Project Title Chapter Title

CHAPTER 6

TESTING AND CONCLUSION

6.1 Introduction
Software testing is defined as an activity to check whether the actual results
match the expected results and to ensure that the software system is Defect free.
It involves execution of a software component or system component to evaluate
one or more properties of interest. Software testing also helps to identify errors,
gaps or missing requirements in contrary to the actual requirements. It can be ei-
ther done manually or using automated tools. Some prefer saying Software testing
as a White Box and Black Box Testing. In our system we have done manual test-
ing as well as stress testing to check the breakpoint of the network. The manual
testing was done using selenium software while stress testing was done manually
with the help of hundreds of nodes that were rented from an online server.
The first Testing was done in the first module i.e. Data Preprocessing which is to
ensure that the dataset does not contain any missing value or unknown value.
The original CSV file is taken as input and data cleansing is performed success-
fully. The second and third testing is done in second module i.e. Feature Extrac-
tion to reduce the dimensionality of dataset . This chapter gives the outline of all
testing methods that are carried out to get a bug free system. Quality can be
achieved by testing the product using different techniques at different phases of
the project development. The purpose of testing is to discover errors. Testing is
the process of trying to discover every conceivable fault or weakness in a work
product. It provides a way to check the functionality of components sub-assem-
blies and/or a finished product. It is the process of exercising software with the in-
tent of ensuring that the Software system meets its requirements and user expec-
tations and does not fail in an unacceptable manner. There are various types of
test. Each test type addresses a specific testing requirement. Test Environment
Testing is an integral part of software development. Testing process certifies
whether the product that is developed compiles with the standards that it was de-
signed to.

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

6.2 Testing Methods

In this work, the system is trained with the dataset of covid and non-covid cough in the
form of audio samples which were collected from kaggle and virufy. The features of audio
samples were extracted using MFCC (Mel-Frequency Cepstral Coefficients). Deep learn-
ing model is used for the trained data. Fig. 4 displays the basic structure of deep learning
model. We used the extracted features from our collected database for network training
and testing. Training and testing phases are crucial for our system. The training and test-
ing were made using python, where TensorFlow has been utilized as the deep learning li-
brary. We propose a machine learning model that is deep learning to obtain the result
using dataset with MFCC as input.

Deep Learning Model

To evaluate our system, we use the accuracy and validation accuracy.

These results were obtained based on, namely, cough sounds. The
best accuracy is achieved for cough sound is 99.16% and validation
accuracy is 92.16%. Our system fetch faster and accurate COVID-19
detection using deep learning. Our analysis shows that we can rely in
the first place on collecting cough sounds to make a COVID-19 detec-
tion system. Figure 5 and 6 illustrates the accuracy and loss of the
trained model.

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Fig.5. The Plot of model accuracy

Fig.5. The Plot of model loss

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

CONCLUSION

The paper presented a new approach in classifying COVID19

cough samples that could be the first step in introducing
contactless identification of COVID-19 patients. Here the ma-
chine learning model uses deep learning to predict as well as
and give good accuracy of the proposed system. Given that the
work initiated is part of a large project to integrate machine
learning classification or COVID-19 recognition software, the re-
sults will be interesting from the point of view of the underlying
model choice.

Although the systems we describe require more rigorous valida-

tion on a larger data set, the results we present are promising
and indicate that screening for COVID-19 based on cough
sound classification is Feasibility. In addition, it can be applied
remotely, thus avoiding contact with medical personnel.

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Dept. of ISE 2021-22 Page No.

 Project Title Chapter Title

Dept. of ISE 2021-22 Page No.