by taking thousands of images of donkeys and horses. Thus, the model understands different properties of donkeys and horses. After that, a strange horse can be tested using the created model. So, the created model draws a decision boundary between the data of donkey and horse and decides support vectors. With the help of support vectors, it will decide whether it is donkey or horse.
Tiredness, fever, loss of taste , and cough are the most repeated indications of COVID-19. Diarrhea, sore throat, red eyes, headache , rashes on skin, pains are less repeated indications [2]. One must drink a lot of fluids so that the body does not get dehydrated and take proper rest.
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 1971