SURAKSHA KAVACH FLUTTER APP

ABSTRACT:-

            The sector is facing a major health crisis due to the rapidity of Coronavirus transmission (covid-19). According to international health organization rules, the strongest anti-covid-19 protective measure carries a mask in public places and Crowded places. It’s incredibly difficult to look at people with their hands in their Places. These days, RT-PCR is only one way to detect the COVID-19 infection, it is limited by the lack of time-consuming, and the need for specialized labs. However these 2 approaches cannot continuously be used for patients screening because of the radiation doses high costs and also the low variety of accessible devices hence there’s a requirement for a less costly and quicker diagnostic model to spot the positive and negative cases of covid-19.

So in this paper, we predict the vulnerability of peoples according to their age and states. In this paper, we used the four prediction models using four different classifiers (i.e Logistic regression, Naive byers, Random forest, SVM) for detecting the vulnerability of peoples from their age and state. We create all of four models with all the classifiers. The results showed that the Random forest classifier is the most accurate classifier for predicting the vulnerability of COVID-19 cases based on the age and state. The results could help in the early diagnosis of COVID-19, specifically when the RT-PCR kits are not sufficient for testing the infection and assist countries, specifically the developing ones that suffer from the shortage of RT-PCR tests and specialized laboratories. So this system is very helpful in COVID-19.

In this paper, no diagnostic model has been proposed to identify the positive and negative cases of COVID-19 using several clinical features. Therefore, this research aims to predict the COVID-19 positive or negative cases based on age and state using machine learning classification algorithms.

OBJECTIVES OF THE PROJECT:-

The objectives of the systems development and event management are:

• To get the meaning of vulnerability to older persons themselves.
• To detect the vulnerability of a person based on age and state by using machine learning and various classification algorithms.

EXISTING SYSTEM :-

Several machine classification algorithms have been used to identify the positive and negative aspects of COVID-19. Flow of mining process data. The procedure involves data purification, data modification, and application of machine learning algorithms in the fourteen clinical features of COVID-19 patients. In the first step, the actual patient data of COVID-19 is obtained for analysis. During the second step, a panel of experts, consisting of three medical doctors, reached a consensus on what factors should be involved in the prognostic model. However, the data collected contains some noise and needs to be cleaned, as it cannot be processed directly by machine learning algorithms. In this step, data purification is also used to correct data conflicts and noise removal. In the third step, to adjust the data for machine learning algorithms, data conversion is applied. In that case, the negative and positive COVID-19 cases were converted to 1 and 2, respectively. In the final step, the machine learning algorithms are applied to the final data to separate COVID-19 patients into positive or negative situations. Some of this section is divided into two sections. The first paragraph introduces the collected data for COVID-19. The second section describes the classification algorithms used in this study, as well as the findings.

PROPOSED SYSTEM :-

In this system, we predict the vulnerability of people based on age and state. For that we had the data preprocessing. In data preprocessing, we clean the data, remove missing values from the dataset. Data cleansing is one among the foremost vital components of machine learning. It plays a crucial role in modeling. never the most effective part of machine learning and at a similar time, there are not any hidden tricks or secrets which will be discovered. However, the success or failure of a project depends on cleanup of the relevant information. Specialist data scientists usually invest the majority of their time during this step. 

This includes deleting duplicate / redundant or unsuitable values from your dataset. Duplicate observations most often arise throughout data assortment and irrelevant observations are those who don’t really the particular those who drawback that you’re attempting to resolve. Fixing the errors that arise throughout measure, transfer of data, or different similar things are referred to as structural errors. Missing data could be a misleadingly difficult issue in machine learning. we cannot simply ignore or take away the missing observation. they need to be handled carefully as they will be a sign of something important. Use Label code in data and refers to converting labels into numerical formats to convert them into machine-readable form. Machine learning algorithms will then be determined during the advanced route but those labels should be used. It is a very important step for pre-processing in a systematic database in supervised reading.

This data is further divided into 2 parts. First part is the training dataset and secode is for testing. A machine learning model is a file that has been trained to recognize certain types of patterns. You train a model over a set of data, providing it an algorithm that it can use to reason over and learn from those data. Once you have trained the model, you can use it to reason over data that it hasn’t seen before, and make predictions about that data. We mainly used four different classifiers like Logistic regression, Naive byers, Random forest, SVM. But the random forest model has more accuracy than other models So we selected this model in our project.

MODULES:

• User Registration: User have to register to become a part of this flutter app.
• User Login: User have to login himself to check the vulnerability.
• Check vulnerability: User have to select the age and state and model predict the level of vulnerability and display to the user.
• Localization: User can shift to local language to the app.

ADVANTAGES:-

• Users check the vulnerability level before traveling in another state according to their age.
• We will ban the high vulnerability peoples to prevent them from COVID-19 infection.
• This model helps the government to utilize the rules according to the vulnerability level.

HARDWARE AND SOFTWARE REQUIREMENTS

HARDWARE:-

• Processor: Intel Core i3 or more.
• RAM: 4GB or more.
• Hard disk: 250 GB or more.

SOFTWARE:-

• Operating System : Windows 10, 7, 8.
• Python
• Anaconda
• Spyder, Jupyter notebook, Flask.
• MYSQL