=Paper= {{Paper |id=Vol-3171/paper63 |storemode=property |title=Intelligent Analysis Impact of the COVID-19 Pandemic on Juvenile Drug Use and Proliferation |pdfUrl=https://ceur-ws.org/Vol-3171/paper63.pdf |volume=Vol-3171 |authors=Natalia Vlasova,Myroslava Bublyk |dblpUrl=https://dblp.org/rec/conf/colins/VlasovaB22 }} ==Intelligent Analysis Impact of the COVID-19 Pandemic on Juvenile Drug Use and Proliferation== https://ceur-ws.org/Vol-3171/paper63.pdf

Intelligent Analysis Impact of the COVID-19 Pandemic on
Juvenile Drug Use and Proliferation
Natalia Vlasova1, Myroslava Bublyk1
1
Lviv Polytechnic National University, S. Bandera Street, 12, Lviv, 79013, Ukraine

Abstract
This paper examines the state of drug use and sales during a lockdown caused by a pandemic
COVID-19. The focus group is juveniles in the United States, as there has been a sharp change
in drug mortality for this group in the United States during quarantine. The change in the death
rate from drugs among minors has been identified. The impact of drug prohibition and
legalization in the US economy on the level of drug use has been studied. Data on drug use and
distribution by juveniles were analyzed using descriptive statistics, data visualization,
smoothing (Kendall, Pollard, median, exponential), data correlation, and cluster analysis. The
results show that for minors aged 12-16, quarantine conditions have benefited by reducing the
trend of drug use, not only after quarantine but also in later life, and confirm the hypothesis of
a positive effect of lockdown on drug use reduction among minors in the United States.
Recommendations are proposed to increase the attention of the state and its implementation of
additional control measures, including conducting political and educational measures among
adolescents to prevent drug use and reduce the popularity of drug use for each succeeding
generation. It will positively benefit young people as drug prevention, and it will help reduce
drug mortality in the United States.

Keywords 1
Statistical Analysis, Information Technology, Intelligent Analysis, COVID-19 Pandemic,
Juvenile Drug Use, Juvenile Drug Proliferation, Business Analysis, Data Processing

1. Introduction
The problem of socio-economic development of each country, according to researchers [1-6], is very
sensitive to changes in external influences [7-11], critical of which the last two years are the pandemic
COVID-19 [12, 13]. During the pandemic in the United States, a record number of people died from
drug overdoses, about 100 thousand Americans [14-16]. Mortality rates have increased by 35%
compared to 2020. In 2019, the number of deaths due to drug exposure did not exceed 73 thousand. It
is the largest number of overdose deaths registered in a year. According to the National Institute on
Drug Abuse [15], this is the largest increase in drug overdose mortality since 1999 [17-19].
The fight against drugs has been going on for more than a century. The author [20] traces the history
of drug use since the 19th century. In the 20th century, the cause of death from drug use was that drug
addicts neglected treatment for a long time. It has been found that a large percentage of deaths are heroin
users born from the 1990s to the 2000s during the baby boom [20-23]. During the baby boom, a
generation was born that became a global drug user, and by 2022, the highest number of overdose deaths
was recorded among drug addicts of this generation. Over 50 years, this has led to a sharp increase in
drug use and frequency, as evidenced in all official documents and reports. From an economic point of
view, it also led to the rapid growth of the drug business and its criminalization [14-16, 20-24]. The
purpose of the work is as following.

COLINS-2022: 6th International Conference on Computational Linguistics and Intelligent Systems, May 12–13, 2022, Gliwice, Poland
EMAIL: nataliia.vlasova.sa.2019@lpnu.ua (N. Vlasova); my.bublyk@gmail.com (M. Bublyk)
ORCID: 0000-0002-3235-4714 (N. Vlasova); 0000-0003-2403-0784 (M. Bublyk)
©️ 2022 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
• Application of basic visualization methods, graphical display and primary statistical processing
of numerical data on the impact of the COVID-19 pandemic on juvenile drug use and proliferation,
presented by a sample.
• Study of trends in the behaviour of drug use by minors during the lockdown, using the basic
methods of identifying trends in the behaviour of addictions that represent the nature of the trend of
use,
• Presentation of the obtained results using MS Excel spreadsheet to confirm or refute the
hypothesis of a positive effect of lockdown on reducing drug use among minors.
• Using methods of correlation analysis of experimental data to establish the relationship between
copper data collected during the pandemic period.
• Application of the cluster analysis method to establish the cluster of the most drug-dependent
age groups of minors.
The task is to study the impact of COVID-19 on the level of drug use by minors on the example of
the largest data set on drug use in the United States. Identify the cluster of the most drug-dependent age
groups of minors to develop ways to counteract the growth of drug use among young people.

2. Literature review
The problem of drug use by minors became acute after the Second World War. Several important
documents have been adopted to control the spread of drugs. The Opium Convention was signed in
1909 in Shanghai [25]. It includes 13 countries of the International Opium Commission. It restricts
exports as opposed to banning or criminalizing the use and cultivation of opium, coca and cannabis.
The Convention provided that States would make every effort to control or seek to control all persons
producing, importing, selling, distributing and exporting morphine, cocaine and their related salts, and
buildings in which such persons are engaged in such industry or trade [25]. The Convention was
replaced by the 1961 Single Convention on Narcotic Drugs. Ukraine was ratified by the Convention in
2001, but on the website of the Verkhovna Rada of Ukraine on December 2, 2020, the Commission on
Narcotic Drugs decided to remove cannabis from List IV of the Convention after the proposals were
published by the World Health Organization in 2019 [25].
However, today the problem is not solved in Ukraine or worldwide. New reports of increasing
adolescent mortality from drug overdose are emerging [26-36]. During the quarantine of the COVID-
19 pandemic, retailers adapted to new conditions [37-49]. Quarantine through COVID-19 increased
unemployment and according to researchers [50-58], a certain part of the population was forced to look
for means of survival that were quite easy to obtain.
Impact of quarantine on juvenile use [59-69]:
1. Forced isolation due to the difficult epidemiological situation with COVID-19 has affected
young people differently.
2. Some have reduced consumption for reasons such as lack of parties and company, moving
parents from the metropolis to the suburbs and provinces.
3. And others, on the contrary, began to use much more due to a large amount of free time; this
category believes that buying drugs during the crown of the virus is safer than going to the
supermarket.
Our work is based on data from research by the National Center for Health Statistics (NCHS), one
of the leading statistical agencies under the US government [67]. It is located within several different
organizations within the Ministry of Health and Social Services and, since 1987, has been part of the
Centers for Disease Control and Prevention. They conduct four data collection programs: National Vital
Statistics System (NVSS), National Health and Nutrition Examination Survey (NHANES), National
Health Interview Survey (NHIS), and National Health Care Surveys (NHCS) [40-45].
The National Drug and Health Survey (NSDUH) is a significant source of statistics on illicit drug,
alcohol, and tobacco use and on the mental health of US civilians over the age of 12 [46-58]. The survey
tracks trends in specific interventions for substance use and mental illness and assesses the
consequences of these conditions by examining and treating mental and substance use disorders [59-
66, 68].
3. Methods
The following methods were used to solve the tasks [ 69-84].
● Data and information collection. Convert data to excel format.
● Descriptive statistics of data.
● Visualization (in polar and Cartesian coordinates; in the form of histograms, etc.).
● Smoothing according to Kendall formulas - a simple moving average, using the different
intervals.
● Smoothing according to formulas from Pollard.
● Exponential smoothing, values of α = 0.1, 0.15, 0.2, 0.25, 0.3
● Median smoothing using the different intervals.
● Cluster data analysis.

4. Experiments and Results
4.1. Data
The work is based on data from the National Center for Health Statistics (NCHS) study, namely the
NSDUH for 2020 [14, 40-41, 67]. The dataset consists of data on the frequency of drug use among ten
age groups of minors in the United States from 12 to 21 years (Table 1). It covers 13 drugs across 10
age groups. The average value of the polled number of people is equal to 2671.

Table 1
US drug use by age dataset
hall
pain- oxyc tran sti sed
alcoh marij coc crac her ucin inha met
n age releiv onti quili mul ativ
ol uana aine k oin oge lant h
er n zer ant e
n
2798 12 3,9 1,1 0,1 0,0 0,1 0,2 1,6 2,0 0,1 0,2 0,2 0,0 0,2
2757 13 8,5 3,4 0,1 0,0 0,0 0,6 2,5 2,4 0,1 0,3 0,3 0,1 0,1
2792 14 18,1 8,7 0,1 0,0 0,1 1,6 2,6 3,9 0,4 0,9 0,8 0,1 0,2
2956 15 29,2 14,5 0,5 0,1 0,2 2,0 2,5 5,5 0,8 2,0 1,5 0,3 0,4
3058 16 40,1 22,5 1,0 0,0 0,1 3,4 3,0 6,2 1,1 2,4 1,8 0,3 0,2
3038 17 49,3 28,0 2,0 0,1 0,1 4,8 2,0 8,2 1,4 3,5 2,8 0,6 0,5
2469 18 58,7 33,7 3,2 0,4 0,4 7,0 1,8 9,2 1,7 4,9 3,0 0,5 0,4
2223 19 64,6 33,4 4,1 0,5 0,5 8,6 1,4 9,4 1,5 4,2 3,3 0,4 0,3
2271 20 69,7 34,0 4,9 0,6 0,9 7,4 1,5 10,0 1,7 5,4 4,0 0,9 0,5
2354 21 83,2 33,0 4,8 0,5 0,6 6,3 1,4 9,0 1,3 3,9 4,1 0,6 0,3

4.2. Descriptive statistics and Cartesian and polar coordinate systems
Descriptive statistics are quantitative characteristics of data [70, 85-91]. To obtain the data results
of descriptive statistics in Excel, in the section "Data," the method "Data analysis" was selected. The
item "Descriptive statistics" was selected. In the menu of "Descriptive statistics," all values from the
table "Alcohol " were set, and the place of output of values was indicated (Table 2 - Table 3). Similar
actions were taken for the other drugs. After all the data, we were obtained. The result of Average,
Standard error, Median, Moda, Standard deviation, Sampling variance, Excess, Asymmetry, Interval,
Minimum, Maximum, Amount, and Account were prepared, namely, formatting. All numbers were
reduced to "00.00".
Fig. 1 shows the structure of 13 drugs used by age in the Cartesian coordinate system. Fig. 2 shows
the structure of 13 drugs used by age in the polar coordinate system.
Table 2
Descriptive statistics of the US drug use by age
Parametre alcohol marijuana cocaine crack heroin hallucinogen
Average 42,53 21,23 2,08 0,22 0,30 4,19
Standard error 8,57 4,19 0,63 0,08 0,09 0,96
Median 44,70 25,25 1,50 0,10 0,15 4,10
Moda - - 0,10 0,00 0,10 -
Standard deviation 27,12 13,24 2,00 0,25 0,29 3,04
Sampling variance 735,26 175,34 4,01 0,06 0,08 9,27
Excess -1,29 -1,60 -1,77 -1,82 0,37 -1,66
Asymmetry -0,08 -0,50 0,41 0,53 1,09 0,06
Interval 79,30 32,90 4,80 0,60 0,90 8,40
Minimum 3,90 1,10 0,10 0,00 0,00 0,20
Maximum 83,20 34,00 4,90 0,60 0,90 8,60
Amount 425,30 212,30 20,80 2,20 3,00 41,90
Account 10,00 10,00 10,00 10,00 10,00 10,00

Table 3
Descriptive statistics of the US drug use by age (continue)
inhal pain-
Parametre ant releiver oxycontin tranquilizer stimulant meth sedative
Average 2,03 6,58 1,01 2,77 2,18 0,38 0,31
Standard error 0,18 0,95 0,20 0,60 0,46 0,09 0,04
Median 1,90 7,20 1,20 2,95 2,30 0,35 0,30
Moda 2,50 - 0,10 - - 0,10 0,20
Standard deviation 0,58 3,02 0,62 1,89 1,46 0,28 0,14
Sampling variance 0,34 9,10 0,39 3,58 2,14 0,08 0,02
Excess -1,40 -1,48 -1,39 -1,48 -1,56 -0,28 -1,17
Asymmetry 0,40 -0,46 -0,50 -0,13 -0,09 0,41 0,10
Interval 1,60 8,00 1,60 5,20 3,90 0,90 0,40
Minimum 1,40 2,00 0,10 0,20 0,20 0,00 0,10
Maximum 3,00 10,00 1,70 5,40 4,10 0,90 0,50
Amount 20,30 65,80 10,10 27,70 21,80 3,80 3,10
Account 10,00 10,00 10,00 10,00 10,00 10,00 10,00

crack
heroin
100,00
oxycontin
80,00 meth
sedative
Using

60,00
inhalant
40,00 cocaine
20,00 stimulant
tranquilizer
0,00
12 13 14 hallucinogen
15 16 pain-releiver
17 18
19 20 21
marijuana
alcohol
Age
Figure 1: Visualization of drug use by age in the Cartesian coordinate system
12
90,00 crack
80,00
21 13 heroin
70,00
60,00 oxycontin
50,00
meth
40,00
20 30,00 14 sedative
20,00
inhalant
10,00
0,00 cocaine

stimulant

19 15 tranquilizer

hallucinogen

pain-releiver

18 16 marijuana

alcohol
17
Figure 2: Visualization of drug use by age in the polar coordinate system

4.3. Histogram and cumulative
We consider the example of marijuana use. To construct a histogram, the values of the boundaries
of the intervals are indicated, and rectangles are constructed on their basis, the height of which is
proportional to the frequencies (or frequencies). Data Analysis >> Histogram was opened, and
parameters were set. Fig. 3 show the histogram of the frequency of marijuana use by age. Fig. 4 shows
cumulative of the frequency of marijuana use by age.

0
1,1 12,06666667 23,03333333 Total

Figure 3: Histogram of the frequency of marijuana use by age
Marijuana
40,00
35,00
30,00
25,00
20,00
15,00
10,00
5,00
0,00
12 13 14 15 16 17 18 19 20 21

Figure 4: Cumulative the frequency of marijuana use by age in the Cartesian coordinate system

12
35,00
21 30,00 13
25,00
20,00
15,00
20 10,00 14

5,00
0,00

19 15

18 16

17
Figure 5: Cumulative of the frequency of marijuana use by age in the polar coordinate system

5. Discussions
Two smoothing methods classes differ in approaches. The first approach is called analytical. Based
on visual analysis, the researcher can set a general view of the function, believing that its graph
corresponds to the nature of the trend. The second approach is called algorithmic. Here, researchers
look at the trend through the use of various smoothing procedures. The algorithmic approach uses the
following methods [70, 72, 82-84].
• Simple or ordinary moving average;
• Weighted moving average;
• Exponential smoothing;
• Median smoothing.
Figure 6 shows the results of using the simple moving average method for marijuana use.
40,00
35,00
30,00
25,00
20,00
15,00
10,00
5,00
0,00
1 2 3 4 5 6 7 8 9 10

Figure 6: The simple moving average of marijuana use by age

Along with simple moving averages, polynomial or weighted averages are also used [92-98]. These
methods allow us to describe the main trend of the series more accurately because when calculating the
weighted average, each level of the series within the smoothing interval is assigned a certain weight,
depending on the distance to the middle of the interval.
The result for marijuana uses by age is shown in Fig. 7, where the moving average is realized using
the minimum smoothing interval w = 5.

40,00
35,00
30,00
25,00
20,00
15,00
10,00
5,00
0,00
1 2 3 4 5 6 7 8 9 10

Figure 7: The moving average of marijuana use by age at w=5

Fig. 8 shows the exponential smoothing result of marijuana use by age at alpha = 0.1.
40,00
35,00
30,00
25,00
20,00
15,00
10,00
5,00
0,00
1 2 3 4 5 6 7 8 9 10

Figure 8: The exponential smoothing of marijuana use by age at alpha = 0.1
5.1. Median filtration
Median smoothing and turning point criteria according to the formula: = IF ((AC3> AA3); (AC3>
AE3); OR (IF (AC3