Application of neural networks in OLAP-systems
The article highlights the main characteristics of OLAP systems that perform online analytical data processing. These systems, based on OLAP technology, are widely used both in government agencies and in private ones. The main characteristics, features and structure of OLAP systems are mentioned. Th...
Збережено в:
| Дата: | 2024 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
PROBLEMS IN PROGRAMMING
2024
|
| Теми: | |
| Онлайн доступ: | https://pp.isofts.kiev.ua/index.php/ojs1/article/view/658 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Problems in programming |
| Завантажити файл: |  |
Репозитарії
Problems in programming| id |
pp_isofts_kiev_ua-article-658 |
|---|---|
| record_format |
ojs |
| resource_txt_mv |
ppisoftskievua/cc/81bf3c83be3c7eebb6f5662beec1bacc.pdf |
| spelling |
pp_isofts_kiev_ua-article-6582025-02-15T15:10:35Z Application of neural networks in OLAP-systems Застосування нейронних мереж в OLAP-системах Nabibayeva, G.Ch. data warehouse; OLAP; artificial intelligence; machine learning; neural network, forecasting; clustering; classification UDC 004.942 сховище даних; OLAP; штучний інтелект; машинне навчання; нейронна мережа; прогнозування; кластеризація; класифікація УДК 004.942 The article highlights the main characteristics of OLAP systems that perform online analytical data processing. These systems, based on OLAP technology, are widely used both in government agencies and in private ones. The main characteristics, features and structure of OLAP systems are mentioned. The article emphasizes that OLAP is a data warehousing tool. OLAP allows analysts to explore and navigate a multidimensional structure of indicators called a data cube or OLAP cube. Indicators (measures) of OLAP cubes play an important role in the decision-making process. To solve some problems, these measures often need to be classified or clustered. Moreover, empty measures are common in OLAP cubes. Empty measures can present due to nonexisting facts in data warehouse or due to empty cells which are unfilled in by mistake. The presence of empty measures negatively impacts strategic decision making. Unfortunately, OLAP itself is poorly adapted for forecasting empty measures of data cubes. Over the years, researchers and analysts have tried to improve the decision-making process in OLAP systems and add forecasting and other options to OLAP applications. Today, in the era of Industry 4.0, with the availability of big data, there is a need to apply new technologies to solve such problems. These technologies include neural networks. The article examines the problem of integrating OLAP and a neural network. In this regard, the article provides information about neural networks: information about their properties, types, as well as their capabilities. The article shows the possibility and advantages of integrating OLAP and neural network. It mentions that in the case of big data, the integration of OLAP and neural networks is very effective for solving problems of classification, clustering and prediction of empty measures of OLAP cubes. An architectural and technological model for integrating OLAP and neural networks is presented. It is noted what types of neural networks can be used to solve the problems of classification, clustering and forecasting specified in the model.Prombles in programming 2024; 2-3: 367-374 У статті висвітлено основні характеристики OLAP-систем, що здійснюють оперативну аналітичну обробку даних. Ці системи, засновані на технології OLAP, широко використовуються як в державних установах, так і в приватних. Наведено основні характеристики, особливості та структуру OLAP-систем. У статті підкреслюється, що OLAP є інструментом сховища даних. OLAP дозволяє аналітикам досліджувати та орієнтуватися в багатовимірній структурі індикаторів, яка називається кубом даних або OLAP-кубом. Індикатори (заходи) кубів OLAP відіграють важливу роль у процесі прийняття рішень. Щоб вирішити деякі проблеми, ці заходи часто потрібно класифікувати або кластеризувати. Крім того, порожні міри часто зустрічаються в кубах OLAP. Порожні показники можуть бути через неіснуючі факти в сховищі даних або через порожні клітинки, які помилково не заповнені. Наявність порожніх заходів негативно впливає на прийняття стратегічних рішень. На жаль, сам OLAP погано підходить для прогнозування порожніх розмірів кубів даних. Протягом багатьох років дослідники та аналітики намагалися вдосконалити процес прийняття рішень у системах OLAP і додати прогнозування та інші параметри до програм OLAP. Сьогодні, в епоху Індустрії 4.0, з доступністю великих даних виникає потреба застосовувати нові технології для вирішення таких проблем. Ці технології включають нейронні мережі. У статті розглядається проблема інтеграції OLAP і нейронної мережі. У зв'язку з цим наводяться відомості про нейронні мережі: їхні властивості, види, а також їхні можливості. У статті розглядаються можливості та переваги інтеграції OLAP та нейронної мережі. Зроблено висновок, що у випадку великих даних, інтеграція OLAP і нейронних мереж дуже ефективна для вирішення проблем класифікації, кластеризації та прогнозування порожніх мір кубів OLAP. Представлено архітектурно-технологічну модель інтеграції OLAP і нейронних мереж. Відзначено, які типи нейронних мереж можна використовувати для вирішення заданих у моделі задач класифікації, кластеризації та прогнозування. Prombles in programming 2024; 2-3: 367-374 PROBLEMS IN PROGRAMMING ПРОБЛЕМЫ ПРОГРАММИРОВАНИЯ ПРОБЛЕМИ ПРОГРАМУВАННЯ 2024-12-17 Article Article application/pdf https://pp.isofts.kiev.ua/index.php/ojs1/article/view/658 10.15407/pp2024.02-03.367 PROBLEMS IN PROGRAMMING; No 2-3 (2024); 367-374 ПРОБЛЕМЫ ПРОГРАММИРОВАНИЯ; No 2-3 (2024); 367-374 ПРОБЛЕМИ ПРОГРАМУВАННЯ; No 2-3 (2024); 367-374 1727-4907 10.15407/pp2024.02-03 en https://pp.isofts.kiev.ua/index.php/ojs1/article/view/658/710 Copyright (c) 2024 PROBLEMS IN PROGRAMMING |
| institution |
Problems in programming |
| baseUrl_str |
https://pp.isofts.kiev.ua/index.php/ojs1/oai |
| datestamp_date |
2025-02-15T15:10:35Z |
| collection |
OJS |
| language |
English |
| topic |
data warehouse; OLAP; artificial intelligence; machine learning; neural network forecasting; clustering; classification UDC 004.942 |
| spellingShingle |
data warehouse; OLAP; artificial intelligence; machine learning; neural network forecasting; clustering; classification UDC 004.942 Nabibayeva, G.Ch. Application of neural networks in OLAP-systems |
| topic_facet |
data warehouse; OLAP; artificial intelligence; machine learning; neural network forecasting; clustering; classification UDC 004.942 сховище даних OLAP штучний інтелект машинне навчання нейронна мережа прогнозування кластеризація класифікація УДК 004.942 |
| format |
Article |
| author |
Nabibayeva, G.Ch. |
| author_facet |
Nabibayeva, G.Ch. |
| author_sort |
Nabibayeva, G.Ch. |
| title |
Application of neural networks in OLAP-systems |
| title_short |
Application of neural networks in OLAP-systems |
| title_full |
Application of neural networks in OLAP-systems |
| title_fullStr |
Application of neural networks in OLAP-systems |
| title_full_unstemmed |
Application of neural networks in OLAP-systems |
| title_sort |
application of neural networks in olap-systems |
| title_alt |
Застосування нейронних мереж в OLAP-системах |
| description |
The article highlights the main characteristics of OLAP systems that perform online analytical data processing. These systems, based on OLAP technology, are widely used both in government agencies and in private ones. The main characteristics, features and structure of OLAP systems are mentioned. The article emphasizes that OLAP is a data warehousing tool. OLAP allows analysts to explore and navigate a multidimensional structure of indicators called a data cube or OLAP cube. Indicators (measures) of OLAP cubes play an important role in the decision-making process. To solve some problems, these measures often need to be classified or clustered. Moreover, empty measures are common in OLAP cubes. Empty measures can present due to nonexisting facts in data warehouse or due to empty cells which are unfilled in by mistake. The presence of empty measures negatively impacts strategic decision making. Unfortunately, OLAP itself is poorly adapted for forecasting empty measures of data cubes. Over the years, researchers and analysts have tried to improve the decision-making process in OLAP systems and add forecasting and other options to OLAP applications. Today, in the era of Industry 4.0, with the availability of big data, there is a need to apply new technologies to solve such problems. These technologies include neural networks. The article examines the problem of integrating OLAP and a neural network. In this regard, the article provides information about neural networks: information about their properties, types, as well as their capabilities. The article shows the possibility and advantages of integrating OLAP and neural network. It mentions that in the case of big data, the integration of OLAP and neural networks is very effective for solving problems of classification, clustering and prediction of empty measures of OLAP cubes. An architectural and technological model for integrating OLAP and neural networks is presented. It is noted what types of neural networks can be used to solve the problems of classification, clustering and forecasting specified in the model.Prombles in programming 2024; 2-3: 367-374 |
| publisher |
PROBLEMS IN PROGRAMMING |
| publishDate |
2024 |
| url |
https://pp.isofts.kiev.ua/index.php/ojs1/article/view/658 |
| work_keys_str_mv |
AT nabibayevagch applicationofneuralnetworksinolapsystems AT nabibayevagch zastosuvannânejronnihmerežvolapsistemah |
| first_indexed |
2025-07-17T09:57:49Z |
| last_indexed |
2025-07-17T09:57:49Z |
| _version_ |
1850410399969574912 |
| fulltext |
367
Експертні та інтелектуальні інформаційні системи, штучний інтелект
УДК 004.942 http://doi.org/10.15407/pp2024.02-03.367
G.Ch. Nabibayova
APPLICATION OF NEURAL NETWORKS IN OLAP SYSTEMS
The article highlights the main characteristics of OLAP systems that perform online analytical data processing.
These systems, based on OLAP technology, are widely used both in government agencies and in private ones.
The main characteristics, features and structure of OLAP systems are mentioned. The article emphasizes that
OLAP is a data warehousing tool. OLAP allows analysts to explore and navigate a multidimensional structure
of indicators called a data cube or OLAP cube. Indicators (measures) of OLAP cubes play an important role
in the decision-making process. To solve some problems, these measures often need to be classified or
clustered. Moreover, empty measures are common in OLAP cubes. Empty measures can present due to non -
existing facts in data warehouse or due to empty cells which are unfilled in by mistake. The presence of empty
measures negatively impacts strategic decision making. Unfortunately, OLAP itself is poorly adapted for
forecasting empty measures of data cubes. Over the years, researchers and analysts have tried to improve the
decision-making process in OLAP systems and add forecasting and other options to OLAP applications.
Today, in the era of Industry 4.0, with the availability of big data, there is a need to apply new technologies
to solve such problems. These technologies include neural networks. The article examines the problem of
integrating OLAP and a neural network. In this regard, the article provides information about neural networks :
information about their properties, types, as well as their capabilities. The article shows the possibility and
advantages of integrating OLAP and neural network. It mentions that in the case of big data, the integration
of OLAP and neural networks is very effective for solving problems of classification, clustering and prediction
of empty measures of OLAP cubes. An architectural and technological model for integrating OLAP and neural
networks is presented. It is noted what types of neural networks can be used to solve the problems of
classification, clustering and forecasting specified in the model.
Keywords – data warehouse, OLAP, artificial intelligence, machine learning, neural network, forecasting,
clustering, classification.
Г.Ч. Набібаєва
ЗАСТОСУВАННЯ НЕЙРОННИХ МЕРЕЖ
В OLAP-СИСТЕМАХ
У статті висвітлено основні характеристики OLAP-систем, що здійснюють оперативну аналітичну об-
робку даних. Ці системи, засновані на технології OLAP, широко використовуються як в державних
установах, так і в приватних. Наведено основні характеристики, особливості та структуру OLAP -сис-
тем. У статті підкреслюється, що OLAP є інструментом сховища даних. OLAP дозволяє аналітикам
досліджувати та орієнтуватися в багатовимірній структурі індикаторів, яка називається кубом даних
або OLAP-кубом. Індикатори (заходи) кубів OLAP відіграють важливу роль у процесі прийняття рі-
шень. Щоб вирішити деякі проблеми, ці заходи часто потрібно класифікувати або кластеризувати.
Крім того, порожні міри часто зустрічаються в кубах OLAP. Порожні показники можуть бути через
неіснуючі факти в сховищі даних або через порожні клітинки, які помилково не заповнені. Наявність
порожніх заходів негативно впливає на прийняття стратегічних рішень. На жаль, сам OLAP погано
підходить для прогнозування порожніх розмірів кубів даних. Протягом багатьох років дослідники та
аналітики намагалися вдосконалити процес прийняття рішень у системах OLAP і додати прогнозу-
вання та інші параметри до програм OLAP. Сьогодні, в епоху Індустрії 4.0, з доступністю великих
даних виникає потреба застосовувати нові технології для вирішення таких проблем. Ці технології
включають нейронні мережі. У статті розглядається проблема інтеграції OLAP і нейронної мережі. У
зв'язку з цим наводяться відомості про нейронні мережі: їхні властивості, види, а також їхні можливо-
сті. У статті розглядаються можливості та переваги інтеграції OLAP та нейронної мережі. Зроблено
висновок, що у випадку великих даних, інтеграція OLAP і нейронних мереж дуже ефективна для вирі-
шення проблем класифікації, кластеризації та прогнозування порожніх мір кубів OLAP. Представлено
архітектурно-технологічну модель інтеграції OLAP і нейронних мереж. Відзначено, які типи нейрон-
них мереж можна використовувати для вирішення заданих у моделі задач класифікації, кластеризації
та прогнозування.
Ключові слова – сховище даних, OLAP, штучний інтелект, машинне навчання, нейронна мережа, про-
гнозування, кластеризація, класифікація.
©Г.Ч. Набібаєва, 2024
ISSN 1727-4907. Проблеми програмування. 2024. №2-3
368
Експертні та інтелектуальні інформаційні системи, штучний інтелект
Introduction
People have always been eager to create
all the necessary conveniences for everyday life
for themselves. As a result of their efforts,
proper technologies and machines appeared to
assist them. One of the areas of such technolo-
gies is Artificial Intelligence (AI) [1]. The term
“Artificial Intelligence” was first coined in his-
tory in 1956 by John McCarthy. The term refers
to any system capable of performing creative
functions and solving problems that typically
require human intelligence, that is, which are
traditionally performed by humans. In effect, AI
imitates human intelligence in machines that are
programmed to think and act like humans. AI
plays a critical role in today’s world by enabling
automation, improving decision making, in-
creasing efficiency and productivity. It opens up
new opportunities for innovation and growth in
a variety of industries, including healthcare, fi-
nance, manufacturing, transportation, e-com-
merce, education, and more.
AI is frequently discussed and explored
jointly with machine learning (ML). ML is a
branch of AI. The idea of ML is that machines
should be able to learn and adapt through ex-
perience, making predictions based on statisti-
cal data collected by computers. Thus, the con-
cept of AI is a broader concept compared to the
concept of ML.
One of the areas of artificial intelli-
gence is neural networks (NN). NNs are used
to recognize hidden patterns in raw data, for
clustering [2] and classification [3], as well as
for solving tasks in the field of AI.
Recently, both government agencies
and private ones have widely used OLAP sys-
tems based on OLAP (Online Analytical Pro-
cessing) technology [4]. They are applied, for
example, in banking, medicine, industry, tele-
communications, trade, etc. [5] describes the
use of OLAP in the terminology environment,
namely in the terminology system to expand its
capabilities and for more efficient functioning.
NN and OLAP are essential tools for
quickly and efficiently discovering valuable,
non-obvious information from a large collec-
tion of data.
The goal of this paper is to study the
possibility of integrating OLAP and NN, and
to identify the benefits of this integration.
The first section describes the main
characteristics of OLAP systems. The second
section provides information about NN: histor-
ical background, types of NN and their func-
tions. The third section reviews related work.
The fourth section is devoted to the problem of
integrating NN and OLAP, and also presents
an architectural and technological model for
integrating OLAP and NN. Finally, the fifth
section presents the final conclusions of this ar-
ticle.
Main characteristics of OLAP
systems
An OLAP system is an information and
analytical data processing system developed
on OLAP technologies. The popularity of
OLAP is explained by the fact that it is possible
to solve many problems with its help, namely:
to implement operational processing of infor-
mation, including issuing information in vari-
ous sections and dynamic report generation
and its analysis based on the data obtained, to
perform monitoring and forecasting [6]. The
OLAP system is designed for generating re-
ports, constructing predictive scenarios and
performing statistical calculations based on
large collection of data with a complex struc-
ture [7]. The key components of the OLAP sys-
tem are a data warehouse (DW), an OLAP
server and applications.
DW is a source of processed infor-
mation accumulated from already existing sys-
tems of geographically distributed units. DW
is a domain-specific, non-volatile, integrated,
time-varying set of data for decision support
[8].
OLAP is an element of the DW and
takes advantage of its information.
The OLAP server is the core of the sys-
tem, with the help of which multidimensional
data structures are processed and communica-
tion between the DW and system users is en-
sured.
Applications are used for user work.
They formulate queries and visualize the re-
sponses received. OLAP applications are used
to store DW analysis contexts in multidimen-
sional data structures, i.e., in OLAP cubes.
OLAP cubes enable analysts to explore infor-
369
Експертні та інтелектуальні інформаційні системи, штучний інтелект
mation and report through interactive, easy-to-
use dashboards. It is OLAP cubes that contain
indicators (measures) used for analysis and
management decision-making.
One of the important goals of OLAP is
to make decisions based on historical data.
Note that OLAP provides any analyti-
cal report within a few seconds due to its wide
visualization functionality.
Neural network as a method in
artificial intelligence
The basic principles of NN operation
were described back in 1943 by Warren
McCulloch and Walter Pitts [9]. In 1957, neu-
roscientist Frank Rosenblatt developed the first
NN. He was the author of the first paper on per-
ceptrons [10]. In 2010, large amounts of train-
ing data opened up the possibility of using NN
for machine learning.
Each NN includes a first layer of neu-
rons called the input layer. This layer does not
perform any transformations or calculations; it
receives and distributes input signals to other
neurons. This layer is the only one that is com-
mon to all NN types.
The main types of NN are as follows:
Perceptron. Perceptrons are single-
layer or multilayer feed-forward artificial NNs
with binary or analog outputs that are super-
vised learning.
Single layer neural network. It is a
structure for the interaction of neurons, in
which signals from the input layer are immedi-
ately sent to the output layer. The output layer
converts the signal and immediately produces
a response.
Fig. 1. Example of a single layer neural
network
Multilayer neural network. This NN,
in addition to the output and input layers, has
several hidden intermediate layers. The num-
ber of these layers depends on the complexity
of the NN.
Fig. 2. Example of a multilayer neural
network
NN can be classified not only for the
number of layers, but also according to the di-
rection of information distribution along syn-
apses (connections) between neurons:
Feed-forward neural network (uni-
directional). In this structure (Fig. 3.) the sig-
nal moves strictly in the direction from the in-
put layer to the output layer.
Input
layer
Hidden
layers
Output
layer
Fig. 3. Architecture of a three-layer
feed-forward neural network.
Recurrent neural networks (with
feedback) (RNN). Here the signal moves both
forward and backward. As a result, the output
result can be returned to the input. Types of
RNN:
˗ One to one
˗ One to many
˗ Many to one
˗ many to many
Self-organizing maps. They include
self-organizing Kohonen maps. They are a
powerful, self-learning clustering engine: the
results are displayed in compact and easy-to-
interpret two-dimensional maps. The Kohonen
Output layerHiden layersInput layer
Output signalsLayer of neuronsInput signals
370
Експертні та інтелектуальні інформаційні системи, штучний інтелект
map is used for exploratory data analysis. It is
able to recognize clusters in data and also es-
tablish class proximity. In addition, the Ko-
honen card is able to predict client behavior. If
it is built a Kohonen map containing clusters
for each group of clients according to their de-
gree of loyalty, then with its help the expected
behavior of the client can be predicted and ap-
plied appropriate marketing policies to them.
The Kohonen map is also capable of detecting
anomalies. It distinguishes clusters in the train-
ing data and assigns all data to one cluster or
another. If after this the map encounters a data
set that is unlike any of the known samples,
then it will not be able to classify such a set and
thereby reveal its anomaly [11].
There are also other criteria for NN
classification:
˗ depending on the types of neurons:
homogeneous and hybrid;
˗ depending on the NN learning
method: supervised learning, unsupervised
learning, reinforcement learning;
˗ according to the type of input
information, NNs are: analogous (use
information in the form of real numbers),
binary (operate with information presented in
binary form); figurative (operate with
information presented in the form of images,
signs, hieroglyphs, symbols);
˗ according to the nature of synapse
setup: with fixed connections (NN weight
coefficients are selected immediately based on
the conditions of the problem, with dW/dt=0,
where W denotes NN weight coefficients);
with dynamic connections (when the learning
process is in progress in the settings of synaptic
connections, that is, dW/dt≠0, where W
denotes NN weight coefficients).
Related works
In recent years, there has been a need to
add new capabilities to OLAP. First of all, this
is due to the fact that when solving some
problems, issues arise due to the sharp increase
in the flow of data entering systems from
various sources. Therefore, there are big
problems in solving some tasks. On the other
hand, it should be noted that today the
requirement of the time is the solution of many
intellectual tasks. Intellectual tasks based on
big data are most effectively solved using NN.
Thus, it can be argued that the integration of
two technologies such as OLAP and NN is
very useful and important, as it enriches each
of them: OLAP is the ability to navigate the
multidimensional structure of indicators, and
NN is the ability to intelligently solve tasks on
large amounts of data .
If a large amount of data in intelligent
systems uses OLAP, NNs are effective for
solving clustering and classification tasks.
Note that clustering, unlike classification, does
not have predefined categories into which all
data should be grouped. In this case, the NN
itself generates clusters based on common
features of the data.
Clustering is one of the most important
methods of data analysis. The article [12]
provides a comprehensive overview of
clustering methods such as the self-organizing
Kohonen map, as well as clustering algorithms
such as k-means, fuzzy means algorithms, etc.
One of the classes of NNs primarily
used to solve clustering tasks is the Kohonen
neural network [13]. In this case, learning
occurs without a teacher, that is, only input
data sets are used, and no output values are
required.
k-means is the most popular and
simplest clustering method. Its main
disadvantages are: you need to know the
number of clusters in advance; very sensitive
to the choice of initial cluster centers.
In the decision-making process, you
can encounter many fuzzy tasks. Therefore, the
queries to the DW that the analyst is trying to
formulate may often contain uncertainties.
Clustering applied to the dimensions of an
OLAP cube using NN produces linguistic
variables. This will make it possible to solve
fuzzy tasks as well [14].
For data mining, the task of data
classification plays an important role.
Currently, a large number of different types of
classifiers have been developed, including
those built on machine learning. These include
NN. Although the classification task for NNs
is not the main one, their use has a number of
advantages:
˗ NNs are self-learning models, the
operation of which requires almost no user
intervention;
371
Експертні та інтелектуальні інформаційні системи, штучний інтелект
˗ NNs are universal approximators
that allow you to approximate any continuous
function with suitable accuracy;
˗ NNs are nonlinear models. It
allows to solve effectively classification tasks
even in the absence of linear separability of
classes (Fig. 4) [15].
˗
Linearly separable
classes
Linearly inseparable
classes
Fig. 4. Options for linear separability of
classes
By means of NN, forecasting problems
that are of great importance in the production,
economic and financial spheres are also
solved. Forecasting in OLAP is important be-
cause when looking at the contents of a cube,
it can often be sparse, meaning it is missing
some measures, and may also be missing di-
mensions. This happens due to missing infor-
mation or input errors. The absence of any
measures and measurements is undesirable and
can lead to incorrect analysis when making
strategic decisions.
In OLAP systems, NN can be used in
parallel with OLAP, i.e. OLAP cubes are cre-
ated on historical DW data, and NN forecast-
ing work is based on the same historical data
[16]. The disadvantage of this approach is that
there are no training data sets.
The approach proposed in [17] includes
two stages. First, principal components are an-
alyzed to reduce the dimensionality of the data
cube and special training sets are created.
Then, in the second stage, a new OLAP-ori-
ented multi-layer perceptron network (MLP)
architecture is proposed whereby training is
implemented on each training set and predicted
dimensions are generated.
In [18], the possibilities proposed in
[17] are expanded. First, the authors introduce
a generalized framework, i.e., Multi-perspec-
tives Cube Exploration Framework (MCEF),
for applying the classical data mining algo-
rithm to OLAP cubes. Secondly, the authors
refer to modular NNs that apply a neural ap-
proach to predicting multidimensional cubes
(NAP-NN). Modular NNs are a collection of
several different networks that operate inde-
pendently and contribute to the result. Each
NN has its own set of input data. These net-
works do not interact with each other during
task execution. The main advantage of modu-
lar NN is that the huge computational process
can be divided into several subprocesses. This
reduces computational complexity and in-
creases computational speed. But ultimately,
the processing time will depend on the number
of neurons and their participation in calculat-
ing the results. Note that NAP-NN includes a
preprocessing step. In this step, principal com-
ponent analysis (PCA) is performed to reduce
the size of the OLAP cube of the proposed
method. Modular neural networks work effec-
tively in cases where several directions of the
system are simultaneously processed.
Note that the article presents experi-
mental results showing the effectiveness of NN.
OLAP and neural networks
integration model
Big data includes huge volumes of het-
erogeneous and rapidly arriving digital infor-
mation that cannot be processed with tradi-
tional tools. Very effective analysis of big data
is carried out using machine learning methods,
in particular NN. It is very important that the
advantage of NN, such as the detection of hid-
den patterns that are invisible to humans, also
works well on big data. The integration of
OLAP and NN also provides these benefits.
Figure 5 illustrates the architectural and
technological model for integrating OLAP and
NN.
According to the Figure 5, data from
various sources, before entering the DW or
data mart (DM), first goes through ETL tech-
nology. During the ETL process, data is
cleared of duplication, contradictions and ty-
pos and brought into a common format. OLAP
cubes are built based on DW (or DM) data. The
figure shows the integration of NN with an
OLAP cube to classify and cluster OLAP cube
data and predict empty measures.
372
Експертні та інтелектуальні інформаційні системи, штучний інтелект
Fig. 5. Architectural and technological model of integration OLAP and NN
The integration of NN and OLAP is
that the NN environment is built into OLAP
applications that operate on a multidimen-
sional structure and a large volume of data cu-
bes.
Note that at this stage, the execution of
processes characteristic of traditional OLAP is
also ensured, namely: analytical queries are
implemented on data for their rapid viewing
and analysis, reports are issued based on the
OLAP cube data, which can be with either in-
termediate or final results. It is also possible to
view the same data from different angles.
Depending on the task set, the most ap-
propriate NN is selected from the above types.
For example,
- single-layer and multi-layer per-
ceptrons are used for classifica-
tion;
- single-layer and multi-layer per-
ceptrons are used for classifica-
tion;
- single-layer or multilayer percep-
trons and Kohonen map are used
for forecasting [11, 19].
Finally the results obtained will serve
to make management decisions.
Conclusion
Currently, in the era of Industry 4.0,
there is a dramatic increase in the flow of data.
requests,
reports,
Data Mining
methods
Data sourсes
……….
ETL
DW (DM)
OLAP-application
OLAP-cube1 OLAP-cuben
……
NN
Classification Clustering Forecasting
Data analysis
Decision making process
373
Експертні та інтелектуальні інформаційні системи, штучний інтелект
This creates great complications when solving
some problems related, for example, to classi-
fication, data clustering, forecasting, and even
in some cases makes solving these problems
impossible. AI can provide quick and effective
solutions to such problems. AI refers to any
system capable of performing creative func-
tions and solving problems that would typi-
cally require human intelligence. AI contrib-
utes to development and innovation in various
industries, such as healthcare, finance, manu-
facturing, transport, e-commerce, education,
etc. NNs are one of the areas of AI. NNs, being
implemented into systems, can solve important
tasks. Such systems include OLAP systems
based on OLAP technology. Recently, these
OLAP systems have been widely used both in
government agencies and in private ones.
OLAP enables analysts to explore and navigate
a multidimensional structure of metrics called
an OLAP cube. The purpose of the article is to
study the possibility of integrating OLAP and
NN, as well as to demonstrate the benefits of
such integration.
This article presented an architectural
and technological model, according to which
data analysis is performed using NN. The inte-
gration of NN and OLAP is achieved by em-
bedding the NN framework into OLAP appli-
cations that operate on a multidimensional
structure and a large volume of data cubes.
Further studies will develop methods
for embedding NN environment into OLAP
applications to integrate NN and OLAP.
References
1. Sh. Iskanderova, The impact of artificial intel-
ligence on the modern world, in: Science and
Education Scientific Journal (2023), Impact
Factor 3.848. Vol. 4, issue 4, pp. 564-570. [in
Russian]
2. L. Yunusova, A. Magsumova, Clustering using
neural networks and searching for dependen-
cies, 2019. Accessed: 22.01.2024.
https://cyberleninka.ru/article/n/klasteri-
zatsiya-s-pomoschyu-neyronnyh-setey-i-
poisk-zavisimostey-1. [in Russian].
3. M. Artikova, Sh. Rasulova, Data classification
using neural networks, in: Scientific Collection
«InterConf»: Scientific Goals and Purposes in
XXI Century, 2021, (78), pp. 403-409. Ac-
cessed: 15.02.2024. https://ojs.ukr-
logos.in.ua/index.php/interconf/arti-
cle/view/15124, [in Russian].
4. What is Online Analytical Processing (OLAP)?
Accessed: 18.12.2023.
https://aws.amazon.com/ru/what-is/olap/. [in
Russian].
5. R. M. Alguliyev, G. Ch. Nabibayova, A. M.
Gurbanova, Development of a Decision Sup-
port System with the use of OLAP-Technolo-
gies in the National Terminological Infor-
mation Environment, in: International Journal
of Modern Education and Computer Science
(IJMECS) (2019). Vol.11, no. 6, pp. 43-52.
6. E.F Codd, S.B. Codd, C.T. Salley, Providing
OLAP (on-line Analytical Processing) to User-
analysts: An IT Mandate, 1993. Accessed:
16.10.2024.
http://www.estgv.ipv.pt/paginaspes-
soais/jloureiro/esi_aid2007_2008/fichas/codd.
pdf.
7. Main characteristics of OLAP systems, 2020.
Accessed 18.12.2023.
https://hsbi.hse.ru/articles/osnovnye-kharak-
teristiki-olap-sistem/. [in Russian].
8. W.H. Inmon, Building the Data Warehouse,
John Wiley & Sons, 2005, p. 596.
9. W. McCulloch, W. Pitts, A logical calculus of
the ideas immanent in nervous activity, in: Bul-
letin of Mathematical Biology (1943), no 5, pp.
115–133.
10. F. Rosenblatt, The perceptron: a probabilistic
model for information storage and organization
in the brain, Psychological Review (1958),
65(6), pp. 386–408.
11. Kohonen map. Accessed: 18.11.2023.
https://basegroup.ru/deductor/function/algo-
rithm/kohonen [in Russian].
12. K.-L. Du, Clustering: A neural network ap-
proach, in: Neural Networks (2010), vol. 23, is-
sue 1, pp. 89-107.
13. Kohonen neural network, self-organizing
maps, learning, 2022. Accessed 20.03.2024.
https://microtechnics.ru/nejronnaya-set-ko-
honena-samoorganizuyushhiesya-karty-
obuchenie/ [in Russian].
14. Kumar K., Krishna R, Kumar De S. Fuzzy
OLAP Cube for Qualitative Analysis / Proceed-
ings of the 3rd International Conference on In-
telligent Sensing and Information Processing
(ICISIP), 2005, pp. 290-295, http://ieeex-
plore.ieee.org/xpls/abs_all.jsp?ar-
number=1529464&tag=1
15. V. Oreshko, Data classification using neural
networks, 2021. Accessed 15.03.2024
https://loginom.ru/blog/neural-classification
[in Russian].
16. M. Gulesian. Using Neural Networks and
OLAP Tools to Make Business Decisions,
2008. Accessed 11.02.2024. https://www.de-
veloper.com/database/using-neural-networks-
and-olap-tools-to-make-business-decisions/
17. W. Abdelbaki, R. Messaoud, S. Yahia, Neural-
Based Approach for Extending OLAP to Pre-
diction, in: Proceedings of science conference
374
Експертні та інтелектуальні інформаційні системи, штучний інтелект
Data Warehousing and Knowledge Discovery
(DaWaK 2012), Springer-Verlag Berlin Heidel-
berg, 2012, pp. 117–129.
18. W. Abdelbaki, S. Yahia, R. Messaoud, Modular
Neural Networks for Extending OLAP to Pre-
diction. Book Transactions on Large-Scale
Data- and Knowledge-Centered Systems XXI,
Springer Nature, 2015, pp. 73-93.
19. Neural networks, perceptron. Accessed:
10.12.2023. https://neerc.ifmo.ru/wiki/in-
dex.php?ti-
tle=Нейронные_сети,_перцептрон#cite_note
-3 [in Russian].
Received: 04.04.2024
Internal review received: 18.04.2024
External review received: 23.04.2024
About the author:
Nabibayova Gulnara Chingiz
PhD in technical sciences
http://orcid.org/0000-0001-8743-7579
Author's place of work:
Ministry of Science and Education
of the Republic of Azerbaijan
Institute of Information Technologies
Tel.: (+994 12) 539 01 67
E-mail: gnabibayova@gmail.com
www.ict.az
|