International Journal of Advanced Science and Technology

Vol. 29, No. 5, (2020), pp. 8944-8957

Effectiveness of a Lock-down to combat COVID-19 using

Regression Analysis: A Case Study in India

Jabeena A.1, Ananya K. R.2

1
Associate Professor, Vellore Institute of Technology, Vellore, Tamil Nadu, India
2
B.Tech Student, Vellore Institute of Technology, Vellore, Tamil Nadu, India
1
ajabeena@vit.ac.in , 2anumanipal99@gmail.com

Abstract
Ever since its outbreak in the city of Wuhan in China, COVID-19 has spread around the
world at a rapid rate and has now been identified in over 187 countries. To combat the
spread of the virus and to isolate the affected, one of the most significant measures taken
by various countries around the world is to enforce a complete lock-down across the
nation. This paper aims to examine the effectiveness of the one such lock-down enforced
in India by the Government on the 24th of March 2020 so as to curb the spread of COVID-
19. In order to be able to estimate the effectiveness, an approach of Regression Analysis,
which is a sub-branch of Predictive Analytics, has been applied using Python
Programming. Upon following this approach, it is observed that the 41-day long lock-
down has been highly effective in limiting the total number of COVID-19 positive cases in
India.
Keywords: Regression, Log-Linear Model, Statistical Analysis in Python, Coronavirus

1. Introduction
The Coronavirus represents a large family of viruses that is held responsible for causing a
gamut of illnesses which include the common cold, Middle East Respiratory Syndrome
(MERS-CoV) and Severe Acute Respiratory Syndrome (SARS-CoV). A novel
Coronavirus (nCoV) may be a new strain that has not been previously identified in
humans. In general, the human population has no acquired immunity to the new
Coronavirus and while most people who contract the virus will experience no or mild
symptoms, some people will develop severe or even life-threatening symptoms. As of 2nd
May 2020, more than 3.39 million cases of COVID-19 have been reported in 187
countries and territories, resulting in more than 241,000 deaths. More than 1.06 million
people have recovered. Owing to this, the Coronavirus has been declared as a global
pandemic by the World Health Organisation (WHO).
In India, the first positive case of COVID-19 was reported on 30th January 2020. Since
then, the number of cases has been rising constantly. In order to curb the spread of the
disease, the Government of India declared a nationwide lock-down on 24th March 2020,
initially for a span of 21 days i.e., up-to 14th April 2020, but subsequently extended the
period up-to 3rd May 2020 and later, with various relaxations, extended it to 17th May
2020.
The primary objective of this paper is to examine the effect of the enforced lock-down
(from 24th March 2020 to 3rd May 2020), for a period of 41 days in curbing the number of
COVID-19 positive cases in the India.

2. Methodology
In order to arrive at the effectiveness of the Lock-down enforced on 24th of March 2020,
the approach of Predictive Analytics has been put to use using Python Software.
Predictive Analytics is a statistical technique that involves studying and analysing the
current data to make predictions about the future through the application of various means
such as Data Collection, Data Modelling and Statistics, to name a few. It is an extremely

ISSN: 2005-4238 IJAST 8944

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

useful technique that can be applied in various spheres to get an estimate of the
behavioural trends and patterns of the data in the future.

This methodology of applying Predictive Analytics has been used to predict the total
number of COVID-19 positive cases had the lock-down not been enforced. This number
has then been compared to the total number of COVID-19 positive cases that were
recorded while the lock-down of 41 days, as ordered by the Government of India, was in
enforcement. This difference in the number of COVID-19 positive cases in both the above
stated scenarios will be used to arrive at an estimate of the effectiveness of the nation-
wide lock-down in India to combat the spread of the coronavirus.
The most crucial steps as followed in the paper to achieve the objective have been listed
as follows:
1. Data Collection
2. Statistical Modelling
3. Linear Regression Analysis
4. Transforming the data into Log-Linear Model
5. Construction of Best-Fit Line
6. Future Prediction
In the following sections, the above mentioned steps have been elaborated in great detail
with various mathematical calculations to supplement it. The analysis carried out in this
paper has been programmed in Python using the Numpy, Pandas, and Matplotlib Software
libraries.

2.1. Data Collection

Data Collection and Data Pre-Processing are the initial steps in any project or research
that involves modelling the data. In short, Data Collection can be defined as the
procedure that aims to gather and measure information on different interests in a
systematic approach, which allows one to draw conclusions to research-oriented
questions, to test various hypotheses, and to predict the outcomes. Accurate data
collection is extremely essential to maintaining the integrity of research. [1]
The primary data that is required in order to achieve this paper's objective is the total
number of COVID-19 positive cases that have been recorded each day since the outbreak
of the Coronavirus in India. This data is primary to the research since all predictions and
conclusions have been drawn by modelling and observing this data. There are various
other factors that propel the spread of the virus such as the number of testing centres,
number of successful tests carried out, number of affected migrants, the quality of
healthcare facilities, presence of public awareness, adherence to issued guidelines in the
country and so on. However, due to lack of adequate and accurate data pertaining to these
factors at the moment, this paper does not take them into consideration.
To maintain the integrity of the research, all the data that has been used in this paper is
sourced from the official COVID-19 website of the Government of India. Apart from
spreading the awareness of the virus amongst its citizens, this website also gives open
access to the daily count of the number of confirmed cases to one and all. This prevents
anomalous and inaccurate data to affect the course of the research.
The thus obtained data is stored in the form of a CSV (Comma-separated values) file as
shown in Table 1. It includes the total number of COVID-19 cases corresponding to each
date since the first positive case of COVID-19 was recorded in India. This number is
cumulative in nature. The data ranges from 30th of January 2020 since the outbreak to the
3rd of May 2020, which marked the end of the second extension of the lock-down. [2]
However, it has been observed that there is absolutely no change in the total number of
COVID-19 positive cases in the month of February 2020. Due to this, it has been regarded
as an outlier to the data and is not considered during the statistical modelling. This is done

ISSN: 2005-4238 IJAST 8945

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

in order to avoid any discrepancies to the accuracy of the predictions of the model and is
thus omitted from Table 1.

Table 1. Daily Count of COVID-19 positive cases in India

Number of COVID-19 positive
Date
cases per day (Cum.)
30-Jan-20 1
01-Feb-20 1
02-Feb-20 2
03-Feb-20 3
02-Mar-20 5
03-Mar-20 6
04-Mar-20 28
05-Mar-20 30
06-Mar-20 31
07-Mar-20 34
08-Mar-20 39
09-Mar-20 44
10-Mar-20 50
11-Mar-20 60
12-Mar-20 73
13-Mar-20 81
14-Mar-20 97
15-Mar-20 107
16-Mar-20 118
17-Mar-20 137
18-Mar-20 151
19-Mar-20 173
20-Mar-20 223
21-Mar-20 283
22-Mar-20 360
23-Mar-20 434
24-Mar-20 519
25-Mar-20 606
26-Mar-20 694
27-Mar-20 834
28-Mar-20 918
29-Mar-20 1024
30-Mar-20 1251
31-Mar-20 1397
01-Apr-20 1834
02-Apr-20 2069
03-Apr-20 2547

ISSN: 2005-4238 IJAST 8946

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

04-Apr-20 3072
05-Apr-20 3577
06-Apr-20 4281
07-Apr-20 4789
08-Apr-20 5274
09-Apr-20 5865
10-Apr-20 6761
11-Apr-20 7529
12-Apr-20 8447
13-Apr-20 9352
14-Apr-20 10815
15-Apr-20 11933
16-Apr-20 12759
17-Apr-20 13835
18-Apr-20 14792
19-Apr-20 16116
20-Apr-20 17656
21-Apr-20 18985
22-Apr-20 20471
23-Apr-20 21700
24-Apr-20 23452
25-Apr-20 24942
26-Apr-20 26917
27-Apr-20 28380
28-Apr-20 29974
29-Apr-20 31787
30-Apr-20 33610
01-May-20 35365
02-May-20 37776
03-May-20 40263

2.2. Statistical Modelling

A Scatter Plot is a graph of plotted points that shows the relationship between two sets of
numeric data. The obtained data in Table 1 is represented in the form of a Scatter Plot
wherein the total number of COVID-19 positive cases has been expressed as a function of
the corresponding date of occurrence, as shown in Figure. 1.

ISSN: 2005-4238 IJAST 8947

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Figure 1. Trend of COVID-19 in India

The Scatter Plot thus obtained closely resembles an exponential curve in nature.
According to the mathematical definition, a quantity is said to be growing at an
exponential rate if it increases with a rate proportional to its current size. Thus, by
observing the obtained graph, it can be implied that greater the number of people initially
affected by the virus in the early stages of its outbreak, greater the number of people who
will get affected in the later stages. The consistent doubling of cases in a fixed period is
the hallmark of exponential growth. [3][4]
The number of new infections that a single infectious individual will cause during their
infectious period is known as the basic reproduction number of a disease. The initial
exponential growth rate of a virus is an important measure of the severity of the spread,
and is also closely related to the basic reproduction number. This number is key to
determining how widespread a virus will become. For COVID-19, estimates of the basic
reproduction number have it somewhere between 1.03 and 1.67. [5][6][7]

2.3. Linear Regression Analysis

Linear Regression is the most commonly used tool of Predictive Analytics. The approach
of Linear Regression involves establishing relationships between two or more variables,
necessarily one dependant variable and one or more independent variables. In this case,
the dependant variable is the total number of COVID-19 positive cases, which is plotted
on the y-axis, and the independent variable or the predictor variable is the date, that is
plotted on the x-axis. [8]
The most common applications of Linear Regression Analysis are to determine the
strength of predictor variables, and to forecast an effect or trend. In other words, the
Regression Analysis is used to gauge the amount of change produced in the dependent
variable when it is subjected to a change in one or more independent variables.
Thus, Linear Regression Analysis is applied to the data in order to gauge the pattern of
growth of the dependant variable, that is, the total number of positive COVID-19 cases.
However, this approach is not applied to the entire data, but just a portion of it. Since the
lock-down was enforced on 24th of March 2020, it is important to apply Linear Regression
only to the data pertaining to the dates prior to it. This is done in order to predict the total
number of COVID-19 positive cases had the lock-down not been enforced. Thus, a
Scatter Plot of the total number of COVID-19 positive cases as a function of the dates
during which the lock-down had not been enforced is plotted as shown in Figure. 2.

ISSN: 2005-4238 IJAST 8948

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Figure 2. Trend of COVID-19 in India before the Lock-down

It is important to note that Linear Regression can only be applied if the curve is of the
linear form, in the equation of which, 'm' denotes the slope of the curve and 'b' denotes the
y-intercept of the curve. In other words, only if the variables share a linear relationship,
the approach of Linear Regression can be applied. The equation of the a linear curve is
stated as follows:
(1)

On the other hand, Coronavirus spreads in a communicable fashion, through which an

infected person affects whosoever is in close vicinity. This leads to the curve that is
depicted in Figure. 2., which is roughly similar to an exponential curve. Thus, the
approach of Linear Regression cannot be applied to it. The equation of an exponential
curve is of the following form, in which the base 'k' is a constant and 'x' is a variable:

(2)

This discrepancy in the approach is resolved by transforming the data into the Log-Linear
Model as described in the following section.
2.4. Transforming the Data into Log-Linear Model
In order to remove the discrepancy in the approach and to apply Linear Regression, it is
necessary to transform the exponential curve obtained in Figure. 2., into a linear curve. To
carry out the above mentioned process, a few mathematical simplifications are necessary.
It is known that the exponential curve is of the form that is depicted in Equation. 2. It can
be re-written as follows:

(3)

Taking logarithm on both sides of the above equation, the following equation is obtained:

(4)

ISSN: 2005-4238 IJAST 8949

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Upon further simplification of the above equation, one can arrive at the following
equation:

(5)

On application of the logarithm rules, the above equation can be simplified to:

(6)

It is known that,

(7)

Thus, Equation. 6 can be re-written as:

(8)

Assuming that,

(9)

(10)

Now, Equation. 8 can be written as:

(11)

Upon comparing Equation. 11 to Equation. 1, it is observed that they are now in similar
forms:

(12)

(13)

This process, in which an exponential curve is modified into a linear curve through the
application of logarithms is known as Logarithmic Transformation. [9]

In the scatter plots plotted throughout the course of this paper, in general, the x-axis
represents the date and y-axis represents the total number of COVID-19 positive cases
recorded on the corresponding date. As seen in the above mathematical derivation,
Logarithmic Transformation involves computing the logarithm of the y-axis values. The
logarithmic y values are now plotted on the y-axis in place of the previous values.

ISSN: 2005-4238 IJAST 8950

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Plotting the newly transformed y-axis values versus the x-axis values will result in a new
plot with a nearly straight line. This plot is termed to be in the form of a 'Log-Linear'
model. When one uses natural log values for the dependent variable (y) and keep the
independent variables (x) in their original linear scale, the resulting graph is called a Log-
Linear model.
The above explained process of Logarithmic Transformation is applied to the data
recorded before the lock-down was enforced as well as to the overall data collected during
the enforcement of the lock-down. The Scatter Plots, thus obtained, have been depicted in
Figure. 3. and Figure. 4. respectively.

Figure 3. Trend of COVID-19 in India before the Lock-down (Log-Linear)

ISSN: 2005-4238 IJAST 8951

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Figure 4. Trend of COVID-19 in India in Lock-down Period (Log-Linear)

As observed in Figure. 3. and Figure. 4., the Log-Linear Models obtained are in close
similarity to the Linear Curves, and resemble a straight line. This enables one to perform
Linear Regression Methods on these logarithmic-ally transformed curves.

2.5. Construction of Best Fit Line

The Line of Best Fit (or "trend" line) is a straight line that best represents the data on any
plot. This line may pass through some of the points, none of the points, or all of the
points. In other words, the line of best fit is the line for which the sum of the squares of
the residual errors between the individual data values and the line is at its minimum. [10]

A Line of Best Fit can be roughly determined using an eyeball method by drawing a
straight line on a scatter plot so that the number of points above the line and below the
line is about equal (and the line passes through as many points as possible). However, a
more accurate way of finding the line of best fit is the Least Square Method, as described
below.
The straight line that constitutes the Best Fit to a set of data points in the x-y plane is
typically calculated by minimizing the sum of the squares of the distances from the points
to the line—a method that was introduced by Legendre and Gauss more than two hundred
years ago.
The Line of Best Fit is drawn for the Logarithmic-ally transformed scatter plot for the
data that was recorded before the Lock-down was enforced. To do so, the approach of
Least Squares Method has been put to use as described in detail below. The line of best fit
is of the standard form:

(14)

where 'm' represents the slope of the line and 'b' represents the y-intercept of the line.
To arrive at the equation of the Best-Fit line, the following values are first calculated by
taking the sum of the x-axis values, logarithmic-ally transformed y-axis values, the square
of x-axis values, and the product of the x-axis and logarithmic-ally transformed y-axis
values in accordance to the data presented in Table I as shown below. The total number of
observations 'n' is computed as well.

(15)

(16)

(17)

(18)

(19
)

ISSN: 2005-4238 IJAST 8952

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Substituting the above values, the slope 'm' and the y-intercept 'b' of the Best-Fit Line are
computed using the following formulae:

(20)

(21)

where 'x' and 'y' represent the respective axis values.

After the application of the above formulae on the data, the values of 'm' and 'b' are
obtained to be:

(22)

(23)

Upon arriving at the values of the slope and intercept using the above formulae, the Best-
Fit line has been plotted over the graph in Figure. 2., to obtain the graph as depicted in
Figure. 5.
This Line of Best Fit depicted in Figure. 5. is the line that is the representation of all the
points in the Scatter Plot, since it is at an equal distance from all the points. In other
words, this Line of Best Fit represents the the data of all the COVID-19 positive cases that
were recorded before the lock-down had been enforced.

ISSN: 2005-4238 IJAST 8953

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Figure 5. Construction of Best-Fit Line

2.6. Future Prediction

A best-fit line is meant to mimic the trend of the data. Thus, in order to examine the
effectiveness of the lock-down enforced, it is necessary to extend the line of Best-Fit
constructed for the data when the lock-down had not been enforced throughout the data.
This is done since it is assumed that had the lock-down not been enforced, the number of
COVID-19 positive cases would have risen in a similar fashion. The Logarithmic-ally
Transformed linear graph for the data when the lock-down was in enforcement has been
depicted in Figure. 4. The Best-Fit line that was plotted in Figure. 5. is extended and
superimposed over the graph in Figure. 4. The plot thus obtained is depicted in Figure. 6.

Figure 6. Extension of Best-Fit Line to entire Data

In Figure. 6., the red line represents the Best-Fit Line that was drawn for the data that was
recorded before 24th of March 2020. In other words, it is the trend-line for the total

ISSN: 2005-4238 IJAST 8954

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

number of COVID-19 positive cases before the lock-down was enforced in the country.
The green line represents the extension of the Best-Fit line that is fitted on to the rest of
the data which includes the total number of COVID-19 cases recorded in the country
while the lock-down was in enforcement. The Observations of the Methodology followed
in the above sections has been elucidated in the following section.

2.7. Observations
The most striking observation made with respect to Figure. 6. at the end of this process is
the significant difference in the slopes of the extended Best-Fit Line and the Logarithmic-
ally Transformed Linear line. It can be observed that the slope of the line depicting the
total number of COVID-19 positive cases during the lock-down is much lesser when
compared to the slope of the line that represents the total number of COVID-19 positive
cases had the lock-down not been enforced. This gives a visual proof of the effectiveness
of the 41-day long lock-down. To strengthen the presented argument, the mathematical
proof of the observation has been presented below.
The two cases that need to be compared in order to arrive at a suitable conclusion are as
follows:
1. The total number of COVID-19 positive cases after 41 days had the lock-down not
been enforced. (As represented by the Best-Fit Line in Figure. 6.)

2. The total number of COVID-19 positive cases after 41 days of the lock-down
enforcement. (As represented by the scatter plot that has been obtained via Logarithmic
Transformation in Figure. 6.)

The mathematical observation can be made by arriving at the y-axis values corresponding
to the x-axis date of 3rd of May 2020. The values obtained upon doing so are 'y1' and 'y2'
respectively for the above mentioned cases:

(24)

(25)

However, it is important to note that Figure. 6. represents a Log-Linear model. This

implies that the x-axis values are linear in nature whereas the y-axis values are
logarithmic in nature.
Thus, the values of 'y1' and 'y2' that have been obtained in Equation. 24 and Equation. 25
are logarithmic in nature.

In order to compute the number of COVID-19 cases, it is necessary to convert the

logarithmic values back into the linear values. This can be accomplished by calculating
the anti-logarithmic values of 'y1' and 'y2', as shown below:

(26)

(27)

The obtained values from Equation. 26 and Equation. 27. are representative of the total
number of COVID-19 positive cases without and with the enforcement of the lock-down
respectively. It is thus concluded that the total number of cases in the country if the lock-
down had not been enforced would be approximately 4.8 million cases. This is an
exorbitant number. On the other hand, it is obtained that the number of cases with the

ISSN: 2005-4238 IJAST 8955

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

lock-down in place is close to 40 Thousand cases till the end of the second lock-down
extension of 3rd May 2020.

3. Conclusion
In this paper, an approach of Regression Analysis has been applied to the data of per-day
count of the number of COVID-19 positive cases in India that has been recorded ever
since the outbreak of the Coronavirus since 30th of January 2020 till the end of the second
extension of the enforced Lock-down, that is the 3rd of May 2020.
Upon following the methodology that involved the steps of Data Collection, Statistical
Modelling, Linear Regression Analysis, Logarithmic Transformation in order to obtain a
Log-Linear Model, Construction of the line of Best-Fit and obtaining the Future
Predictions, various visual and mathematical observations have been made as elucidated
in the previous section.
According to the results obtained through data visualisation and modelling, it can be
concluded that had the lock-down not been enforced, there would be close to a 4.8 million
COVID-19 positive cases in India by the 3rd of May 2020. On the other hand, since the
lock-down has been enforced on 24th of March 2020, the total number of COVID-19
positive cases in the country by the 3rd of May 2020 is around 40 Thousand.
Thus, it is concluded the lock-down of 41 days enforced by the Government of India to
tackle the Coronavirus pandemic has been highly effective in reducing the number of
active COVID-19 cases in the country by reducing the potential 4.8 million positive cases
to around 40 thousand positive cases in the through the duration of 41 days.

References
[1] Knatterud.,G.L., Rockhold, F.W., George, S.L., Barton, F.B., Davis, C.E., Fairweather, W.R., Honohan,
T., Mowery, R, O’Neill, R. (1998). "Guidelines for quality assurance in multicenter trials: a position paper".
Controlled Clinical Trials, 19:477-493.
[2] J. Mitlöhner, S. Neumaier, J. Umbrich and A. Polleres, "Characteristics of Open Data CSV Files," 2016
2nd International Conference on Open and Big Data (OBD), Vienna, 2016, pp. 72-79.
[3] J. Ma, J. Dushoff, B.M. Bolker, D.J.D. Earn "Estimating initial epidemic growth rates", Bulletin of
Mathematical Biology, 76 (2013), pp. 245-260
[4] E. Pourabbas, A. d’Onofrio, M. Rafanelli "A method to estimate the incidence of communicable
diseases under seasonal fluctuations with application to cholera", Applied Mathematics and Computation, 118
(2001), pp. 161-174
[5] J. Wallinga, M. Lipsitch "How generation intervals shape the relationship between growth rates and
reproductive numbers", Proceedings of the Royal Society B: Biological Sciences, 274 (2006), pp. 599-604.
[6] F.J. Richards "A flexible growth function for empirical use", Journal of Experimental Botany, 10
(1959), pp. 290-300.
[7] L. Zhong, L. Mu, J. Li, J. Wang, Z. Yin and D. Liu, "Early Prediction of the 2019 Novel Coronavirus
Outbreak in the Mainland China Based on Simple Mathematical Model," in IEEE Access, vol. 8, pp. 51761-
51769, 2020.
[8] Astrid Schneider, Dipl. Math.,1 Gerhard Hommel, Prof. Dr. rer. nat.,1 and Maria Blettner, Prof. Dr. rer.
nat.*,1. "Linear Regression Analysis" Dtsch Arztebl Intv.107(44); 2010 Nov; PMC2992018.
[9] M. T. Abuelma'atti and N. A. Tassaduq, "A new implementation for the logarithmic/exponential
function generator," 2014 International Symposium on Intelligent Signal Processing and Communication
Systems (ISPACS), Kuching, 2014, pp. 127-132.
[10] M. B. Jain, M. K. Nigam and P. C. Tiwari, "Curve fitting and regression line method based seasonal
short term load forecasting," 2012 World Congress on Information and Communication Technologies,
Trivandrum, 2012, pp. 332-337.

Authors

ISSN: 2005-4238 IJAST 8956

Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 5, (2020), pp. 8944-8957

Jabeena A., is an Associate Professor in School of Electronics

Engineering, VIT University, Vellore-632014, India. She has
completed her B.E [Electronics and Communication Engineering] in
Bhrathiar University, Coimbatore, an M.E. in Applied Electronics and
has received her PhD. in Optical communication at Vellore Institute
of Technology, Vellore. She has more than 25 years of teaching
experience and her Research Interest includes application of
Evolutionary algorithms to optimization problems in Wireless
Optical Communication, Satellite communication, Optical Sensors
and visible light communication. She is a member of Indian Society
for Technical Education, IEEE, Institution of Engineers and Indian
Science Congress Association. She has published more than 60 of her
contributions in Scopus journals and IEEE conferences. She is also
reviewer of several international conferences and Scopus indexed
journals.

Ananya. K. R., is a pre-final year student pursuing a Bachelor of

Technology degree in Electronics and Communication in Vellore Institute
of Technology, Vellore-632014, India. She is also a core-committee
member of the IEEE Student Chapter - VIT (affiliated to Madras Section,
India of IEEE). With a great interest in Machine Learning Algorithms and
its Applications, she plans to specialise in this field in the future. Her
Research Interest includes applying Machine Learning Techniques and
Statistical Modelling to solve real-world problems.

ISSN: 2005-4238 IJAST 8957

Copyright ⓒ 2020 SERSC