=Paper=
{{Paper
|id=Vol-3288/short4
|storemode=property
|title=Web-Based Management System for Customer Interaction in E-Trade with Adaptive Interface (short paper)
|pdfUrl=https://ceur-ws.org/Vol-3288/short4.pdf
|volume=Vol-3288
|authors=Valentyna Pleskach,Viacheslav Zosimov,Oleksandra Bulgakova,Oleg Nagornyi
|dblpUrl=https://dblp.org/rec/conf/cpits/PleskachZBN22
}}
==Web-Based Management System for Customer Interaction in E-Trade with Adaptive Interface (short paper)==
https://ceur-ws.org/Vol-3288/short4.pdf
Web-Based Management System for Customer Interaction
in E-Trade with Adaptive Interface
Valentyna Pleskach1,2, Viacheslav Zosimov2, Oleksandra Bulgakova2,3, and Oleg Nagornyi2
1
Vilnius Gediminas Technical University, 11 Saulėtekio al., Vilnius, LT-10223, Lithuania
2
Taras Shevchenko National University of Kyiv, 24 Bohdan Hawrylyshyn str., Kyiv, 04116, Ukraine
3
V. O. Sukhomlynsky National University of Mykolaiv, 24 Nikolska str., Mykolaiv, 54000, Ukraine
Abstract
The paper present a modeling of mechanisms for determining consumer priorities in the e-trade
market and the development of models, methods and information technologies on this basis that
ensures the implementation of effective mechanisms for managing interaction with customers
of e-trade enterprises. The main focus is to research the mechanisms of formation and
identification of consumer priorities in the e-trade market and the development of artificial
intelligence technology that can be used to study complex patterns of interaction in e-trade and
be the basis for developing of modern customer information management technologies.
Keywords 1
E-trade, modeling, consumer priority, CRM information technology, intellectual analysis,
clustering, artificial intelligence.
1. Introduction using information technology is very relevant.
This task is especially relevant in the current
context of the COVID-19 pandemic, war state,
Today it is impossible to imagine the modern
when e-trade is constantly growing, and e-trade
development of the economy without the
itself becomes an effective and reliable channel
widespread implementation and full use of digital
for economic activity and an important factor in
technologies. The pace of digital innovation is
ensuring socio-economic and social development.
growing every year, and, accordingly, the number
The object of research is the mechanisms of
of Internet users is increasing.
formation and identification of consumer
Hootsuite Global Digital research shows that
priorities in the e-trade market, interaction
on average, more than a million new users of the
processes, business processes and information
global Internet are added every day. In January
flows in e-trade. The subject of research is the
2021, the number of active Internet users was 4.66
theoretical-methodological and organizational
billion people. This is 316 million (7.3%) more
principles of functioning of e-trade enterprises,
than at this time last year (2020). According to
methods, models and information technologies of
research for 2021, the global Internet penetration
interaction management in e-trade. Among the
is 59.5% [1]. The need to develop e-trade is one
main tasks of the article is the development of
of the innovative sectors of the national economy,
technology for determining consumer priorities in
the development of which directly depends on the
e-trade, taking into account the consumer profile
level of implementation of information and
and a set of personal customer characteristics
communication technologies. Businesses have the
using intellectual data analysis and artificial
opportunity to increase their profits by creating
intelligence methods and on this basis the
web-based representations with the presentation
development of web-based management system
of their e-services.
for customer interaction in e-trade [3–7].
The problem of finding effective mechanisms
for determining consumer priorities in e-trade
CPITS-2022: Cybersecurity Providing in Information and Telecommunication Systems, October 13, 2022, Kyiv, Ukraine
EMAIL: v_pleskach@ukr.net (V. Pleskach); zosimovvv@gmail.com (V. Zosimov); sashabulgakova2@gmail.com (O. Bulgakova);
jeffhardyte@ukr.net (O. Nagornyi)
ORCID: 0000-0003-0552-0972 (V. Pleskach); 0000-0003-0824-4168 (V. Zosimov); 0000-0002-6587-8573 (O. Bulgakova); 0000-0002-
7881-5491 (O. Nagornyi)
©️ 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)
82
2. The Ukrainian E-Trade 3. Interface Adaptation based on
Development in the Context Cyberentities
of COVID-19 and War State
Interfaces provide software products
The processes of the COVID-19 pandemic and management and communication of users with the
martial law in Ukraine have become factors in program. The advantages of using software
accelerating the development of e-commerce products aimed at improving the usability of
systems in order to survive e-commerce interfaces include: simplifying the perception of
enterprises in these difficult conditions. The rate the software product business logic by a specific
of growth of e-commerce systems in Ukraine is user; flexibility of the information presentation
constantly increasing, the number of electronic model; improving the performance of the
orders for goods increases by approximately 25– interface. Dynamic change should be understood
30%, and the turnover in hryvnias is by 40–60%. as a change in the display of the interface
The digital transformation of the economy (adaptation), because of the execution of scenario
changes the generally accepted ideas about how based on the user behavioral portrait [10–13].
business is structured, how consumers receive A criteria set СR {сr1, ..., сrСR } that will
services, goods, and how business structures change, that is, adapt to the user,
should adapt to these challenges and regulate Fact { fact1 , ..., fact Fact }
them. In 2021–2022, the coronavirus pandemic is this is a factors set
significantly motivated customers around the that will influence the choice of one or another
world to make purchases without leaving their criterion. Based on the factors identified in
homes. Nowhere has there been more advance that influence the interface, the criteria
unprecedented and unpredictable growth than in for building the interface will change. Interface
the digital and e-commerce sectors, which have adaptation is been from specific observational
flourished amid the COVID-19 crisis. Against the data, that is, factors, to a general model that
background of a slowdown in economic activity, includes many changing criteria.
COVID-19 led to a surge in e-commerce and So, you can build a function Ф that reflects the
accelerated the digital transformation of society. adaptation process:
Businesses and consumers are increasingly (сr1 ( fact1 ,..., fact k ),...,сrm ( fact k ,..., fact F ))
“moving to digital markets. Statistics show that a m k
total of 59% of respondents said they will invest 0 i сri ( fact1 ,..., fact F )
i 1 F 1
more in e-commerce channels in the near future as m m k
a result of COVID-19. Companies in these ij сri сrj ( fact1 ,..., fact F ) (1)
i 1 j 1 F 1
categories are more likely to do it in the following m m m k
areas: consumer electronics (78%); packaged ijk сri сrj сrk ( fact1 ,..., fact F ) ...
food and beverages (77%); baby care products i 1 j 1 k 1 F 1
(75%); personal hygiene products (74%) [8]. Software requirements:
In the United States, the B2C e-commerce The software product must provide: collection
industry is expected to hold on to the global and storage of information about users of the
pandemic accelerated sales in 2022, with big involved web application; pseudo-identify users
retailers, including Amazon and Walmart, set to based on the collected data; automate the
benefit from the growing online shopping trends adaptation of pseudo-identified user interfaces.
among consumers in the country. While the The software product must match the
growth in sales declined in 2021, the total following requirements:
purchases remained far above the pre-pandemic 1. Controlled collection of information on
levels [9]. users of the web application.
Another important factor with the surge of the 2. Creation and support of databases of
COVID-19 pandemic was cybercrime, which received information.
grew more than any other criminal activity. The 3. Pseudo-identification of web application
FBI notes that cybercrime reports have users based on the collected information.
quadrupled during the COVID-19 pandemic. 4. Automated adaptation of the user
These problems are described in more detail in interface of the web application based on the
[2].
83
data collected and processed by the software characteristics collected on the basis of the
product. software product interface objects
Fig. 1 shows the interaction scheme of the programmed to track changes, the application
software complex. of personal settings “Involvement settings”
and subsequent monitoring of changes in
potentially changeable settings of the
“Tracking status changes” are monitored. All
changes to the tracked web interface settings
are saved “Saving the latest user settings” to
the database at each stage of their
implementation.
If the user visits the “User not identified”
web service for the first time and no
identification matches are found, the specified
changes by the user are tracked and the
“Saving changes made by the user” is further
Figure 1: The interaction scheme of the software saved at the stage of their implementation.
complex
“AAUI is Automatic Adaptation User
Interface” is the initial module from which
interaction with the components of the software
product, “Identification” is implements the
formation of an entity from the collected
information about the object that has shown
activity, “Operating system” is collection of
information about the software of the active user,
“Browser identification” is responsible for
collecting information about the software
involved in interaction with the final software
product, “Human behavior” is provides a check
for the possible past presence of a pseudo-
deanimized object, “Adaptation” is the block is
responsible for the embedded software adaptation
implemented on the basis of the resulting imprint
of the pseudo-deanimized object of interaction
with the final product, “Database” is the block is
responsible for saving the analyzed information,
pseudo-deanimation tokens and interface
adaptation rules, “Process identifier” is the block Figure 2: The scheme of modulation applied
is responsible for matching the identification of software product
existing interaction objects.
Fig. 2 show a block diagram of modulation
when using a software product.
4. Management System for
“LOG IN” block is displays the user’s visit to Customer Interaction in E-Trade
the service; “Set token” block is displays the
generation of an identifier by the “AAUI” An online store selling handmade goods was
software system, the generation of an identifier created to test the technology of intelligent
occurs in parallel with the collection of additional interface customization.
data for stricter user identification. Several types of UML diagrams will be
Checking visit is check for a possible past visit schematically depicted below, each of which
by the user to the service: models the subject area from different
In the case of identifying a user “User perspectives. The online handmade goods store is
identified”, the “Analysis behavioral the main object of the simulation, so all diagrams
characteristics” of the behavioral will be created based on the main component.
84
Fig. 3 show a diagram of the options for using
the developed web resource. The web resource
was created according to the classical online
stores scheme and has all the necessary
functionality: a catalog, a shopping cart, an online
order, an authorization system.
Figure 5: UML sequence diagram without
registration and authorization
Fig. 6 show the catalog page fragment of the
developed online store. Here is the site header,
Figure 3: UML-diagram of precedents (options of which contains logo, contact information, main
use) menu of the site. The left column contains the
filter block, and the products are presented on the
Fig. 4 show the UML class diagram that right.
formed the basis for creating a web resource. It
contains classes: Catalog, Category, Product,
Website, User, and Order.
Figure 6: System for customer interaction in e-
trade
5. Interface Adaptation for
Figure 4: UML class diagram Application Solution
Fig. 5 show the sequence of user actions from The main advantages of the software system
the moment he entered the site and until the order are the applied model of identification of
acceptation. At the top of the diagram, the roles of anonymous users of final software products for
users are indicated: Client, Administrator, as well the further use of a dynamic identifier in order to
as functional modules which the client interacts automatically adapt the interface to the identified
with: Product catalog, Cart. After entering the site, user. “AAUI” is a server-side software product
the client interacts with the product catalog. The written in the PHP scripting programming
result is the added to the necessary products. At language.
the stage of the client interaction with the Cart The software product provides:
module, the order is placed. At the final stage, the Software product administration
Administrator who carries out the final (managing the list of active identification
confirmation of the order is included in the filters, changing, editing, creating adaptation
scheme [14]. rules).
Storage and processing of information
about users.
Ensuring the storage of information about
sessions.
85
The data source is the database. It stores all The block for obtaining information about
information about pseudo-deanonymized users user-set languages, time zone, in listing is
and rules for dealing with established groups. presented.
End-users interact with the software system The function of automatic localization
through primary interaction with the interface. adaptation of the web application based on the
The server application “AAUI” interacts with collected data is presented in listing:
the user of the web application at the moment of
his stay and interaction with the latter's interface. public function handle($request, Closure $next)
{
The level of absorption of collected information if(!\Session::has('locale'))
about the visitor is set by the administrator of the {
software system. \Session::put('locale',
\Config::get('app.locale'));
The main source of data storage is the }
database. It stores information about the received $usersrule = DB::table('data')-
entities of pseudo-deanonymized users. >select('rule', 'rule')->get();
if
The free MySQL relational database (checkrule($usersrule)==Config::get('app.
management system is used to store information. locale'))
app()-
The structure of the database is developed on the >setLocale(\Session::get('usersrule'));
basis of the developed user identification return $next($request);
methodology. The implementation of the user }
pseudo-identification system is interesting in this
product. We will consider the interaction between 6. The Protection of Confidential
parts of the software product using the following
code parts as an example: Data
App: :before(function($request){ To solve the problem of storing structured
$url_lang = Reques::segment (1); information cross the Internet in local storage and
$cookie_lang = Cooki::get(‘language’);
$browser_lang = substr(Request: fragments of information transferred to the
:server(‘HTTP_ACCEPT_LANGUAGE’), 0, 2); browser from the site visited by the user, the paper
iF(!empty($ur1_lang) AND in_array($url_lang, proposed a method of storing the JSON web key
Config::get(‘app.language’)))
{ in a local variable inside the closure. In this case,
if($url_lang !=$cookie_lang) the token cannot be obtained during an XSS and
{ CSRF attack on the partition from when the token
Session::put(‘language’,
$url_lang); is stored in LocalStorage or a cookie, because it is
} stored in memory, and placed it in local storage
App::setLocale($url_lang);
}
given an advantage in saving data, so that now the
else if(!empty($cookie_lang) AND private keys not became public. Every time an
in_array($cookie_lang, Config::get(‘app. attacker tries to gain privacy using access token,
languages’)))
{
he will spend more time to find and get the token
App::setLocale($cookie_lang); along with other methods of saving access token,
} so this method can be better than storing token in
else if(!empty($browser_lang) AND
in_array($browser_lang, Config:: get(‘app.
LocalStorage or cookie.
languages’))) In addition, this method allows you to protect
{ the user's storage data even when the web
if ($browser_lang != $cookie_lang)
{ resource is accidentally closed. Suppose the user
Session::put( ‘language', exits the current session, closes the browser tab.
$browser_lang); Now, when the user enters the application again,
}
$timezone = \Auth::user()->timezone; the system looks like this (see Fig. 7):
$datetime - $this->asDateTime($value); 1. If there is no JWT in memory, the silent
DB: : table( ‘essences *)->insert( update workflow [15, 16] is started.
[‘langcode’ => , $browser_lang) ‘votes’ -
> 0] 2. If the refresh token is still valid (or has
[‘date’ => , $datetime + $timezone) ‘votes’ not been revoked), a new JWT is sent [17].
=> 0] In this way, it is possible to maintain the
);
} client's authorization inadvertently at the end of
the life of the access token.
86
development of web-based management system
for customer interaction in e-trade.
The presented software development is aimed
at automated adaptation of interfaces to the needs
of users. The software product provides pseudo-
identification of users (building a database of
anonymous users and rules based on their
presence in web applications). The main
advantages of the software system are the applied
model of identification of anonymous users of
final software products for the further use of a
dynamic identifier for the purpose of automatic
adaptation of the interface to the identified user.
Figure 7: Workflow diagram Prospects for further development include
increasing the number of identification markers to
increase the probability of user identification,
7. Conclusions implementing embeddability in content
management systems, and optimizing data loss.
The aim of the article is to model mechanisms
for determining consumer priorities in the e-trade 8. References
market and to develop models, methods and
information technology for managing interaction
with customers of e-shops based on intellectual [1] Hootsuite Global Digital Research.
data analysis and artificial intelligence methods. Available from, version 02, 2022.
The article aims to solve the complex problem https://www.hootsuite.com/
of finding effective mechanisms for determining [2] V. Pleskach, et al., Current State and Trends
consumer priorities in e-trade based on modern in the Development of E-Commerce
methods, models and information technologies. Software Protection Systems. CEUR
The study is especially relevant in the current Workshop Proc., 2021, vol. 3179, pp. 79–88.
context of the COVID-19 pandemic, war state [3] J. Reynolds, The Complete E-Commerce
when e-trade is constantly growing, and e-trade Book: Design, Build & Maintain a
itself becomes an effective and reliable channel Successful Web-based Business, 2nd Ed.,
for economic activity and an important factor in 2004.
ensuring socio-economic and social development [4] Retail E-Commerce Sales Worldwide,
[18–20]. The main idea of the article is a modeling Version 02, 2022. https://www.statista.com/
of mechanisms for determining consumer statistics/379046/worldwide-retail-e-
priorities in the e-trade market and the commerce-sales/
development of models, methods and information [5] G. Sreedhar, Improving E-Commerce Web
technologies on this basis that ensures the Application through Business Intelligence
implementation of effective mechanisms for Techniques. N.Y.: IGI Global, 2018.
managing interaction with customers of e-trade [6] V. Grechaninov, et al., Formation of
enterprises. The main focus of the article is to Dependability and Cyber Protection Model
study the mechanisms of formation and in Information Systems of Situational
identification of consumer priorities in the e-trade Center, in Workshop on Emerging
market and the development of artificial Technology Trends on the Smart Industry
intelligence technology that can be used to study and the Internet of Things, vol. 3149, 2022,
complex patterns of interaction in e-trade and be pp. 107–117.
the basis for developing of modern customer [7] V. Buriachok, et al., Invasion Detection
information management technologies. Among Model using Two-Stage Criterion of
the main tasks of the article is the development of Detection of Network Anomalies,
technology for determining consumer priorities in Cybersecurity Providing in Information and
e-trade, taking into account the consumer profile Telecom. Systems, 2020, pp. 23–32.
and a set of personal customer characteristics [8] M. Pleskach, et al., Mechanisms for
using intellectual data analysis and artificial Encrypting Big Unstructured Data.
intelligence methods and on this basis the Technical and Legal Aspects 2021 11th
87
International Conference on Advanced com/chart/13139/estimated-worldwide-
Computer Information Technologies, ACIT mobile-e-commerce-sales/
Proceedings, 2021, pp. 504–509. [20] Personalized Product Recommendations in
[9] Digital Payment Trends for 2021, version 04, Ecommerce, ver. 8, 2022. https://www.per
2022. https://www.entrepreneur.com/article/ zonalization.com/blog/personalized-
363921 product-recommendations-in-ecommerce/
[10] V. Zosimov, O. Bulgakova, V. Pozdeev,
Semantic Profile of Corporate Web
Resources, CEUR Workshop Proc., 2021,
vol. 3179, pp. 389–397.
[11] V. Zosimov, O. Bulgakova, Web Data
Displaying Approach Based on User's
Semantic Profile Templates. International
Scientific and Technical Conference on
Computer Sciences and Information
Technologies, 2020, pp. 428–431.
doi: 10.1109/csit49958.2020.9321839
[12] O. Bochkarov, V. Holembo, Vykorystannia
intelektualnykh tekhnolohii zboru danykh u
avtonomnykh kiberfizychnykh systemakh,
Lviv Polytechnic National University
Institutional Repository Zbirnyk naukovykh
prats, 2015, vol. 830, pp. 7–11.
[13] W. Karwowski, A Review of Human Factors
Challenges of Complex Adaptive Systems:
Discovering and Understanding Chaos in
Human Performance, Human Factors, vol.
54, no. 6, 2012, pp. 983–995.
[14] S. Obushnyi, et al., Ensuring Data Security in
the Peer-to-Peer Economic System of the
DAO, in Cybersecurity Providing in
Information and Telecommunication
Systems II, vol. 3187, 2021, pp. 284–292.
[15] Password Stealing from HTTPS Login Page
and CSRF Protection Bypass with Reflected
XSS, ver. 9, 2022. https://medium.com/
@MichaelKoczwara/passwordstealing-
from-https-login-page-and-csrf-bypass-
with-reflected-xss76f56ebc4516
[16] Cross-Site Request Forgery Prevention
Cheat Sheet, ver. 8, 2022.
https://cheatsheetseries.owasp.org/cheatshee
ts/CrossSite_Request_Forgery_Prevention_
Cheat_Sheet.html
[17] JSON Web Tokens, ver. 8, 2022.
https://jwt.io
[18] Digital Wallets to Represent Half of Global
Ecommerce Sales by 2023, ver. 09, 2022
URL: https://thepaypers.com/ecommerce/
digital-wallets-to-represent-half-of-global-
ecommerce-sales-by-2023-fis-study–
1240916#
[19] Estimated Worldwide Mobile E-Commerce
Sales, ver. 8, 2022. https://www.statista.
88