The paper simulates the process of interaction between virtual enterprises (participants) in the Internet world of e-commerce based on Petri nets. The general formal model of mutual relations using payment systems (processes) between participants of e-commerce is described. Regardless of the type of e-commerce, the primary indicator of the impact on the profit of ecommerce is the involvement of potential audiences to visit the relevant e -commerce sites and their further motivation to make the applicable orders. Appropriate e-commerce referral systems should withstand the worst of motivating the user to place an order. Timely further analysis of the statistics of visits and conversions depends on the formation of new goals to encourage potential e-commerce customers. The e-commerce system model based on the Petri net allows determining the set of financial relations between the participants of e -business. Analysis of labelled systems based on the Petri net enables you to build a tree of achievement of e-commerce goals and finding the relationships between site visits and e -business profits. Further application of the methods of correlation and cluster analysis to the obtained data allows classifying the purposes of conversions depending on the market reaction and the current time for the necessary adjustment. E-commerce system, information process marking, e-business, Petri net, time series, reach tree, trend detection, information technology, electronic content commerce system, intelligent system, payment system, average method, financial transaction, virtual enterprise, internet shop, successful conversion, cluster analysis, machine learning COLINS-2021: 5th International Conference on Computational Linguistics and Intelligent Systems , April 22-23, 2021, Kharkiv, Ukraine
The twenty-first century has challenged humanity in the form of a worldwide “web” of the Internet and the emergence of an interactive business or virtual economy. An on-line business is a business built on the joint actions of a business process in a businessperson and a computer or other means of communication. A virtual economy is an economy based on interactive business and the primary law of man - saving time. Already in developed countries, there are many cases of implementation of this law, namely: using communication means, without leaving home, to operate technological lines in production or financial and commercial activities; maintain accounting; carry outdistance learning, reading books and periodicals; to buy goods; perform banking, exchange and other financial transactions. Nowadays, the Internet influences both the external relations between companies and their partners or clients and the internal structure of the companies themselves. E-commerce is based on the design of traditional commerce, and the use of electronic networks gives it flexibility.
E-commerce is the product/service sale for commercial electronic media, either via the network or via the Internet. Data may include ordering, payment and delivery of goods/services. Component ecommerce (EC) is presented on Fig. 1 [1-3]. The process that remains the EC cycle [4-6]: available
2021 Copyright for this paper by its authors. content, formal ordering, payment, order placement, after-sales service and support. EC types [7]: trade in content, trade, services.
EC Categories [1-7]: Business-to-Business or B2B (which is a company that uses reasoning to order offers, required options and pay); Business-to-Consumer or B2C(electronic retail); Business-to-Administration or B2A (operations used between companies and government organizations); Consumer-to-Administration or C2A (does not yet exist, but it is pre-existing two categories used in different fields as social visits); Consumer-to-Consumer or C2C (the user who uses to reduce commercial content, informs, audits the trade representation between individuals) this reduction is the ability of firms to take on the functions traditionally performed by middleware professionals since the Internet has a more efficient ability to interact with consumers and at the same time allows monitoring consumer information. On the technical side, this is due to developing technologies for building and maintaining databases and automatic processing of incoming requests.
Features of the Internet as a means of retail trade [22-27]: The consumer interacts with the Internet and requests for purchase directly through the Internet; The consumer controls the content on products and services and receives it at any time of the day; Organizing retailing on the Internet requires less investment than traditional retailing.
Compared to television or catalogue trade, consumers may request additional content sufficient to make a purchase, including on-line. Advantages of the Internet market [28-30]: Active position of the consumer; A comparatively smaller amount of money invested by firms to enter the market; The global nature of the market; 24/7 market availability; Possibility of obtaining the necessary amount of content
The development of the Internet environment as a market requires the implementation of three primary conditions [31-33]: The conquest of firms using the Internet as a channel of distribution of products, consumer confidence; Ensuring the reliability of participants and their operations; Creating security for transmission and further storage of data on the Internet.
Methods of gaining trust [34-37]: 1. Increase in guarantors shifting the risk of buying from buyer to seller; 2. Significant discounts for first-time customers. The first operation significantly improves the trust relationship between the buyer and the firm, which were almost absent before; 3. Details of the firm, including its history, business philosophy and biography, approval marks from other firms certifying the seller, etc.
Not being the only technology, e-commerce is characterized by versatility. It brings together a wide range of business operations, including [38-42]: Establishing contact, for example, between a potential customer and a supplier; Information exchange; Pre- and after-sales support (detailed product and service content, product instructions, customer answers); Selling; electronic payment (using electronic transfer of money, credit cards, electronic checks, electronic cash); Management of delivery and tracking of physical products, direct delivery of products that can be distributed electronically; Virtual enterprises are groups of independent companies working together to provide products and services that are not available to individual companies; The company and its trading partners jointly manage shared business processe s.
Table 1 presents E-commerce features and benefits [43-45]. Examples of specific e-commerce business benefits are as follows [46-51]: 1. Reducing advertising costs; 2. Reduction of shipping costs, mainly for goods that can be obtained electronically; 3. Reduction of design and strategic planning costs; 4. Great opportunities for market research niches in the market; 5. Equal access to the market (for large corporations and small firms); 6. Access to new markets; 7. Involvement of customers in the development and implementation of new products and services Among the most critical legal issues that require an urgent decision with the participation of the world community are the following [52-57]: The order of taxation of transactions in electronic form; Tariffs; Requirements for the form of agreements and liability; Regulation of cryptography; Authentication rules; Protection of content; Consumer protection.
E-commerce now affects the economy and citizens’ rights. Legal requirements are required to develop a global and open market by harmonizing legislation and simplifying the rules and procedures applicable in different countries. Developing such norms requires in-depth cooperation between business and government, which would provide significant benefits to manufacturers and consumers worldwide. Today, underdevelopment in general or fragmentation of e-commerce legal norms and significant contradictions between the laws of different countries are obstacles to the online economy’s successful functioning. Businesses often exchange commercial content and make payments electronically and with their customers. Traditionally, systems based on EDI (Electronic Data Interchange) and EFT (Electronic Financial Transfer) standards have used to impose stringent conditions on the forms of content transmitted. Mainly large companies, corporations with private VAN networks [58] use such systems. Nowadays, both companies and individuals can conduct business transactions over the Internet without EDI. Commercial content transmitted over the Internet can be divided into two categories: information transactions; financial transactions.
Information provisioning is a fundamental and expensive element of e-commerce. Business information can take several forms [59-63]: Statistics (figures, graphs, analyses); Corporate information (telephone numbers, addresses, organization structure); Product or service information; Paid content (news, periodicals, access to databases, etc.) There are three main classes of financial transactions [59-63]: Company-to-company (payments between banks, other financial institutions, transfers of funds from one firm’s account to another’s account are usually made by electronic transfer or by check); Company-to-client (the primary payment method is cash, checks, debit and credit cards); Client-to-client (the primary payment method is cash and reviews).
Using the Internet to perform these types of transactions allows you to replace the presentation or display of cash, checks, and credit cards with their electronic equivalents.
Today, there are specially designed electronic versions of payment systems through which you can: order and pay for a purchase, bank payments, investments (insecurities), etc.
The electronic payment system is an authorized information system designed to make payments on the Internet between financial, commercial, manufacturing, governmental organizations and individual users. One of the main requirements for payment systems is [58-63]: 1. Confidentiality; 2. Preserving the integrity of content; 3. Providing payment methods; 4. Providing authentication; 5. Authorization; 6. Minimization of transaction fee.
Payment systems can be classified [64-67]: 1. By payment type:
Credit card-based systems; Systems based on Internet banking; Systems based on electronic checks; Systems using electronic money;
Smart card-based systems; 2.
According to the payment scheme:
Credit;
Debit; 3. By prevalence: a. International;
National. follows:
Each virtual enterprise pі independently chooses the type of payment system tі, which is the best option for conducting financial transactions. However, there is no single C system for conducting financial transactions between virtual enterprises with different payment systems in the virtual ebusiness world. There is no single standard C system for controlling financial transactions between participants in the virtual e-business world. Petri networks can represent the system of economic relationships through payment systems (processes) between virtual enterprises (participants) in t he Internet world of e-commerce. Such a C system will consist of four elements [68]: A finite set of participants Р = {p1, p2,..., pn} of the electronic market system;
Multiple processes of relationships Т = {t1, t2,..., tm}; 3. Input function І:TP, that is, direct links between processes and participants;
Output function О:TP (direct links between participants and procedures). (Fig. 2)
The system of conducting and controlling financial transactions between virtual enterprises is as
= ( , , , ).
The power of many virtual enterprises Р - |P| = n.
The power of multiple payment systems Т - |T| = m.
A participant (a virtual e-commerce enterprise) pi is the input position of the process ti , if pi I ti [68-73]. Similarly, рі is the output position of the process tі, if рі О(tі). Process inputs and outputs are kits of posts. The equipment (the kit)is a generalization of the set. Repeatedly repetitive identical items may be included in the kit. An example of the kit is O(t1) = {p2, p3, p4, p4}.
The multiplicity of the input position pi for the process tj is the number of occurrences of the position in the input kit of process #(pi, І(tj)) [68-73]. Similarly, the multiplicity of the output position pi for process tj is the number of position occurrences in the original set of process #(pi, О(tj)).
#(tj, І(pi)) = #(pi, O(tj)).
#(tj, O(pi)) = #(pi, І(tj)).
The graphical representation of the financial relationship system in the virtual world of e-commerce is a bipartite oriented multigraph. The structure of the system consists of a set of participants and processes. Accordingly, the graph of a system consists of two types of nodes – the circle denotes the participant and the bar | denotes the process. Therefore, since the vertices of a graph can be divided into two sets - participants and methods, the chart is bipartite.
Oriented arcs (arrows) connect participants and processes, with some arcs directed from participants
to methods and some from operations to participants. Such arcs indicate the input and output kits.
Multiple arcs represent multiple outputs. Since there can be various arcs between vertices, we have a
multigraph. The arcs are directional, so this multigraph is oriented.
(
Thus, graph G of the financial relationship system in the virtual world of e-business is a bipartite oriented multigraph,
G = (V,A),
where V = {v1, v2,..., vs} is the set of vertices, А = {а1, а2,..., аr} is set of directional arcs, аі = (vi, vk),
where vi, vk V. The set V can be divided into two non-intersecting subsets - Р і Т, so V=PT, PT=
and for any arc аі(vi, vk) V one of two conditions is fulfilled: or vi Т, vk Р, or vi Р, vk Т. The
kit of directional arcs А, corresponding to this system is defined as follows [68]:
#((рі, tj), А) = #(pi, І(tj)); (
The dual graph of the system С = (Р, Т, І, О is the graph of the system C = (T, P, І, О), that is, the graph in which participants and processes are reversed. Such diagrams are of no practical interest, so we do not consider them.
Marking is assigning a certain number of chips (monetary units) to the members of the relationship system. Marking are user resources, that is, the resources of a virtual e-commerce enterprise that it can use to conduct transactions in the scheme of financial relationships of e-commerce participants.
The marking of the system С = (Р, Т, І, О) is a function that reproduces the participants Р set into the nonnegative integers N set. The marking can also be represented as an n-dimensional vector = (1, 2,..., n), where n = |P|, i N. The vector defines for each participant рі the number of chips in this position і. More commonly, marking is used as a function (рі) = і. The marked system М = (С, ) is the set of the structure of the system С = (Р, Т, І, О) and the marking . The chips on the graph are indicated as a point inside the circle, indicating the given position of the participant.
Running the system С - is a redistribution of chips by running processes. When a function is started, the fragments are removed from their input member positions and added to their output. The process can only be activated when allowed. A transition is called allowed if each of its input positions has a chip count of at least the number of arcs from a place in the process. The chips in the entry position that allows the process is called permits. The process tj T in the marked Petri net С = (Р, Т, І, О) with marking is allowed if the condition for all рі Р is satisfied [68]
(рі) #(рі, І( )). (
The process is started by removing all the allowance chips from their output positions and then adding the chips to each of their initial positions, one chip for each arc. The process of tj in a marking system with marking can be started whenever it is allowed. As a result of triggering the allowed transition tj a new marking is formed , which is determined by this ratio [68]:
(рі) = (рі) - #(pi, I(tj)) + #(pi, O(tj)). (7)
Its marking determines the status of system С. Running any of the processes changes the state of the system by changing its marking. The states space of a system having n participants is the set of all markings, i.e. Nn. A change in system state is caused by a process start-up and is determined by a function , called the following state function. The part of the next state for the system С = (Р, Т, І, О) with marking and the process tj Р is defined if and only if (рі) #(pi, I(tj)) for all pi P. If (рі) is defined, then (, tj) = , where (рі) = (рі) - #(pi, I(tj)) + #(pi, O(tj)) for all рі Р [68]. Suppose that we have the system С = (Р, Т, І, О) with initial marking 0. The sequential starting of processes can execute this network. Starting a legal tj process will result in the formation of a new mark 1 = (0, tj). Any other allowed process, such as tk, can be started in the resulting marked graph. We obtain the marking 2 = (1, tk). You can continue this process until at least one method is allowed.
When running the system, we get two sequences. The sequence of markings (0, 1, 2,... ) and the sequence of processes that are started (tj1, tj2, tj3,... ). These sequences are related by the following relation [68]:
(k, tjk) = k+1, k = 0,1,2,... (8)
With the help of this relation, having the structure of the system С = (Р, Т, І, О), the initial marking 0 and the sequence of states (tj1, tj2, tj3,... ) we can obtain the series (0, 1, 2,... ). The reverse process from markings to the series of processes is possible in most cases, except some degenerate cases. For a system С = (Р, Т, І, О) with marking marking is called directly achievable from , if there is a process tj T, such that (, tj) = , i.e. can be obtained from running one of the allowed processes. The set of reach R(C, ) for the system С = (Р, Т, І, О) with marking - is the most miniature set of markings defined as follows: 1. R; 2. if R, then for all = (, tj), tj T is fulfilled R, that is, if some marking belongs to the set of reach, then all reachable markings from it also belong to this set.
Often use the concept of the advanced function of the next state. The extended input state function is defined for the marking and the sequence of processes = (tj1, tj2,..., tjn) by this rule [68-73]: = (, ) = (. . . ((, 1), 2). . . , ), (9) that is, we first start the process tj1, then tj2 and so on, to tjn.
The reach tree of system С is an illustration of the reach set R(C, ). Since the reach is infinite in many cases, specific rules allow it to be reflected by the finite reach tree.
The following rules are used to bring the reaching tree to a complete view [68-73].
Passive markings - markings in which no processes are allowed are terminal vertices of the reaching tree (do not have their subtree reach).
If the marking received has already encountered in the reaching tree before, such markings are called duplicate vertices. Consideration of the same vertices is unnecessary because the resulting subtree will be similar. If in the reaching tree we get a mark that is greater than one of the previous (on the way to the root) markings, then in all positions we put the symbol .
With these three rules, a set of the reach of any system can be represented as a finite tree.
For example, we will build a system of financial relationships between the following participants (Fig. 2): user - client (р1), Internet - auction (р2), customer’s bank and on-line auction (р3), Internet – shop (р4), bank Internet - shop (р5). Similarly, we have the following processes: payment for a product/service by a customer (t1), financial interaction process between an Internet - auction, an Internet - shop and a customer’s bank (t2), financial transactions between an Internet - shop and its bank (t3), transfer money at the level of banks and an Internet - shop (t4, t5).
For the example of the system above, the extended input and output functions show in Fig. 3. t1 t2 p3 p3 t5 p4 Figure 5: The marking received due to the start of the transition t4 of the graph depicted in Fig. 4.
(
t2
t1
t2
t3
t3
(
(
2 t
(
(
For example, consider the system shown in Fig. 4. This initial marking 0 = (
Consider the sequence of starts t1t1t1. Each time, the marking received differs from the previous one
only by the number of chips in p2 - p3 (increases by one) and in p4 (increases by two units). Running the
process t1 many times, you can get any number of chips in it. Since starting the process increases the
number of chips in one (several) positions while leaving the number of fragments unchanged in others,
t1 t2t4 t5
t5
(
(
Regardless of the type of e-commerce, the primary indicator of the impact on the profit of ecommerce is the involvement of potential audiences to visit the relevant e-commerce sites and their further motivation to make the applicable orders. Appropriate e-commerce referral systems should withstand the worst of motivating the user to place an order. Timely further analysis of the statistics of visits and conversions depends on the formation of new goals to encourage potential e-commerce customers. The e-commerce system model based on the Petri net allows determining the set of financial relations between the participants of e-business. Analysis of labelled systems based on the Petri net enables you to build a tree of achievement of e-commerce goals and finding the relationships between site visits and e-business profits. Further application of the methods of correlation and cluster analysis to the obtained data allows classifying the purposes of conversions depending on the market reaction and the current time for the necessary adjustment. The requires: Build a correlation field; Determine the value of the correlation coefficient; Calculate the correlation ratio; Plot graphs of autocorrelation functions; Break one of the sequences into three equal parts; Build a correlation matrix for them; - find the coefficients of multiple correlation.
First, will pre-process the data of the sessions of visiting the e-commerce site when the conversion is achieved and graphic representation of these results (Fig. 8-10): df <- read.csv("D:/V20.csv", header = TRUE) df1 <- read.csv("D:/V20.csv", header = TRUE) head(df) summary(df) str(df) df$Індекс_дня <- as.Date(df$Індекс_дня, format = "%m/%d/%y")
df$Month <- as.Date(cut(df$Індекс_дня, breaks = "month")) df$Week <as.Date(cut(df$Індекс_дня, breaks = "week", start.on.monday = TRUE)) df$Day <- as.Date(cut(df$Індекс_дня, breaks = "day")) ggplot(data = df, aes(x = df$Індекс_дня, y = df$Сеанси)) + geom_point(data = df) + labs(x = "Month", y = "Total Sessions", title = "Sessions/month") + geom_line(data = df) r1 <- ggplot(data = df, aes(x = df$Month, y = df$Сеанси)) + stat_summary(fun.y=sum, geom="line", colour="mediumvioletred")+ scale_x_date(labels = date_format("%m"), breaks = "1 month",
guide = guide_axis(n.dodge=1)) + labs(x="Month", y="Total Sessions", title="Sessions per month")
Detailed analysis of sessions per weeks is more informative than sessions per month (Fig 11-12). r2 <- ggplot(data = df, aes(x = df$Week, y = df$Сеанси)) + stat_summary(fun.y = sum, geom = "line") + scale_x_date(labels = date_format("%m-%d"), breaks = "1 week", guide = guide_axis(n.dodge=3)) + labs(x = "Day", y = "Total Sessions", title = "Sessions per week")
data <- sessions_sum ggplot(data) + geom_point(aes(x=data$Week, y=data$sum_precip)) + geom_line(aes(x=data$Week, y=data$sum_precip), colour = "gold") + xlab("Month") + ylab("Num") +
ggtitle("Sessions sum by month")
Fig. 16-18 show the graphic representation of cumulative attendance sessions sessions_sum <- df %>% group_by(Month) %>% group_by(Day) %>% summarise(sum_precip = sum(Сеанси)) %>% print(sum(df1$Сеанси)) plot(ecdf(df[,"Сеанси"]), col = "green") num.of.sessions = df1$Сеанси breaks = seq(0, 36, by=1) num.cut = cut(num.of.sessions, breaks, right=FALSE) num.freq = table(num.cut) freq = seq(0, cumsum(num.freq)/max(cumsum(num.freq)), by=0.1) # кумулята plot(breaks, freq, main="Sessions frequency", xlab="Num of sessions", ylab="Cumulative frequencies", xlim=c(0,36), ylim=c(0.0,1.0)) lines(breaks, freq)
polar <- r2 + coord_polar() data.SMA3 <- SMA(data.ts,n=3) data.SMA5 <- SMA(data.SMA3,n=5) data.SMA7 <SMA(data.SMA5,n=7) data.SMA9 <- SMA(data.SMA7,n=9) data.SMA11 <SMA(data.SMA9,n=11) data.SMA13 <- SMA(data.SMA11,n=13) data.SMA15 <SMA(data.SMA13,n=15)
tP1 <- turnpoints(na.remove(data.SMA3)) plot(tP1) tP2 <turnpoints(na.remove(data.SMA5)) plot(tP2) tP3 <turnpoints(na.remove(data.SMA7)) plot(tP3) tP4 <turnpoints(na.remove(data.SMA9)) plot(tP4) tP5 <turnpoints(na.remove(data.SMA11)) plot(tP5) tP6 <turnpoints(na.remove(data.SMA13)) plot(tP6) tP7 <turnpoints(na.remove(data.SMA15)) plot(tP7)
Fig. 28 shows detection of the trend of the time series by the weighted moving average method. ma(data.ts, order = 12, centre = TRUE) autoplot(data.ts, series = "Data") + autolayer(ma(data.ts, order = 12, centre = T), series = "normal") + ggtitle("Daily Sessions") + labs(x = NULL, y = "Sessions")
Fig. 30 shows detection of the trend of the time series by smoothing Kendall formulas.
sm <- ksmooth(format(df$Day,'%d'), df$Сеанси, kernel = "normal", bandwidth
= 0.5) plot(sm,xlab='day',ylab='num')
lines(sm,col='blue',lwd=2.5)
coll <- colorRampPalette(c("blue", "white", "red"))(60) heatmap(x = res1,
col = coll, symm = TRUE, margins = c(7,7), scale="column", cexCol=0.7,
cexRow=0.7) legend(x="bottomright", legend=c("min", "ave", "max"),
fill=colorRampPalette(c("blue", "white", "red"))(
Fig. 33 shows detection of the trend of the time series by Autocorrelation analysis of time sequences. acf(data.ts, lag.max = 10, plot = FALSE) acf(na.remove(data.SMA15))
Fig. 35 shows detection of the trend of the time series by Median smoothing. plot(smooth(data.ts, kind = "3", twiceit = FALSE, endrule = c("Tukey", "copy"), do.ends = FALSE), ylab='num')
Fig. 36 shows detection of the time series trend by hierarchical agglomerative cluster analysis of multidimensional data.
Fig. 37-38 show construction of the proximity matrix through the rationing of the table "objectproperty" (first 5 records). array = data.values # separate array into input and output components X = array[:,1:7] Y = array[:,0] scaler = Normalizer().fit(X) normalizedX = scaler.transform(X) # summarize transformed data numpy.set_printoptions(precision=3) print('Сеанси', 'Відсоток_нових_сеансів', 'Нові_користувачі',
'Показник_відмов', 'Сторінок_за_сеанс', 'Сер_тривалість_сеансу') print(normalizedX[:10,:])
Fig. 39-41 show conducting agglomerative hierarchical cluster analysis. classifiers = [] model1 = xgboost.XGBClassifier() classifiers.append(model1) model2 = svm.SVC() classifiers.append(model2) model3 = tree.DecisionTreeClassifier() classifiers.append(model3) model4 = RandomForestClassifier() classifiers.append(model4) for clf in classifiers:
clf.fit(X_train, y_train) y_pred = clf.predict(X_test) acc = accuracy_score(y_test, y_pred) print("Accuracy of {} is {}".format(clf, acc)) cm = confusion_matrix(y_test, y_pred)
print("Confusion Matrix of {} is {}".format(clf, cm))
E-commerce-related legal issues can profoundly affect its development, but they are still waiting to be resolved. They relate to privacy, e-commerce taxation and control over the export of cryptographic technology and legal, financial relationships between participants in the virtual e-commerce world. No matter how the company works, electronic transformation is no longer a matter of choice. It is already a necessity for successful business development.
[7] K. Alieksieieva, A. Berko, V. Vysotska, Technology of commercial web-resource processing, in: Proceedings of 13th International Conference: The Experience of Designing and Application of CAD Systems in Microelectronics, CADSM 2015-February, 2015. [8] V. Vysotska, L. Chyrun, L. Chyrun, The Commercial Content Digest Formation and Distributional Process, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2016, pp. 186-189. [9] V. Lytvyn, V. Vysotska, V. Shatskykh, I. Kohut, O. Petruchenko, L. Dzyubyk, V. Bobrivetc, V.
Panasyuk, S. Sachenko, M. Komar, Design of a recommendation system based on Collaborative
Filtering and machine learning considering personal needs of the user, volume 4(
Data Component, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 358-373.
[13] R. Yurynets, Z. Yurynets, I. Myshchyshyn, N. Zhyhaylo, A. Pekhnyk, Optimal Strategy for the
Development of Insurance Business Structures in a Competitive Environment, volume Vol-2631
of CEUR Workshop Proceedings, 2020, pp. 79-94.
[14] R. Yurynets, Z. Yurynets, M. Kokhan, Econometric Analysis of the Impact of Expert Assessments
on the Business Activity in the Context of Investment and Innovation Development, volume
Vol2604 of CEUR workshop proceedings, 2020, pp. 680-694.
[15] H. Lipyanina, A. Sachenko, T. Lendyuk, S. Nadvynychny, S. Grodskyi, Decision tree based
targeting model of customer interaction with business page, volume Vol-2608 of CEUR Workshop
Proceedings, 2020, pp. 1001-1012.
[16] I. Oksanych, I. Shevchenko, I. Shcherbak, S. Shcherbak, Development of specialized services for
predicting the business activity indicators based on micro-service architecture, volume 2(
Problems of Economics, 2010, pp. 147-152. [20] A.Y. Berko, Methods and models of data integration in E-business systems, volume 10 of Actual
Problems of Economics, 2008, pp. 17-24.
[21] A. Berko, Consolidated data models for electronic business systems, in: Proceedings of The
Experience of Designing and Application of CAD Systems in Microelectronics, CADSM, 2007,
pp. 341-342.
[22] O.cKuzmin, M. Bublyk, Economic evaluation and government regulation of technogenic
(manmade) damage in the national economy, in: Computer sciences and information technologies
(CSIT), 2016, pp. 37–39.
[23] A. Demchuk, V. Lytvyn, V. Vysotska, M. Dilai, Methods and Means of Web Content
Personalization for Commercial Information Products Distribution, volume 1020 of Advances in
Intelligent Systems and Computing, 2020, pp. 332–347.
[24] A. Gozhyj, I. Kalinina, V. Vysotska, S. Sachenko, R. Kovalchuk, Qualitative and Quantitative
Characteristics Analysis for Information Security Risk Assessment in E-Commerce Systems,
volume Vol-2762 of CEUR Workshop Proceedings, 2020, pp. 177-190.
[25] V. Vysotska, A. Berko, M. Bublyk, L. Chyrun, A. Vysotsky, K. Doroshkevych, Methods and tools
for web resources processing in e-commercial content systems, in: Proceedings of 15th
International Scientific and Technical Conference on Computer Sciences and Information
Technologies, CSIT, 1, 2020, pp. 114-118.
[26] O. Artemenko, V. Pasichnyk, N. Kunanets, K. Shunevych, Using sentiment text analysis of user
reviews in social media for e-tourism mobile recommender systems, volume Vol-2604 of CEUR
workshop proceedings, 2020, pp. 259-271.
[27] B. Rusyn, L. Pohreliuk, O. Kapshii, J. Varetskyy, A. Demchuk, I. Karpov, A. Gozhyj, V. Gozhyj,
I. Kalinina, An Intelligent System for Commercial of Information Products Distribution Based
SEO and Sitecore CMS, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 760-777.
[28] B.E. Kapustiy, B.P. Rusyn, V.A. Tayanov, Peculiarities of Application of Statistical Detection
Criteria for Problem of Pattern Recognition, volume 37(
Rusyn, L. Pohreliuk, T. Salo, Identifying Textual Content Based on Thematic Analysis of Similar Texts in Big Data, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2019, pp. 84-91. [34] O. Veres, B. Rusyn, A. Sachenko, I. Rishniak, Choosing the mathod of finding similar images in the reverse search system, volume Vol-2136 of CEUR Workshop Proceedings, 2018, pp.99-107. [35] V. Husak, O. Lozynska, I. Karpov, I. Peleshchak, S. Chyrun, A. Vysotskyi, Information System for Recommendation List Formation of Clothes Style Image Selection According to User’s Needs Based on NLP and Chatbots, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 788818. [36] S. Makara, L. Chyrun, Y. Burov, Z. Rybchak, I. Peleshchak, R. Peleshchak, R. Holoshchuk, S.
Kubinska, A. Dmytriv, An Intelligent System for Generating End-User Symptom Recommendations Based on Machine Learning Technology, volume Vol-2604 of CEUR Workshop Proceedings, 2020, pp. 844-883. [37] Y. Matseliukh, V. Vysotska, M. Bublyk, Intelligent System of Visual Simulation of Passenger
Flows, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 906-920. [38] M. Bublyk, Y. Matseliukh, U. Motorniuk, M. Terebukh, Intelligent system of passenger transportation by autopiloted electric buses in Smart City, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 1280-1294. [39] I. Bodnar, M. Bublyk, O. Veres, O. Lozynska, I. Karpov, Y. Burov, P. Kravets, I. Peleshchak, O.
Vovk, O. Maslak, Forecasting the risk of cervical cancer in women in the human capital development context using machine learning, volume Vol-2631 of CEUR workshop proceedings, 2020, pp. 491-501. [40] L. Podlesna, M. Bublyk, I. Grybyk, Y. Matseliukh, Y. Burov, P. Kravets, O. Lozynska, I. Karpov, I. Peleshchak, R. Peleshchak, Optimization model of the buses number on the route based on queueing theory in a smart city, volume Vol-2631 of CEUR workshop proceedings, 2020, pp. 502515. [41] D. Koshtura, M. Bublyk, Y. Matseliukh, D. Dosyn, L. Chyrun, O. Lozynska, I. Karpov, I.
Peleshchak, M. Maslak, O. Sachenko, Analysis of the demand for bicycle use in a smart city based on machine learning, volume Vol-2631 of CEUR workshop proceedings, 2020, pp. 172-183. [42] O. Soprun, M. Bublyk, Y. Matseliukh, V. Andrunyk, L. Chyrun, I. Dyyak, A. Yakovlev, M.
Emmerich, O. Osolinsky, A. Sachenko, Forecasting temperatures of a synchronous motor with Sciences and Information Technologies, CSIT, 2019, pp. 15-20. DOI: 10.1109/STCCSIT.2019.8929790 [60] A. Vysotsky, V. Lytvyn, V. Vysotska, D. Dosyn, I. Lyudkevych, N. Antonyuk, O. Naum, A.
Vysotskyi, L. Chyrun, O. Slyusarchuk, Online Tourism System for Proposals Formation to User Based on Data Integration from Various Sources, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2019, pp. 92-97. DOI: 10.1109/STCCSIT.2019.8929849 [61] L. Chyrun, A. Kowalska-Styczen, Y. Burov, A. Berko, A. Vasevych, I. Pelekh, Y. Ryshkovets, Heterogeneous Data with Agreed Content Aggregation System Development, volume Vol-2386 of CEUR Workshop Proceedings, 2019, pp. 35-54. [62] L. Chyrun, Y. Burov, B. Rusyn, L. Pohreliuk, O. Oleshek, A. Gozhyj, I. Bobyk, Web Resource Changes Monitoring System Development, volume Vol-2386 of CEUR Workshop Proceedings, 2019, pp. 255-273. [63] L. Chyrun, A. Gozhyj, I. Yevseyeva, D. Dosyn, V. Tyhonov, M. Zakharchuk, Web Content Monitoring System Development, volume Vol-2362 of CEUR Workshop Proceedings, 2019, pp. 126-142. [64] N. Chukhray, N. Shakhovska, O. Mrykhina, M. Bublyk, L. Lisovska, Methodical Approach to Assessing the Readiness Level of Technologies for the Transfer, volume 1080 of Advances in Intelligent Systems and Computing IV, 2020, pp. 259-282. [65] N. Chukhray, N. Shakhovska, O. Mrykhina, M. Bublyk, L. Lisovska, Consumer aspects in assessing the suitability of technologies for the transfer, in: Computer sciences and information technologies, (CSIT), 2019, pp. 142–147. [66] M. Bublyk, O. Rybytska, A. Karpiak, Y. Matseliukh, Structuring the fuzzy knowledge base of the
IT industry impact factors, in: Computer sciences and information technologies (CSIT), 2018.
[67] M.I. Bublyk, O. M. Rybytska, The model of fuzzy expert system for establishing the pollution
impact on the mortality rate in Ukraine, in: Computer sciences and information technologies
(CSIT), 1, 2017, pp. 253–256.
[68] J. L. Peterson, Petri nets, volume of 9(
Dynamic System Modeling, 2007, p. 24.
[71] M. Wang, G. Liu, C. Yan, C. Jiang, Modeling and vulnerable points analysis for e-commerce
transaction system with a known attack, in: International Conference on Security, Privacy and
Anonymity in Computation, Communication and Storage, 2016, pp. 422-436.
[72] V. Jain, S. Wadhwa, S. G. Deshmukh, e-Commerce and supply chains: Modelling of dynamics
through fuzzy enhanced high level petri net, volume 30(