|
Submit Paper / Call for Papers
Journal receives papers in continuous flow and we will consider articles
from a wide range of Information Technology disciplines encompassing the most
basic research to the most innovative technologies. Please submit your papers
electronically to our submission system at http://jatit.org/submit_paper.php in
an MSWord, Pdf or compatible format so that they may be evaluated for
publication in the upcoming issue. This journal uses a blinded review process;
please remember to include all your personal identifiable information in the
manuscript before submitting it for review, we will edit the necessary
information at our side. Submissions to JATIT should be full research / review
papers (properly indicated below main title).
|
|
|
Journal of
Theoretical and Applied Information Technology
December 2023 | Vol. 101
No.24 |
Title: |
A LIGHT WEIGHT OF PARALLEL ENCRYPTION WITH DIGIT ARITHMETIC OF COVERTEXT
ENCRYPTION MODEL |
Author: |
WIDIYANTO TRI HANDOKO, EKA ARDHIANTO , HARI MURTI , RARA SRIARTATI REDJEKI |
Abstract: |
The speed of sending documents over the internet network is influenced by the
file size. Confidential documents require fast processing, reducing the risk of
hacking while on the network. So, secret documents must have a light ciphertext
size. The Parallel Encryption with Digit Arithmetic of Covertext (PDAC)
encryption model produces covertext with an average size of 124.88% larger than
plaintext. Thus, the process in the network will take longer, and give intruders
a lot of time to translate it. This research aims to design a proposed new PDAC
method to produce lighter ciphertext. This research was conducted using
experimental methods. Proposed PDAC model that has a different covertext
generator and encryption key generator design. The results obtained are a
plaintext and ciphertext ratio of 100%. This means that the size of the
ciphertext is the same as the size of the plaintext. Thus, the proposed new PDAC
method can reduce the size of the ciphertext, thereby speeding up the transfer
process over the network, saving storage space, and making it difficult for
intruders to retrieve. |
Keywords: |
Encryption, PDAC, Ciphertext, Cryptography. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
DESIGN OF A NOVEL PI FUNCTIONAL OBSERVER BASED LOAD FREQUENCY CONTROLLER FOR
INTERCONNECTED POWER SYSTEM |
Author: |
DUNDI JESRAJ TATAJI, ANAND GONDESI, K R SUDHA, A. CHANDRASEKHAR |
Abstract: |
The increase in population has led to the increase in the demand of the power in
terms the size and. To meet this demand a decentralized power structure has
emerged. The main advantage of decentralized power system is to overcome the
delay in decision making unlike the centralized power system. In the present
paper, load frequency control (LFC) in the decentralized scenario is analyzed
using state space. The stability of the power system can be observed through a
state-feedback controller rather than being directly monitored by the system
Instead of guessing the system states, a Functional Observer (FO) is made to
evaluate the control input. The stability can also be guaranteed by using the
suggested controller, as the observer gains can be calculated theoretically. An
industry-standard IEEE test system is used to evaluate the effectiveness of the
suggested methods. It has been found that functional observers perform more
effectively than both functional and traditional state observers. |
Keywords: |
Inter Connected Power System, Load Frequency Control (LFC), PI Functional
Observer, Leunberger Observer |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ANONYMOUS AND ONLINE HATE SPEECH WHEN GOVERNMENT RAISED SUBSIDIZED FUEL PRICES:
AN ANALYSIS ON INSTAGRAM ACCOUNT OF INDONESIAN STATED OWN COMPANY |
Author: |
RYNALDI SETIABUDI, LA MANI1, DAVID WILLIEM, WILIATER SITUMORANG |
Abstract: |
The study aims to investigate the impact of anonymity on the frequency of online
hate speech in the comments section of the instagram account @pertamina due to
government's decision to increase the prices of subsidized fuel (Pertalite and
Solar) in Indonesia. Hate speech can be delivered online, known as cyberhate,
and anonymity in online communication can lead to aggressive behavior and the
exploitation of anonymity for hate speech. This study employed a positivist
research paradigm with a quantitative approach to examine social phenomena and
their relationships. Data were collected by documenting the top 10 comments on
@pertamina instagram posts using a coding sheet for intercoder analysis on
August 31 2022 to September 7 2022. The variables of interest are anonymity and
online hate speech, operationalized through some indicators. Our study found a
significant relationship between anonymity and the frequency of online hate
speech, with anonymous users showing a 2.42 times higher prevalence of engaging
in such behavior compared to non-anonymous accounts. This underscores the role
of anonymity as a potential risk factor contributing to the initiation of hate
speech. Despite these findings, it is important to acknowledge that the research
has limitations, such as not considering other factors that may influence online
hate speech, such as educational level and sociodemographic factors.
Nonetheless, the study provides valuable insights into the impact of anonymity
on the prevalence of online hate speech on social media platforms like
instagram. |
Keywords: |
Anonymous, Online Hate Speech, Fuel Prices, Subsidized, Instagram, Pertamina |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
FUZZY SIMPLE ADDITIVE WEIGHTING FOR OPTIMIZATION TOURIST DESTINATION
RECOMMENDATIONS |
Author: |
ADIE WAHYUDI OKTAVIA GAMA, GEDE HUMASWARA PRATHAMA, DESAK MADE FEBRI PURNAMA
SARI, KADEK JANUARSA ADI SUDHARMA, DEWA AYU PUTU ADHIYA GARINI PUTRI, IDA BAGUS
PUNIARDHI ISABHA PIDADA |
Abstract: |
The difficulty of choosing destinations according to tourist preferences
requires the development of an appropriate Decision Support System to optimize
tourist destination recommendations. This research aims to develop a Decision
Support System using the Fuzzy SAW algorithm to support the growth of the
tourism sector. This research develops a decision support system. The objects
used are 13 tourist destinations in Karangasem Regency, Bali. The algorithm used
is Fuzzy Simple Additive Weighting, where the algorithm considers costs and
benefits. Determine the criteria, then determine alternative costs and benefits
and give fuzzy weight to the criteria for use. The result of this study is the
ranking of tourist destination recommendations according to Fuzzy SAW
calculations. This research contributes to implementing the appropriate SAW
algorithm in the tourism sector. In addition, the results of this study make it
easier for tourists to choose tourist destinations that suit their preferences
so that they can support the growth of the tourism sector. |
Keywords: |
Fuzzy Simple Additive Weighting, Decision Support System, Tourist Destination,
Tourism Growth, Karangasem Regency |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
CLT-CRP : A CROSS LAYER TECHNIQUE FOR CLUSTER BASED ROUTING PROTOCOL IN MOBILE
WIRELESS SENSOR NETWORK |
Author: |
D. SWAPNA, M. NAGARATNA |
Abstract: |
The Wireless Sensor Network(WSN)s attains tremendous changes in the field of
communication networks. Due to the nature of ubiquitous capability of WSN, used
in different applications such as monitoring of environment, surveillance in
military and health care applications. The innovations in WSN are widely used in
data transmission applications. Basically the WSN consists of sensor nodes and
clusters. The sensor nodes can move from one cluster to another cluster
uniformly and randomly. The phenomenon leads to high energy consumption and data
loss in the network. The field of WSN attracts many researchers to work with
different routing protocols. However, the previous routing protocols were unable
to reduce energy consumption and data loss. Many issues in data transmission
with the sensor nodes and nature of resource constrained. To address the
limitations of WSN, in this research paper proposed a Cross Layer Technique for
Cluster Routing Protocol (CLT-CRP) for Mobile Wireless Sensor Network(MWSN). The
proposed protocol is designed with a combination of MAC layer and Network layer
configurations. The implementation of the proposed CLT-CRP is made using the NS2
simulator. The proposed protocol outperforms with the comparison of LEACH, CRPD
protocols in terms of residual energy, delay, throughput and packet delivery
ratio. The results revealed that the proposed protocol improves the performance
of WSN. |
Keywords: |
Wireless Sensor Network, Cluster Head, Energy Consumption, Data Transmission |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
PREDICTION OF LUNG DISEASE SEVERITY BY APPLYING MACHINE AND DEEP LEARNING
TECHNIQUES |
Author: |
SURYA PRAKASH GUTTULA, SANJEEV KUMAR GUPTA, LAXMI SINGH, K VIDYA SAGAR |
Abstract: |
Virus infected diseases are increasing rapidly. SARS covid -19 is one emerged
into human body to extinct the human life. Prediction of the rapid changes and
meticulous interpretation of the type of decease is challenging. Various stages
of the risk severity prediction and interpretation is challenging. This paper
discussed various machine learning algorithms applied on X-radiation chest
images to predict the severity of the decease. Significant features are
extracted using Principal component analysis (PCA). Bagging, Ada boost, XG
boost, KNN machine learning methodologies applied to achieve the reliable
performance. Bagging methodology shows dominant performance over other machine
learning methodologies with 98.82% precision value and 98.67% accuracy. The F1
score and recall value is significantly good with 98.755 and 98.69 respectively.
Bagging methodology is much reliable to interpret X-radiation images for Sars
Covid-19 infections. |
Keywords: |
SARs Covid-19, X-radiation images, Bagging, Ada boost, KNN, XG boost, PCA. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
METHODS OF MODELING, CALCULATION AND ANALYSIS OF ROBOTIC SYSTEMS FROM COMPOSITE
MATERIALS UNDER DYNAMIC INFLUENCES |
Author: |
KAMIL KHAYRNASOV |
Abstract: |
The methods of modeling, calculation and analysis of a robotic system are
considered: a semi-natural simulation stand designed to imitate flight
characteristics in ground conditions. The originality of this study is in the
uniqueness of the design of the stands and the modeling of the stand from a
composite material. A method has been developed for approximating bearings,
gearboxes and gear rims by rod systems identical to them in terms of rigidity. A
method for creating a stand of maximum rigidity by locating the base of
composite material along the lines of maximum stresses has been developed. A
study of composite of materials and magnesium alloy, traditionally used in the
manufacture of stands for dynamic operational loads, was made. A method for
creating a three-layer stand structure that provides maximum strength and
rigidity has been developed. The problem by the finite element method is solved.
The convergence of the results of the study was determined by thickening the
mesh of finite elements and comparing the results obtained. The stress-strain
state of stands made of composite material and magnesium alloy has been
obtained. Comparison of the results shows that the stand made of composite
material has advantages over the stand made of magnesium alloy. A layer-by-layer
stress state of a five-layer composite material has been obtained. The results
of the failure of the composite were determined based on the proven criteria for
the failure of layered materials. The considered research methods are applicable
to a wide class of robotic systems under dynamic influences, containing
bearings, gear rims, gearboxes and motors. The study revealed the structure of
the arrangement of the base layers of the composite material, in which the
construction has the greatest strength and rigidity. The production of dynamic
stands from a composite material exceeds the strength characteristics of the
magnesium alloy traditionally used in the manufacture of stands. |
Keywords: |
Robotic Systems; Stands; Composite Materials; Calculation; Analysis; Finite
Element Method; Dynamic Effects |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
TRANSFORMER-BASED MODEL FOR HANDWRITTEN RECOGNITION ARABIC WORDS AL-SOUDANI
MAGHREBI SCRIPT |
Author: |
SIDI AHMED MAOULOUD, M. H OULD MOHAMED DYLA, CHEIKH BA |
Abstract: |
Automatic handwriting recognition is a crucial element for various applications
across different domains. It’s a complex problem that has garnered significant
attention over the past three decades. For the Arabic language, research has
primarily focused on recognizing historical Eastern Arabic scripts and
manuscripts. However, fewer studies have been conducted on Maghrebi Arabic
scripts and their variants. In this article, we introduce a novel dataset of
Arabic words written in the widely used Maghrebi Al-Soudani script, primarily
found in West Africa, extracted from three different manuscripts. Our dataset
comprises 30,430 words. We also propose a Handwritten Text Recognition (HTR)
model based on an Encoder-Decoder architecture, utilizing a Convolutional Neural
Network (CNN) for image encoding and a Transformer for image-to-text decoding.
We train our model, as well as a recent reference model (FPHR), on this dataset.
The results demonstrate promising performance of our model, outperforming the
FPHR model with accuracy rates of 10% for Character Error Rate (CER) and 9.9%
for Word Error Rate (WER). |
Keywords: |
Dataset Transformer Encoder-Decoder Handwritten recognition Al-Soudani Arabic
script Manuscript Image-to-Sequence |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
USING DOMAIN ONTOLOGY UNCERTAINTY OR ONTOLOGY MAPPING UNCERTAINTY TO IMPROVE
LEARNING |
Author: |
TATYANA IVANOVA |
Abstract: |
A grand amount of research, related to the application of ontologies in
education has been done in recent years. Hot topics are personalization of
tutoring and learning and knowledge modeling. Most of ontologies are mainly
implemented in software systems (i.e. personalized e-learning environments) and
are rarely used directly by learners or teachers. This is because of the needs
of specific knowledge and skills. We think that in our days some teachers and
learners mainly in higher education have all the knowledge and skills for using
ontologically represented knowledge directly in learning and tutoring. Knowledge
uncertainty exists in all real domains, but it is rarely discussed in e-learning
courses. Most of semantic technologies are based on crisp logics and cannot deal
with uncertain knowledge. In this research we analyze the specifics and sources
of uncertain knowledge and technologies for its semantic modeling. We also
discuss strategies for direct use of uncertainty both in domain knowledge and
ontological models to improve learning. In such a way we search possibilities
for direct usage of methods of one of the most important artificial intelligence
fields, as knowledge modeling for improving learning and tutoring. We
propose a model of interactive ontology and ontology alignment evaluation
environment, aimed at involving users in solving uncertainty problems during
learning. We discuss strategies for application of domain and ontological models
uncertainty for e-learning purposes and situations of its practical usage. |
Keywords: |
Ontology, Uncertainty, E-Learning, Ontology Evaluation, Ontology Mapping |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
MODELING, CALCULATION AND ANALYSIS OF ROBOTIC STRUCTURE MADE OF COMPOSITE
MATERIAL UNDER DYNAMIC IMPACT |
Author: |
KAMIL KHAYRNASOV |
Abstract: |
The paper develops methods for calculating and analyzing robotic systems:
multi-step dynamic systems under dynamic loading. At present, stands are made of
magnesium alloys, which have good specific strength characteristics, and
composite materials with higher specific strength characteristics are not used
in the manufacture of stands. Therefore, the issues of researching stands made
of composite material are important and relevant. The paper gives dependencies
for determining the given characteristics of a multilayer composite material.
Modeling and approximation of the stand by finite elements is given. The stand
model had a three-layer structure: the model received the characteristics of the
filler, and the surface of the model had the characteristics of a five-layer
composite material. A technique for approximating the bench elements: bearings,
gear rims and gearboxes by a system of rod elements of the same stiffness is
proposed. An algorithm and a program for determining the stiffness of these
elements have been developed. A technique has been developed for arranging a
base made of a composite material along the lines of maximum stresses to ensure
maximum rigidity of the stand. A comparative analysis of a stress-strain stand
made of a composite and magnesium material has been carried out. Research
methods can be applied to a wide class of robotic systems. |
Keywords: |
Robotic Systems; Benches; Modelling; Finite Element Method; Composite Materials;
Stress-Strain State; Dynamic Loading |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
MANTA RAY FORAGING OPTIMIZATION ALGORITHM WITH DEEP LEARNING ASSISTED AUTOMATED
PHISHING URL DETECTION MODEL |
Author: |
K SUBASHINI, DR. V NARMATHA |
Abstract: |
In current scenario, phishing attacks are vital threats to cyberspace security.
Phishing is one of the common types of scams that attract individuals to access
mischievous URLs (Uniform Resource Locators) as well as their personal data like
IDs, passwords, and others. Many intelligent attacks have been launched to cheat
users by retrieving a trustworthy website or any online platform in order to get
data. Phishing URL classification is one of the crucial cybersecurity tasks
intended to classify and moderate malevolent web addresses considered to cheat
consumers by revealing sensitive data. Numerous researchers in cyberspace are
interested in generating intelligent techniques as well as offering security
services on a phishing website that grows more clever and malicious daily.
Therefore, this study introduces a manta ray foraging optimization with deep
learning-based phishing website detection (MRFODL-PWD) technique. The major
intention of the MRFODL-PWD technique is to recognize and classify the presence
of legitimate or phishing URLs. In the presented MRFODL-PWD technique, several
stages of pre-processing to transfer data into a useful setup, and BERT is
applied for feature extraction. Moreover, deep belief network (DBN) model can be
used for automated phishing URL detection. Furthermore, the MRFO algorithm
selects the hyperparameter values of the DBN model. An extensive comparison
study stated that the MRFODL-PWD technique accomplishes enhanced phishing URL
detection results over other models. |
Keywords: |
Phishing Attacks; Cybersecurity; Deep Learning; Parameter Tuning; Manta Ray
Foraging Optimization |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
A NEW SMART COMMUNICATION PROTOCOL AND INTERNET OF THINGS (IOT) FOR WASTE
MANAGEMENT SYSTEM |
Author: |
Dr.B. RAGHURAM, Dr.C.SRINIVAS, Dr.S.VENKATRAMULU, Dr.V.CHANDRASHEKAR RAO4,
Dr.K.VINAYKUMAR, UMA N DULHARE |
Abstract: |
Rapid population growth throughout the world has resulted in the improper
handling of garbage in many nations, leading to several health problems and
environmental contamination. Every week, the garbage vehicles collect garbage
only once or twice. Due to improper wastage collection procedures, the contents
of the trash can are scattered throughout the streets. This paper proposes a
solution for wise and effective waste management based on deep learning (DL) and
the Internet of Things (IoT) to combat this problem. The proposed method
combines the ResNet-50 and Inception network models, which receive input and
provide the correct solution for identifying and classifying Waste and
organizing it into the appropriate waste receptacle (recyclable, organic, and
hazardous wastes) without human intervention. An ultrasonic sensor is inserted
in each wastage container to measure the amount of garbage filling. The
integrated GPS module monitors the bin's location in real-time. Utilizing the
LoRa communication protocol, the receptacle's location, real-time, and filling
level are transmitted. RFID module is embedded for personnel identification in
waste management. Recycled and organic residues that have been categorized can
be used for greater purposes in the future. This procedure will assist the
environment is becoming more valuable and ecologically secure, as well as assist
us in creating a thriving verdant ecosystem and a brighter future. This method
will help us establish a lush, verdant ecology and a brighter, hopeful future
while making the environment more valued and environmentally secure. The
proposed Res50+IncV3 network obtains 0.98 accuracy,0.97 precision, and 23% loss. |
Keywords: |
Waste management, Internet of Things, neural networks, ultrasonic sensor,
environmental pollution |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
NABGAMES: NASH BARGAINING GAME FOR IMPROVING COVERAGE IN UNMANNED AERIAL
VEHICLES (UAV) |
Author: |
UMA N DULHARE, Dr.B.RAGHURAM, Dr.C.SRINIVAS, S. VENKATRAMULU, Dr.V.CHANDRASHEKAR
RAO, Dr.K.VINAYKUMAR |
Abstract: |
The deployment and positioning of unmanned aerial vehicles (UAVs) are heavily
reliant on the amount of power accessible, a workable situation, and a protected
connectivity. Appropriate height is required for the practical deployment of UAV
modules to reach the mobile networks. Resources wastage or operation
inefficiencies result from the under or overestimation of nodes' air time. The
UAV coverage issue is taken into account in this research, and then a Nash
bargaining game model (NaBGames) is suggested to enhance the range by utilizing
UAV nodes which are controlled and managed for re - allocation and can shift
positions as needed. Cellular users are permitted to flexibly alter their
service cost during every round of negotiation in order to maximize their
earnings. Afterwards, every user adopts "bargaining" and "warning" to create
their finest reaction. Such method could function even in hostile circumstances
because it is dependent on the connections between UAVs and its nearby neighbors
in a distributed system. By employing the least number of UAVs required to
encompass the potential specific region, the altitudes of the UAVs are modeled
based on the transmitters and other installation needs. The proposed NaBGames is
compared with three base line methods, whereas it achieves 18.3 ms delay,
587Mbps throughput, 16.4% of power consumption and 97.8% of Trade success
probability. |
Keywords: |
Game theory, Nash bargain, Unmanned Aerial Vehicles, Coverage, Power
consumption. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
GRØSTL STOCHASTIC GRADIENT DEEP MULTILAYER PERCEPTIVE BLOCKCHAIN BASED SIDE
CHANNEL ATTACK DETECTION FOR ENHANCED SECURITY IN CLOUD COMPUTING |
Author: |
RAMAKRISHNA SUBBAREDDY, Dr.P. TAMIL SELVAN |
Abstract: |
Data security has a vital role in Cloud Computing. Cache side channel attacks
are a type of cryptanalysis in the cloud for acutely threatens the security of
the cryptosystem. Therefore, a novel cryptography model named Grøstl stochastic
gradient deep multilayer perceptive Blockchain (GSGDMPB) model is designed for
detecting side channel attacks in the cloud computing environment. At first, the
Grøstl cryptography function generates hash value for every user data. Then
Borda positional voting consensus algorithm is also applied for identifying the
active blocks based on the majority votes. Secondly, Lai-Massey stochastic
gradient deep multilayer perceptive learning is employed to perform encryption
and decryption. After that, the generated cloud user block validation is
performed based on the simplex matching coefficient. Then the max-out activation
function is employed to provide final attack classification outcomes through
enhanced accuracy. Experimental assessment of proposed model is performed by
dissimilar metrics by a different number of traces. The results of GSGDMPB model
improves data communication security through enhanced channel attack detection
accuracy, throughput, and minimum overhead and time than the conventional
methods. |
Keywords: |
Cloud Computing, Security, Side Channel Attacks, Grøstl Cryptography, Borda
Positional Voting Consensus Algorithm, Lai-Massey Stochastic Gradient Deep
Multilayer Perceptive Learning, Simplex Matching Coefficient, Maxout Activation
Function. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
A SURVEY ON COMPONENTS THAT GOVERN THE EXECUTION OF CRYPTOSYSTEMS |
Author: |
YAHAYA GARBA SHAWAI, MOHAMAD AFENDEE MOHAMED, USMAN HARUNA, MOHAMMED AMIN
ALMAIAH ,ABDALWALI LUFTI, SULAIMAN IBRAHIM MUHAMMAD |
Abstract: |
The aim of the review was to carry out a survey on the components that govern
the experimental simulation of a cryptosystems, which includes input images,
hardware and software facilities, algorithm formulations. The components were
analysed with regards to the performance of the encryption algorithms, based on
specific finding of individual execution time. The review paper discovered the
existing literature within the fields of image encryption based chaotic system,
with aid of adapting new proposed systematic review framework known on as YAFSU.
The search strategy process based on search and selection were considered as to
be the pre-requisite of the framework, with regards to the extracted data and
synthesis implementation. It is believed that the review may put impact on the
researcher’s with interest on chaotic cryptosystems based on images, to discover
the input images, hardware and software facilities that should be considered for
the experimental simulations. In addition, performances of existing encryption
algorithms were analysed, to highlights ways to be followed in developing robust
image encryption algorithm that may outperform the existing ones. The
limitations of the reviewed literature were discovered, that some image
encryption algorithm simulations suffered from fewer input images utilizations,
which in many cases resolved to poor performance or in-efficiency of the
encryption algorithm. Moreover, it was observed that no research exist to prove
the claim “The effectiveness of the execution time depends on the Hardware
capacity of the computer systems”. Considering reviewed literatures, it was
discovered that both Saleh et al and Mohamed et al. Used high capacity computers
than that of Nadeem et al. But, various execution time obtained by the
researcher’s disproving the earlier claim.. It was recommended that more finding
should be discovered based on utilization of many input images during the
experimental simulation, which will assist in monitoring both the security and
performance of the proposed encryption schemes. Both the gray/coloured images
need a discovery that may clear the claim which said “The effectiveness of the
execution time does not depend on the Hardware capacity of the computer
systems”.. in addition, it was observe that recent articles used older version
of software tool being utilized during the experimental simulation. |
Keywords: |
Encryption Scheme, Experimental simulations, Images, Inputs, Hardware and
Software, Cryptography |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ANALYSIS OF TECHNICAL INDICATORS OF EFFICIENCY AND QUALITY OF INTELLIGENT
SYSTEMS |
Author: |
KRYVORUCHKO O., KOSTIUK Y., DESIATKO A., STEPASHKINA K., TYSHCHENKO D., FRANCHUK
T., HNATCHENKO D., ZAKHAROV R., BRZHANOV R. |
Abstract: |
The efficiency of intelligent systems refers to their ability to achieve set
goals and perform tasks with high quality and efficiency. It is determined by
the effectiveness of the system in the context of its functionality,
productivity, accuracy, and speed of solving tasks. The efficiency of
intelligent systems can be evaluated according to several criteria. For example,
efficiency can be related to the time it takes to complete tasks, the amount of
resources used, the quality of problem-solving, or the system's ability to adapt
to changing conditions. Various indicators and metrics can be used to measure
the performance of intelligent systems, such as response time, classification
accuracy, data processing speed, use of resources (for example, memory or
computing power), etc. The performance of intelligent systems is an important
aspect of their development and application, as it determines their ability to
achieve positive results and perform tasks with maximum productivity and
quality. |
Keywords: |
Intelligent Systems, Multi-Agent Systems, Adaptive Systems, Efficiency
Indicators |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
PROPOSED ENHANCED FEATURE EXTRACTION FOR MULTI-FOOD DETECTION METHOD |
Author: |
SUHAILA ABUOWAIDA, ESRAA ELSOUD, ADAI AL-MOMANI, MOHAMMAD ARABIAT, HAMZA ABU
OWIDA, NAWAF ALSHDAIFAT, HUAH YONG CHAN |
Abstract: |
This research presents a comprehensive system that utilizes computer vision and
deep learning techniques to develop the detection of multiple food methods.
Despite the incorporation of deep learning techniques, the effectiveness of the
existing detection method for different food products is unsatisfactory due to
the utilization of ResNet-101 for feature extraction. The features maps of the
ResNet-101 exhibit a size reduction or may vanish entirely following the
down-sampling process. The ResNet-101 blocks may have been subject to varying
degrees of repetition, with certain blocks receiving a restricted number of
repeats and others being excessively repeated. There is an ongoing need to
develop the rate of detection in the field of food recognition. The procedure
under consideration consists of a series of primary steps. The optimization of
the ResNet-101 block entails the careful selection of an appropriate number of
repetitions. A supplementary convolutional layer is suggested. The results
produced from this approach were later compared to the outcomes reached by the
latest algorithms in object detection, specifically Mask R-CNN and CASCADE
R-CNN. The evaluated algorithm exhibits exceptional performance in accuracy AP
over multiple thresholds. The numbers regularly exceed the relevant criteria of
three commonly utilized techniques. The thresholds demonstrate higher magnitudes
than the thresholds of three frequently utilized methods. |
Keywords: |
Deep Learning, Object Detection, Resnet-101, Features Maps, Down-Sampling
Process. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
OLGV3 NET: OPTIMIZED LIGHTGBM WITH INCEPTIONV3 FOR ACCURATE MULTI-CLASS BREAST
CANCER IMAGE CLASSIFICATION |
Author: |
D. VETRITHANGAM, B. ARUNADEVI, ANGATI KALYAN KUMAR, S. NALINI, NEHA , NIRMAL
MAHESH |
Abstract: |
In todays world, computer models, especially those using deep learning, are
helpful in diagnosing breast cancer by analyzing special images called
histopathological images. Understanding and classifying these images for breast
cancer diagnosis is crucial in the field of medical information technology. The
existing deep learning models for breast cancer image classification include a
lack of diversity in the training dataset, leading to reduced model robustness
and an inability to accommodate variations in different imaging conditions.
Furthermore, there exists a deficiency in the model's sensitivity and
generalization capabilities, accompanied by suboptimal hyperparameter
configurations. This inadequacy has the potential to hinder the model's
efficiency in breast cancer classification. Additionally, the absence of
regularization options heightens the susceptibility to overfitting. These
identified gaps directly impact the effectiveness of current technologies in
addressing crucial issues encountered in clinical practice and biomedical
research concerning breast cancer diagnosis and prognosis. This research aims to
overcome these challenges by focusing on important factors like making the model
work well with different types of images, avoiding unnecessary information,
ensuring efficient performance, and handling difficulties when there are only a
few cancer cells present. The proposed solution is a new model called OLGV3 Net
Classifier, which combines enhanced Inception V3 for understanding images and
LightGBM for making accurate classifications. By using Sequential Model-Based
Optimization (SMBO) to fine-tune the model's settings, this research achieved a
remarkable accuracy of 99.80%, surpassing other models and making a significant
improvement in breast cancer image classification. |
Keywords: |
Multi-classification, Breast cancer, Inception V3, LightGBM, Optimization |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
INDUSTRY 5.0: AN OVERALL ASSESSMENT OF USING ARTIFICIAL INTELLIGENCE IN
INDUSTRIES |
Author: |
T S KARTHIK, BEATRIZ LUCIA SALVADOR BIZOTTO, MITHILEYSH SATHIYANARAYANAN |
Abstract: |
In recent years, the increasing complexity of digital technologies and the
emergence of Artificial Intelligence (AI)-based solutions have posed challenges
to maintaining a competitive edge. Industry 4.0, known as the fourth industrial
revolution, entails an increased degree of automation with the goal of improving
operational productivity and efficiency in both the digital and physical domains
of an organization. The importance of AI computing is seeing significant growth
in tandem with the advancements made during the Fourth Industrial Revolution.
Notwithstanding the persistent challenges, several enterprises continue to face
obstacles in their endeavor to digitalize their operations via the utilization
of technologies such as the Internet of Things (IoT), AI, and other analogous
improvements. To address this challenge, the introduction of the fifth
industrial revolution, sometimes referred to as "Industry 5.0," aimed to address
individualized production processes and enhance people across different
industries. The advent of Industry 5.0 signifies the emergence of a new era
characterized by personalized approaches, which enable digital production
systems to achieve improved outcomes. In recent times, there has been a notable
surge in the prominence of AI-based wireless applications, owing to the rapid
advancements in remote devices and the emergence of AI-driven technologies.
Creative AI-based advancements are emerging in industrial applications to
address demands that cannot be fulfilled by traditional remote handling and
communication technologies. These demands include high throughput, high
mobility, low latency, heterogeneity, and flexibility. The primary aim of this
research paper is to explore the effect of AI in the Industries. In order to
accomplish this, initially, it provide a thorough examination of AI in the
Industries, including its technological, economic, and regulatory dimensions.
This analysis will include all facets of the technical assessment and impact
assessment of AI in industries. |
Keywords: |
Industry 5.0, Artificial Intelligence, Technological Assessment, Impact
Assessment |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
DETECTION OF EPILEPTIC SEIZURES USING EEG SIGNALS |
Author: |
DR. SHARMISHTA DESAI, PUJA A. CHAVAN |
Abstract: |
According to the World Health Organization (WHO), epilepsy is a chronic
neurological circumstance illustrated by an excessive and uncontrolled
electrical explosion. It is a rare occurrence that affects people of various
ages. An electroencephalogram (EEG) of brain activity is a well-known tool for
studying epileptic convulsions and recording changes in electrical activity in
the brain. Consequently, epilepsy prediction and early diagnosis are required to
give timely preventative measures to free patients from the detrimental
repercussions of epileptic seizures. Despite decades of research, accurately
projecting these seizures remains an unsolved challenge. This paper proposed
epilepsy seizure detection and classification using a deep learning model called
CNN and LSTM. In the convolutional layer, numerous features are extracted from
EEG signal files, while in the optimization layer, non-essential elements are
eliminated. In an extensive experimental analysis, we validated the system with
a real-time EEG dataset, where we obtained 82.5% accuracy for epilepsy detection
for the entire testing dataset. |
Keywords: |
EEG Signals, Epilepsy Seizure Detection, Brain Computer Interaction, Feature
Extraction, Classification, Deep Learning |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
EXPLORING PRIORITY OF UCD IN AN AGILE DEVELOPMENT ENVIRONMENT: A CASE STUDY IN
IRAQ |
Author: |
FOUAD ABDULAMEER SALMAN, BAKHTAWAR BALUCH, ZURIANA ABU BAKAR |
Abstract: |
User-Centred Design (UCD) and agile software development share common values,
such as customer satisfaction, continuous improvement, and flexibility. However,
integrating UCD into agile software development poses several challenges that
must be addressed to achieve successful outcomes. This paper investigates the
existing practices of UCD, which are carried out in an actual development
environment. It also explores the obstacles that software development team
members may face while applying UCD practices alongside agile development
activities. The qualitative and quantitative approaches are used to collect the
primary data from the practitioners in different job positions. Two methods are
used to gather primary data: semi-structured interviews and closed-ended
surveys. The results reveal a growing realization of the usability concept in
software development among Iraqi agile practitioners. Further, the results
provide insight into how well they can incorporate UCD activities within agile
development circumstances. |
Keywords: |
User-Centred Design (Ucd), Usability, Usability Engineering, Agile Development
Process |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
EXPLORING THE CAPABILITIES OF DEEP LEARNING FOR ADVANCING CREDIT CARD FRAUD
DETECTION: A REVOLUTIONARY APPROACH |
Author: |
ABDERRAHMANE DAIF, SOUMAYA OUNACER, SOUFIANE ARDCHIR, MOHAMED GHAZOUANI, MOHAMED
AZZOUAZI |
Abstract: |
The use of credit cards for both online and in-person purchases has become
increasingly prevalent in our daily lives. However, this convenience also
exposes users to the risks associated with credit card fraud. Credit card fraud
presents a significant challenge for banks, merchants, and consumers,
emphasizing the crucial need for the swift and accurate detection of such
fraudulent activities. In response to this challenge, recent research has delved
into the application of deep learning techniques for credit card fraud
detection. This article presents a study that combines a Bi-LSTM (Bidirectional
Long Short-Term Memory) with an attention layer to identify fraudulent
transaction patterns and achieve a balanced classification of data. The results
of this study demonstrate the method's high accuracy, surpassing the performance
of other fraud detection approaches. Notably, this innovative approach
efficiently identifies critical transactions within input sequences,
significantly improving the prediction accuracy for fraudulent transactions.
This research provides a unique perspective on the use of deep learning to
enhance security in credit card transactions. |
Keywords: |
Deep Learning, Machine Learning, Credit Card Fraud, Bi-LSTM, Attention
Layer, SMOTE. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
OPTIMIZED FUZZY C-MEANS (FCM) CLUSTERING FOR HIGH-PRECISION BRAIN IMAGE
SEGMENTATION AND DIAGNOSIS USING DENSENET FEATURES |
Author: |
PRIYA KANTAMANENI, D.VETRITHANGAM, M MAITHILI SAISREE3, S.SHARGUNAM, S.SRINIVAS
KUMAR, ASHOK BEKKANTI |
Abstract: |
Brain imaging techniques play a crucial role in identifying the causes of brain
cell injury. Consequently, early diagnosis of such conditions can yield
significant benefits, improving treatment prospects and minimizing potential
patient complications. Among the most formidable challenges in medical image
analysis is brain tumor segmentation. Challenges include limited spatial
context, increased occurrences of false positives and negatives, the inability
to distinguish tumor components, and a lack of preprocessing. To address these
issues, we propose an approach that combines Optimized Fuzzy C-Means (FCM)
Clustering with DenseNet Features and employs efficient preprocessing
techniques. Our improved DenseNet architecture meticulously extracts relevant
features from preprocessed images. FCM assigns each feature vector to one or
more clusters based on their degrees of membership and its output encompasses
cluster centers and membership values, indicating the degree of association for
each data point with each cluster. Hence, FCM improves interpretability by
distinctly delineating tissue regions through the utilization of these features.
Markov Random Field (MRF) Optimization is probabilistic graphical model that
capture spatial dependencies among neighboring pixels or regions in an image. As
each MRI modality possesses the unique ability to emphasize distinct tissue
characteristics. All the MRI Modalities (Flair, T1, T1c, T2) can be combined to
get valuable and complementary wealth of information regarding the tissues and
structures undergoing examination. Our optimized FCM model is experimented on
the Original FLAIR -MR images of patients and Combined MRI Modalities (Flair,
T1, T1c, T2).The Optimized Fuzzy C-Means (FCM) Clustering achieved train Dice
Coefficient score of 99.18% and test Dice Coefficient score 98.64%, and the
Optimized Fuzzy C-Means (FCM) Clustering with Combined MRI modality feature
achieved train Dice Coefficient score 100% and test Dice Coefficient score of
99.413%.The results shows that the proposed model out performs the existing
models. |
Keywords: |
Brain Image, Segmentation, DenseNet, Fuzzy C-Means, Optimization, Diagnosis |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
BILAE-GAN FRAMEWORK FOR ANOMALY DETECTION IN VIDEO SURVEILLANCE |
Author: |
SWAPNA.C, DR.B.PADMAJA RANI |
Abstract: |
In recent years, increasing the use of surveillance cameras with less manpower
makes automatic video surveillance systems to become more important. Recent
advances in video anomaly identification have mostly focused on improving
performance with available datasets. We propose a Bidirectional Long-Short term
memory-based Convolutional Autoencoder Generative Adversarial Network
(BiLAE-GAN) method for video surveillance. During training the model learns the
normal data distribution of data in the Generator and the detection of anomalies
in the discriminator. Bidirectional Long-Short term network in Convolutional
Autoencoder in Generator for reconstruction, Encoder features of Generated image
and real image to discriminator to identify Anomaly. At the anomaly detection
phase, anomalies are identified based on reconstruction error and discrimination
results. Our proposed method validation benchmark datasets such as UCSD Ped1,
UCSD Ped2, and CHUCK Avenue dataset with performance metrics AUC, EER. |
Keywords: |
BiLAE-GAN, Encoder-Decoder, BILSTM, Svdloss, Advloss |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
MAHALANOBIS-TAGUCHI SYSTEM AND TIME-DRIVEN ACTIVITY-BASED COSTING INTEGRATION OF
PRINTED CIRCUIT BOARD |
Author: |
NURUL HAZIYANI ARIS, MOHD YAZID ABU, SRI NUR AREENA MOHD ZAINI, MUHAMMAD
ARIEFFUDIN MOHD JAMIL, NUR SYAFIKAH PINUEH, WAN ZUKI AZMAN WAN MUHAMMAD, FAIZIR
RAMLIE, NOLIA HARUDIN, EMELIA SARI |
Abstract: |
System integration involves creating or implementing a customized architecture,
as well as integrating hardware, packaged and customized software, and
communications. The electronic component demand is rising nowadays. However, a
weekly rejection report for rejected parts is necessary in reflect the severity
level as it can influence the criteria used to identify relevant parameters.
Meanwhile, activity-based costing (ABC) is outdated since it employs interviews
or surveys to get numerical data, which can lead to erroneous corporate data.
This research aim is to integrate Mahalanobis-Taguchi system (MTS) with
time-driven activity-based costing (TDABC) at the electronic component
workstation. MTS optimizes 22 printed circuit board (PCB) assembly top side
workstation environments. After generating time equation and capacity cost rate,
TDABC considers critical elements to identify underutilized capacity. As a
result, October and December have the most significant parameters (16) and
August the fewest (10). Next, the TDABC confirmed that the PCB input into auto
loader machine sub-activity has a large unused capacity of time (239,550
minutes) and cost (MYR18,120.67), which allows this workstation to minimize
unused capacity by shifting employees to another load workstation. In spite of
to minimize capacity utilization in sub-activity solder paste preparation, extra
labor should be given since it has an overutilized capacity of time (-790,100
minutes) and cost (MYR-1,000,535.13). Lastly, the reduction of waste and better
forecasting is obtained after MTS and TDABC integration is implemented. |
Keywords: |
MTS, TDABC, Integration, Classification, Optimization. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
SKIN LESION SEGMENTATION AND CLASSIFICATION USING FCN-ALEXNET FRAMEWORK |
Author: |
V. AUXILIA OSVIN NANCY, V.RAJASEKAR, MEENAKSHI S ARYA |
Abstract: |
Malignant melanoma tumors diagnosis relies as essential tasks for segmentation
and classification of skin lesion. However, previous deep learning methods still
face challenges in effectively handling aspects like boundary identification,
artifact presence, and limited dataset availability. The automatic application
poses significant difficulties, especially considering the intricate nature of
melanoma. Its inconspicuous contrast with surrounding skin makes it a formidable
task for clinicians to detect. This research governs the approach that leverages
medical image segmentation to aid dermatologists in swift melanoma
identification. The AlexNet framework is used in this study to develop a skin
lesion segmentation and classification system. It begins with an
encoder-decodifier fully convolutional network (FCN) to recognise the complex
features of lesions and to learn about their boundaries using the decoder. Using
a succession of skip paths, FCN subnetworks are connected to each other. The
implementation involves the modification of two CNN architectures into fully
convolutional network models. Four distinct FCN architectures, namely modified
FCN-AlexNet, FCN-8s, FCN-16s, and FCN-32s, are employed to autonomously dissect
skin lesions based on their semantic characteristics. The evaluation on the HAM
10000 dataset, a pivotal aspect of this study, marks the pioneering assessment
of their capabilities on this dataset. Through these advancements, this research
seeks to enhance the diagnostic accuracy and efficiency of melanoma
identification in dermatology. |
Keywords: |
Skin Lesion, Classification, FCN, Alexnet, Melanoma, Encoder-Decoder, Deep
Learning |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
DETECTION OF CASING LACERATION USING DENSE CNN |
Author: |
MR.K.PHANI RAMA KRISHNA, DUVVURI ESWARA CHAITANYA DR.M.PREMA KUMAR, VENKATA
BHUJANGA RAO MADAMANCHI, BALAJI. TATA, KURRA UPENDRA CHOWDARY |
Abstract: |
The conveyance of dermatological administrations could be totally changed by the
utilization of teledermatology. Using broadcast interchanges progresses,
teledermatology is utilized to pass clinical information on to experts to
investigate disease. The goal of our investigation is to perceive skin wounds by
gathering the image trial of skin bruises that were gained from various
patients. In this work, input information is taken from the Kaggle. The next
step involves scaling the input images using the PC vision library in order to
focus in more on the irritated area. The dataset is then divided into
preparation dataset and testing dataset. Eighty percent of the dataset is
utilised for planning and twenty percent is used for testing in the accompanying
stage. Through the use of planning datasets, our suggested DenseNet Model—which
has five convolutional layers—is suitable for use at 100 years old. The testing
dataset is attempted by the pre-configured DenseNet model, and the accuracy is
measured and reviewed. Our experimental analyses highlight how information
pictures can identify skin injuries. |
Keywords: |
Casing, Laceration, Uncovering, Dense CNN |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
DYNAFLOW DEVICE OPTIMAL PLACEMENT USING ARTIFICIAL INTELLIGENCE |
Author: |
RAGALEELA DALAPATI RAO, PADMANABHA RAJU CHINDA, KUMAR CHERUKUPALLI |
Abstract: |
Numerous studies on steady-state control problems in power systems have made
heavy use of flexible AC transmission systems (FACTS). For example, the Combined
controllers, also known as the Dynaflow Controller, are just one of the several
available FACTS controllers. This coordinated controller combines a TCPST and a
TSSC, making it a new member of the FACTS family. It also belongs to the FACTS
group of standards. Power flow can be controlled in either the way the Dynaflow
Controller is positioned, or in parallel, using the combined skills of TSSC and
TCPST. To address this issue, the particle movement optimization-based bee
colony algorithm (PMBCA) has been proposed. In order to address the OPF problem
under a wider range of conditions, including normal operation, network
contingency, and overload, the idea of using Decision Making to determine the
optimal location of a Dynaflow device has been developed. The regular case, the
network emergency case, and the network overflow case are all examples. The
outcomes of the IEEE 30-bus system are used to demonstrate the proposed method.
The results indicate that the dynaflow device may be placed most effectively
using MADM techniques. |
Keywords: |
Dynaflow Controller; MADM methods; OPF problem; Particle Movement Bee Colony. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
PREDICTING THE DIFFICULTY OF WEAVING A NEW FABRIC USING ARTIFICIAL INTELLIGENCE
(FUZZY LOGIC) |
Author: |
M. EL BAKKALI, R. MESSNAOUI, O. CHERKAOUI, AZIZ SOULHI |
Abstract: |
This research work presented in this paper is the modelling of the prediction of
the weavability of a new fabric at the time of its creation. Using artificial
intelligence (fuzzy logic), we show the feasibility of a decision support model
for designers and production experts. This model is based on fuzzy set theory
and uses knowledge and expertise to help designers predict the degree of
difficulty in making new fabric and to avoid material waste, time loss, and
material damage. With this model, designers have more opportunity to choose the
right decisions. Below we find an application of this model based on fuzzy
logic. The simulations produced convincing results, and demonstrated the extent
to which the knowledge and expertise of weaving experts can be exploited to
anticipate production problems. |
Keywords: |
Weaving, Weavability Limits, Fuzzy Logic, Modelling, Saturation Index. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
AN ENSEMBLE APPROACH FOR HARVEST VINTAGE FORECAST WITH MACHINE LEARNING
TECHNIQUES: AN EXPERIMENTAL STUDY |
Author: |
DR.SURESH KUMAR PITTALA, DR. KONGARA SRINIVASA RAO, BALAJI.TATA, P.RAVI KUMAR,
N.JAYA, DR Y ANURADHA, KURRA UPENDRA CHOWDARY |
Abstract: |
Since crop yield forecasting directly affects the production and security of
food, it is an essential part of agricultural research and development. The
traditional methods of calculating harvest vintages centered on
agriculturalists' explanations and expertise have become increasingly
challenging as a result of the rapid ups and downs in soil and climatic state of
affairs. Machine-learning techniques have been applied in recent years to find a
solution to this issue. This study centers around the utilization of a few AI
models, for example, nearest neighbor relapse, closest polynomial relapse,
irregular woodland relapse, slope helped tree relapse, and backing vector
relapse, for the expectation of farming efficiency. The raw data was transformed
into a format that is conducive to machine learning using efficient feature
selection techniques. The study's findings showed that, in comparison to other
tactics, when choosing features, using an ensemble technique can lead to more
accurate predictions. By combining several data sources and machine learning
algorithms, the ensemble approach generates a detailed and precise forecast.
This work highlights both the benefits of using machine learning techniques to
forecast agricultural amount produced and the recompenses of using a
collaborative line of attack to feature selection |
Keywords: |
Harvest, Vintage, Ensemble, Forecast, Agriculture. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
AN EFFICIENT CYBER SECURITY AND DATA SCIENCE FOR ANALYZING BIG MEDICAL DATA |
Author: |
MOHAMMAD AL-OMAR, SALEH ALOMARI, TAMER BANI AMER |
Abstract: |
This study aims to the impact of the use of cyber security and data science in
analyzing big medical data. The study sample comprised 120 participants,
including hospital chief information officers, chief information security
officers, and healthcare cyber security professionals, who were selected from
all 33 government hospitals in Jordan connected to the Ministry of Health as the
research sample. The primary independent variable, cyber security, was evaluated
using information security, network security, operational security, and end-user
education. massive amounts of medical data were used as the dependent variable.
The study used SPSS to determine the impact of cybersecurity on the analysis of
big medical data in Jordanian hospitals. The results found that 75% of
participants confirmed that analyzing big data in the medical field will have a
high impact on the evaluation of medical diagnoses and 63.3% of the participants
agreed that analyzing big data in the medical field, it will have a high impact
in predicting the incidence of diseases, the results also found the role of
cybersecurity in protecting the storage of a large amount of data in hospital
information systems (HIS), ranking first with an arithmetic mean of (3.73), The
study recommends that future research should explore the benefits to medical
organizations of analyzing structured and unstructured data in clinical and
administrative fields, such as the limitations they face in these areas.
Additionally, it is suggested that further research should also include medical
institutions from outside Jordan borders to enable international comparative
analyses. |
Keywords: |
Cyber Security; Big Medical Data; Data Analyzing; Jordan Hospitals |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
AN IOT FRAMEWORK FOR ACCURATE DIAGNOSIS OF COVID-19 TO ELIMINATE FALSE POSITIVES
USING RESNEXT |
Author: |
A. PHANI SRIDHAR, DR.P.V.LAKSHMI |
Abstract: |
Background: Novel Corona Virus is increasing day by day; it is needed to
identify the formal techniques and innovative trendy approaches. The responsive
effort in short times is needed. Objective: The aim of this study was to
consequently, display things that recognize COVID-19 pneumonia accurately. This
examine work is pointing to particularly offer assistance with the conclusion of
COVID-19. Methods: The proposed work is centred on the headway of an AI-based
examination of CT pictures of the cutting edge crown infection. The proposed
system classifies CT pictures utilizing balanced Resnet appear name ResNext. The
Convolutional layers of particular sizes were utilized in each of the confined
ways of Resnet illustrated. In Resnext, 32 channels are bound together at the
same bottleneck and convolve them. This made it conceivable to perform
indistinguishable changes to gather convolution in 32 bunches, which compares to
the initial 32 courses. The proposed show effectively separated between viral
pneumonia and COVID-19 influenced lung CT pictures. The accuracies of DenseNet,
mobileNet and VggNet are 90.91,75.24 and 35.75 respectively for testing as shown
in table 4. Results: False positives are identified among normal images, Covid
effected images and viral pneumonia images. Also the validation loss and
validation accuracy of the training process is to be processed and observed. The
training loss is also calculated and observed. The training accuracy and
validation accuracy both reached 100%. The final testing accuracy of the
proposed model is 100%. The performance comparison of existing methods with the
proposed method is also observed. The proposed method obtained better
classification accuracy when compared to the existing deep learning models
DenseNet, mobileNet and VggNet. The accuracies of DenseNet, mobileNet and VggNet
are 89.3, 72.72 and 30.30 respectively for training. Conclusion: The clinical
execution of the PCR test for COVID-19 pneumonia is not accurate. This paper
points to assist make stride the AI examination stage for COVID-19 pneumonia
investigate and make strides the precision of AI choice and judgment. The
proposed demonstrate gotten an exactness of 100% in dispensing with wrong
positives delivering accurate COVID-19 detection. |
Keywords: |
COVID-19, lung CT, Resnet, ResNext, viral pneumonia, Reverse Transcriptase
Polymerase Chain Reaction (RT-PCR) |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ENGENDERING FINEST TEST CASES BY USING G-GENETIC ALGORITHM VARIANT SSGA |
Author: |
T J PRASANNA KUMAR, JASTI SURENDRA, PONNURU ANUSHA, DR M. BABU PRASAD, MANASA
BANDLAMUDI, K.PRAVEEN KUMAR, ANIL KUMAR PALLIKONDA |
Abstract: |
Programming Testing is the most difficult occupation among every one of the
companions of the business. Thorough programming Testing is never conceivable
just streamlined programming testing is conceivable. Thus Programming Testing
can be seen as streamlining issue as it fall under NP Hard. Because of the
enormous number of experiments that are expected to perform adequate testing of
the ideal programming application; the assorted strategies to decrease the test
suite is required. One of the normal concentrated on techniques is eliminating
the repetitive experiments; the explanation is negligible number of experiments
and most extreme number of mistakes disengagement or uncovering. In this
examination work review is led to address the use and viability of Consistent
satisfy hereditary calculation to reduce the quantity of experiments that don't
added unmistakable worth in that frame of mind of test inclusion or where the
experiments can't disconnect blunders. Hereditary calculation is used in this
work to help in limiting the experiments or upgrading the experiments , where
the hereditary calculation produces the fundamental populace arbitrarily,
ascertains the wellness esteem utilizing inclusion measurements, and afterward
particular the posterity in sequential ages utilizing hereditary tasks
determination, get over and change. The hereditary displaying activities are
explicit and in light of the activity might shift to typical hereditary
calculation. This course of age is rehashed until there is no adjustment of the
wellness values for two successive ages, when there is no adjustment of the
information age for two emphases so intermingling achieved or a limited
experiment is accomplished. The aftereffects of review show the way that,
hereditary calculations can essentially lessen the size of the experiments |
Keywords: |
SSGA, NP Hard, Test Case, Error, Testing. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
THE REPRESENTATION OF THE CULTURAL CRISIS IN SOCIAL MEDIA AS A NEW ECONOMIC
REALITY IN THE CULTURE INDUSTRY |
Author: |
AHMAD MULYANA, DEWI SAD TANTI, AMINAH SWAENAWATI, IRMULANSATI TOMOHARDJO |
Abstract: |
In the dynamic landscape of the digital culture industry, messages on social
media go beyond mere expression, realising economic value as commodities
carefully packaged by content creators to garner followers. Operating within the
sphere of mass culture and the culture of irreverence, content representations
align themselves with market tastes, prompting an exploration of the complex
interplay between cultural preferences and economic motives. Using netnographic
methods, this research aims to uncover content creators' intentional design of
messages and identify dominant orientations within the social media industry.
Through a critical lens, this research reveals the ideological workings
intricately woven into the dynamics of digital content, investigating the
reflection and potential perpetuation of cultural norms and values. The results
show that vulgar language on platforms such as Bunda Corla and Nikita Mirzani
includes categories of profanity, racism, explicit content, harassment and
cyberbullying, demonstrating the multifaceted nature of content in the digital
cultural landscape. Beyond its direct impact, this type of content significantly
affects society's economy and culture, evolving alongside technological
advancements and shifting consumer trends. This research emphasises the
understanding required to navigate the intersection between cultural
representation, economic interests and societal evolution in the digital culture
industry. |
Keywords: |
Representation, Youtube, Culture Industry, Cultural Crisis, Netnography |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ASSESSING THE EFFECTIVENESS OF QUILLBOT-MEDIATED INSTRUCTION IN ENHANCING EFL
STUDENTS’ PARAPHRASING SKILLS |
Author: |
TAJ MOHAMMAD, ALI ABBAS FALAH ALZUBI, MOHD NAZIM, SOADA IDRIS KHAN |
Abstract: |
Paraphrasing skills have reportedly been challenging for EFL learners due to
limited linguistic exposure, which hinders their language development and
academic growth. This challenge necessitates effective learning strategies, such
as artificial intelligence (AI)-mediated tools, which have the potential to
improve writing skills. Therefore, QuillBot, an online paraphrasing tool, was
experimented with in this study to investigate its effect on developing
students' paraphrasing skills at university. The quasi-experimental method
research design was followed to achieve the study objectives. Thirty students
enrolled in Technical Writing were recruited. A test and a semi-structured
interview were employed for data collection. The QuillBot-mediated instructional
program highly benefited students’ paraphrasing skills in Technical Writing.
Also, students were satisfied with using QuillBot as it helped them improve
vocabulary, sentence structure, the substitution of grammar units, and
comprehensibility. In addition, the EFL students perceived QuillBot as
user-friendly, simplified, and adaptable. The current findings contribute to the
role of AI tools like QuillBot in improving students' paraphrasing skills;
therefore, teachers should incorporate AI tools in teaching writing. |
Keywords: |
QuillBot, University Students, Paraphrasing Skills, EFL Writing. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
A TECH SAVVY TOMATO LEAF AND FRUIT DISEASE CLASSIFICATION SYSTEM BASED ON
ENSEMBLE DEEP LEARNING MODEL AND SHUFFLED SHEPHERD OPTIMIZATION ALGORITHM |
Author: |
MRS. M. SHARMILA, DR. M. NATARAJAN |
Abstract: |
Deep learning based artificial intelligent solutions are proving to be very
successful in many different fields. It shows promising results in precision
agriculture as well. Hence in this article, disease classification model based
on deep learning has been proposed for identifying and classifying tomato leaf
and fruit diseases. The proposed system acquires data set from Kaggle image
repository and augments them using data warping techniques. The augmented data
is preprocessed using median filter and contrast enhanced using dynamic
histogram equalization. U2 net architecture is used for background removal and
Segnet algorithm is used for disease region segmentation. Features extraction is
achieved through Nasnet Large model and classified using ensemble model. The
obtained results are optimized using shuffled shepherd optimization algorithm
and the proposed system is tested against existing classifiers like and is
proved to work efficiently and accurately than the previous models. |
Keywords: |
Tomato Diseases, U2Net, Dynamic Histogram Equalization, Segnet, Nasnetlarge,
Ensemble Model. |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
MULTIMODAL SIGNAL ANALYSIS FOR DEPRESSION AND ANXIETY PREDICTION: A HYBRID
CNN-RNN APPROACH |
Author: |
MAMTA KUMARI, DR. MURALI KRISHNA PUTTAGUNTA, ANISHA J, DR. TAVITI NAIDU GONGADA,
DILFUZA GULYAMOVA |
Abstract: |
Anxiety is a serious global problem. The cornerstones of management are drug and
psychological therapy. Because some individuals will respond to traditional
medications, regardless of their efficacy, there is a demand for alternative
strategies for preventing and treating depression. The physiological markers
that can be employed in combination with a method to recognize depression are
described in this study. This article contributes to the successful
instructional approach by optimizing a convolutional neural network model. In
this instance, the attributes are extracted using a Convolutional Neural Network
(CNN), which increases the classification accuracy and predictability of
sadness. RNN forecasts the existence or absence of depression. This unique
approach aims to reduce training time and improve accuracy by combining the use
of convolutional neural networks with bi-directional short-term and long-term
memory. Using the suggested method, the CNN-RNN classifier uses ECG data to
predict sadness and anxiety. The proposed method for identifying depressive
illnesses effectively obtained 99.6% accuracy, 99.5% recall 99% precision, and a
99.7%F1 score using a combination of evaluation of heterogeneous emotional
signals to anticipate melancholy and assess stress levels. The success of the
proposed strategy is confirmed through the comparison of its outcome indicators
with the validation data. |
Keywords: |
Anxiety; Psychological Treatments; CNN; RNN; Melancholy |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING-DRIVEN DECISION-MAKING IN SPINAL
DISEASE TREATMENT |
Author: |
DR. SUNDARAMOORTHY K, DR. MOHAMMED SALEH AL ANSARI, DR. S. KOTESWARI, MAMTA
KUMARI |
Abstract: |
Aside from people with significant neurological abnormalities, it's unclear
whether conservative or surgical therapy of spinal disorders is better for each
patient. This research looks into the use of Artificial Intelligence (AI) and
Machine Learning (ML) in the therapy of spinal diseases decision-making. A
supervised ML was trained and tested using the datasets of 70 patients with goal
of accurately predicting the Oswestry Disability Index (ODI) 5 months following
operation or the beginning of conservative therapy. In addition, using tenfold
cross-validation, created an approach that forecasts the ODI of 5 arbitrarily
chosen testing patients. The use of AI in this study also enabled for a
comparison of the genuine patient data after 5 months with different treatment
forecast, revealing variances of up to 19.6%. In the supervised version used
here, ML can detect patients who'd receive from conservative treatment earlier
on and, on the other hand, avoid unnecessary delays and painful for patients
who'd gain from surgical treatment. Furthermore, despite tiny diagnostic
categories, this strategy can be applied in various other fields of medical as
an excellent instrument for decision-making when deciding between competing
treatments alternatives. |
Keywords: |
Artificial Intelligence, Machine Learning, Spine, ODI, Pain, Therapy |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ANALYTIC HIERARCHY PROCESS-BASED EVALUATION APPROACH FOR DIGITAL TECHNOLOGY
SERVICE MANAGEMENT |
Author: |
KATHARI SANTOSH, MR.P.RAVI PRAKASH, L. B. ABHANG, ISKANDAR MUDA, DR. DYUTI
BANERJEE, DR.S.SUMA CHRISTAL MARY, MANIKANDAN RENGARAJAN |
Abstract: |
Rapid digital technology advancement, automating of production procedures, and
effective decision-making in company operations are all results of the industry
4.0 revolution. There is a lack of empirical study examining the desire to
embrace industrial technology for managing services, considering the potential
advantages of industrial technologies in operations administration that have
been described in the existing literature. By discussing how disruptive
innovations that includes the Cloud Computing (CC), Internet of Things (IoT),
and Artificial Intelligence (AI) support service transformation in industrial
firms. In a time of lightning-fast technological development, the contributions
of research in information technology are important because they are essential
in spurring innovation, improving system performance, and tackling difficult
problems all of which help to shape the course of the digital revolution.
Component-level issues like data are transformed into information and knowledge
by the technologies using the Analytical Hierarchy Process (AHP) model, which
enables the evolution of services. Even though it is primarily required for
becoming an availability provider, the study identify that IoT is the
cornerstone of any service change. Advancing to an effective provider profile
requires AI. CC is specifically utilized to apply an industrializer approach,
resulting in established, predictable, and component products, in addition to
offering adaptability across every profile. The study examines the potential for
operations managers to efficiently and effectively recognize technologies
intended to increase productivity in operations according to the findings. |
Keywords: |
Industry 4.0; Digital Technology; Digital Transformation; Service Management;
Analytical Hierarchy Process; |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
HATE SPEECH DETECTION USING LSTM AND NAÏVE BAYES ALGORITHM |
Author: |
ANANDA WONGWATANA NASUTION , ARIEL ZAHRAN, MUHAMMAD FAZAR AKBAR ,GHINAA ZAIN
NABIILAH ,ROJALI |
Abstract: |
Hate speech detection requires effective strategies to ensure a safe and
inclusive online environment. This research paper presents a comparative study
of hate speech detection using Natural Language Processing (NLP) techniques,
specifically Naïve Bayes and Long Short-Term Memory (LSTM) approaches. The
objective is to develop models capable of automatically identifying and
analyzing hate speech in written language. The prevalence and impact of hate
speech are emphasized, as it can lead to psychological harm and incite criminal
acts. NLP offers a valuable tool for automatically detecting potentially
dangerous content and addressing this problem. The study utilizes a dynamically
generated dataset containing diverse words and expressions to train and evaluate
the Naïve Bayes and LSTM models. The results show that the LSTM and the Naïve
Bayes model, achieving an accuracy of 74% and 64%. |
Keywords: |
Hate Speech, Naive Bayes, LSTM, Text Classification,NLP |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
Title: |
ENERGY EXCHANGE PROCESS FOR SMART GRID BASED ON INTEGRATING BLOCKCHAIN WITH
GCN-LSTM |
Author: |
DR. VUDA SREENIVASA RAO, AFSANA ANJUM, DR.S.SUMA CHRISTAL MARY, IBRAHIM AQEEL,
DR. S. KOTESWARI, SHAMIM AHMAD KHAN, MANIKANDAN RENGARAJAN |
Abstract: |
The energy exchange process lies at the heart of the modern smart grid, a
transformative energy infrastructure. This process involves the seamless
transfer of electricity among various grid components, including consumers,
producers, and storage units, to meet dynamic demand patterns efficiently.
Traditional energy exchange systems often lack transparency, building it tough
on behalf of clients to track the source of their energy and verify the fairness
of prices. By addressing these issues and increasing effectiveness, security,
also transparency of energy trading and supply. The modernization of energy
systems has ushered in the era of the smart grid, promising enhanced efficiency
and resilience. In the pursuit of these goals, this study explores a novel
approach by integrating two cutting-edge technologies: Block chain and Graph
Convolutional Networks with Long Short-Term Memory (GCN-LSTM). Block chain,
renowned for its transparency and security features, is leveraged to enable
transparent, tamper-proof energy transactions within the smart grid.
Complementing this, GCN-LSTM, a fusion of graph neural networks and deep
learning, enhances grid intelligence and decision-making, optimizing energy
distribution and consumption patterns. This research delves into the intricacies
of this integration, offering insights into its benefits, challenges, and
potential applications. By combining the decentralized ledger capabilities of
Block chain with the data-driven power of GCN-LSTM, This study hopes to open the
door to a more resilient and adaptive smart grid, heralding a new era in energy
exchange and management. Overall, the proposed methods are highly effective and
have demonstrated their superiority by achieving an impressive classification
accuracy of 99.30%, which outperforms several existing state-of-the-art methods
in the same task. |
Keywords: |
Graph Convolutional Networks; Deep Learning; Long Short-Term Memory; Block
Chain; Smart Grid; |
Source: |
Journal of Theoretical and Applied Information Technology
31st Decmeber 2023 -- Vol. 101. No. 24-- 2023 |
Full
Text |
|
|
|