|
|
|
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
February 2024 | Vol. 102
No.4 |
|
Title: |
VULNERABILITY DETECTION IN SOFTWARE APPLICATIONS USING STATIC CODE ANALYSIS |
|
Author: |
DEEPAK KUMAR A, GNANAPRAKASAM C , PRABU SANKAR N, SENTHAMILARASI N, CHENNI
KUMARAN J, VINSTON RAJA R, SUSEENDRA R |
|
Abstract: |
In this modern era of technology where data and its integrity are vital for
organizations, software security has become a major area to focus on in the
software life cycle. Organizations must preserve the program's security to
ensure the computer program's availability, authenticity, and data integrity
delivered to the clients. The major focus in software security processes is to
find the vulnerabilities displayed in source code prior to the production phase
of the software product. Recognizing the bugs present in the code in the early
stages of the software lifecycle may help resolve the vulnerability findings in
the computer program and help the software developers settle those bugs. This
detection process is effective at runtime, but can also be performed in the
production phase where the computer program is under development and partially
implemented. A static code analysis process is used to detect vulnerabilities.
It can be done computerized or evaluated physically by development and testing
teams. The use of source code scanning tools that are mostly automated for
detecting vulnerabilities is utilized in this paper. These tools review the
source code for its quality based on several code metrics and identify bugs
present in the program. Unlike dynamic analysis methods, static code analysis
helps find the security vulnerabilities in the initial stages of the software
life cycle, where the software product is in the production phase and static
analysis does not require code to be in the execution state. |
|
Keywords: |
Static Analysis, Bug Detection, Vulnerabilities, Software Security |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
STRATEGIC ADVANCEMENT OF BUILDING INFORMATION MODELLING PRACTICES THROUGH
KNOWLEDGE GRAPH CONSTRUCTION USING ONTOLOGY AND DEEP LEARNING |
|
Author: |
HAYAT EL ASRI, ASMAA RETBI, SAMIR BENNANI |
|
Abstract: |
The integration of data from diverse sources continues to present a significant
challenge in the Architecture, Engineering, and Construction (AEC) industry. To
address this issue, our proposed approach leverages ontologies to structure and
organize the disparate data, while employing deep learning algorithms to extract
pertinent information and relationships. The resultant knowledge graph serves as
a valuable resource that facilitates decision-making, enhances the management of
construction projects, and fosters expert knowledge sharing within the domain.
To evaluate the effectiveness of our method, we conducted experiments on a
manually built and annotated dataset, and the results demonstrated the
approach’s commendable accuracy and efficiency. The proposed knowledge graph,
founded upon the synergistic amalgamation of ontology and deep learning, shows
the potential to significantly elevate the efficiency and effectiveness of the
construction industry by streamlining data integration, fostering knowledge
exchange, and empowering stakeholders with informed decision-making
capabilities. |
|
Keywords: |
AEC Industry, Building Information Modeling (BIM), Knowledge Graph, Ontology,
Deep Learning. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
COMPARISON OF OUTSTANDING EMPLOYEE SELECTION AT BENU COFFEE ROASTER USING
BROWN-GIBSON AND SIMPLE ADDITIVE WEIGHTING (SAW) METHODS |
|
Author: |
HANDRIZAL, HERRIYANCE, SYLVIA ALVIEONITA AYU PURNAMA |
|
Abstract: |
The selection of outstanding employees at Benu Coffee Roaster is carried out to
increase employee enthusiasm at work. The Brown-Gibson method and Simple
Additive Weighting (SAW) can be used to assist management in selecting
outstanding employees who have qualities and abilities that suit the company's
needs. The Brown-Gibson method is used to calculate employee performance levels
based on several predetermined performance indicators. Meanwhile, the Simple
Additive Weighting (SAW) method is used to rank employees based on previously
calculated performance values. This research was conducted by collecting
employee achievement data through interviews, observation, and data collection
from the company's management system. The data that has been collected is then
processed and analyzed using the Brown Gibson and Simple Additive Weighting
(SAW) methods. The research results show that the system developed can
contribute to management in selecting outstanding employees more efficiently and
accurately. By using this system, management gains new knowledge in selecting
outstanding employees more quickly and effectively, as well as minimizing
mistakes in selecting outstanding employees. |
|
Keywords: |
Decision Support System, Brown Gibson, Simple Additive Weighting, Benu Coffee,
DBMS |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
RETINAL DISEASE PREDICTION USING LEAKY RECTIFIED LINEAR UNIT BASED GATED
RECURRENT UNIT MODEL |
|
Author: |
JYOTHIRMAI JOSHI, MALINI MUDIGONDA, RAGHAVA MORUSUPALLI |
|
Abstract: |
Detecting retinal disease has become more difficult in recent years, attracting
a lot of attention. However, when predicting retinal disease, problems has
caused due to a lack of a thorough understanding of all components in disease
progression. The complexity and variability of retinal diseases can also have an
impact on prediction accuracy. In this research, a Leaky Rectified Linear Unit
(ReLU)-based Gated Recurrent Unit (GRU) classification model is suggested to
improve retinal disease detection accuracy. Data is obtained from the Retinal
Optical Coherence Tomography (OCT) image and the Noor Eye Hospital dataset. The
preprocessing is then done by using Minmax normalization techniques.
Additionally, the Local Ternary Pattern and Gray Level Co-occurrence Matrix
(GLCM) are used to identify and select the relevant features from the raw data.
Next, the tournament-based Levy Multiverse Optimization Algorithm (LMOA) feature
selection is used to improve model performance by identifying relevant features.
Finally, ReLU-based GRU classification was used to improve accuracy in retinal
eye disease. According to the results, the proposed technique attained a higher
accuracy of 99.01% in the OCT image dataset and 99.59% accuracy on the Noor Eye
dataset, which was greater than the existing approaches. |
|
Keywords: |
Age-related Macular Detection, Choroidal Neo-Vascularization, Gated Recurrent
Unit, Leaky Rectified Linear Unit, Optical Coherence Tomography. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
FAKE DRUG DETECTION USING QR CODES AND CONSENSUS BASED SECURITY ENHANCEMENT IN
DECENTRALIZED BLOCKCHAIN SYSTEM |
|
Author: |
GADDE PRANITHA, P.V.LAKSHMI |
|
Abstract: |
Fake medicines are becoming a growing problem in the pharmaceutical industry's
field of piracy and manufacturing. Implementing QR codes in the manufacturing
process is one way to address the problem of fake medication. But however, it
cannot solve the problem completely and so that the novel Decentralized
blockchain assisted Quick response (QR) code system (DcB assist QR) is
introduced in this research work. In this work, once if the drugs are created by
the manufacturer, the QR is generated for the corresponding drugs. After that
particulars are uploaded in the blockchain once gotten approval of government,
then it can be distributed over the hospitals. The drug details are stored
securely into the blockchain, so that intruders are cannot accessed and modify
the drug details. In blockchain, the data are encrypted using hyper elliptic
curve based cryptosystem (HEllC) model. Further, to secure the blockchain
network and prevents unauthorized users from validating bad transactions, an
Improved Practical Byzantine Fault Tolerance (IPBFT) consensus algorithm is
proposed for effective block verification. During drug transportation, the
temperature monitoring of drugs is enabled by Internet of Things (IoT) sensors
and whenever temperature crosses the threshold, the alert message is send to the
driver of the vehicle. Especially, an Inter Planetary File System (IPFS) is
employed to store drug temperature data in a decentralized way. Moreover, the
performance of the DcB assist QR system evaluated based on various several
performance metrics and compared to existing system. The DcB assist QR system
attained 21.02 seconds of less execution time and of throughput. |
|
Keywords: |
Blockchain, Hyper elliptic curve, Improved Practical Byzantine Fault Tolerance,
Inter Planetary File System, Fake Drug detection, QR code, Encryption. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
EXTERIOR WOUND DETECTION AND CLASSIFICATION USING FEATURES BASED DEEP CNN |
|
Author: |
L. MADHAVI DEVI, P. BHARAT SIVA VARMA, Dr. JAMMALAMADUGU RAVINDRANADH, D. HARI
KRISHNA, K.RAMESH CHANDRA, LAKSHMI RAMANI BURRA, Dr YALLANTI SOWJANYA KUMARI |
|
Abstract: |
The legitimate sickness determination is quite possibly of the main move toward
clinical therapy. As far as analysis, dermatology is quite possibly of the most
unstable and testing field. To make a right finding, dermatologists consistently
need more patients since skin injuries, a dangerous infection, can influence
individuals, all things considered. Altogether, for example, clever frameworks
to analyze skin disease early and all the more precisely, Exterior wound
recognition and order are fundamental. Basal Cell Carcinoma (BCC), melanocytic
Nevus (NV), Melanoma (MEL), Actinic Keratosis (AK), Harmless Keratosis Sore
(BKL), Squamous Cell Carcinoma (SCC), Dermatofibroma (DF), and Vascular injury
(VASC) are all subtypes of skin injuries that are all in all alluded to as
Multiclass skin injuries. The multi-class orders are as yet a troublesome errand
because of the great many skin injuries and their high likenesses. It demands a
lot of investment, and expense to physically recognize different skin sores from
dermoscopy pictures. In this manner, critical to foster mechanized diagnostics
methods can all the more precisely group skin sores of various classes.
Consequently this review presents Multiclass exterior wound and order using
cross breed highlight choice in light of Profound Convolutional Brain
Organization (DCNN). The awareness, exactness, and explicitness of the
engineering that is being given are utilized to assess its exhibition. |
|
Keywords: |
Dermatology, Exterior wound, features, DCNN |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
MULTI-LABEL INTENT CLASSIFICATION FOR EDUCATIONAL CHATBOT: A COMPARATIVE STUDY
USING PROBLEM TRANSFORMATION, ADAPTED ALGORITHM AND ENSEMBLE METHOD |
|
Author: |
AZIZAN ISA, WAN MOHD AMIR FAZAMIN WAN HAMZAH, MOHD KAMIR YUSOF, ISMAHAFEZI
ISMAIL, MOKHAIRI MAKHTAR |
|
Abstract: |
This article presents a comparative study of multi-label intent classification
for educational chatbots using three machine learning (ML) techniques: problem
transformation, adapted algorithm and ensemble method. In the context of
chatbots, user intent can be complex, potentially spanning multiple areas
simultaneously. Current single-label intent classification techniques often fail
to handle such intricate user intentions. Thus, an in-depth study of multi-label
intent classification was conducted, critically analysing the performance of
these techniques based on evaluation metrics such as accuracy, hamming loss,
precision, recall and F1-score. The results highlighted the superiority of the
problem transformation technique, particularly the label powerset method, over
the other two methods across all evaluation metrics. Significantly, the label
powerset methodology demonstrated remarkable performance with a substantial
accuracy rate of 0.9669 and a minimal hamming loss of 0.0132, showcasing its
efficacy in handling tasks associated with multi-label intent classification.
The adapted algorithm and ensemble method displayed positive results but did not
surpass the problem transformation technique. This study offers valuable
insights for researchers and developers seeking to design an efficient and
accurate intent classification for educational chatbots. |
|
Keywords: |
Educational chatbot, Classification, Problem Transformation, Adapted Algorithm,
Ensemble Method. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
MACHINE LEARNING FOR ALZHEIMER DETECTION: A COMPREHENSIVE APPROACH |
|
Author: |
CHINNU MARY GEORGE,SANTHOSH MENON |
|
Abstract: |
Alzheimers disease (AD) represents a significant and growing concern worldwide,
particularly among older adults, as it remains the leading cause of dementia.
The increasing incidence rates of AD, along with its profound impact on
individuals, families, and healthcare systems, highlight the urgent need for
effective diagnostic tools. AD is characterized by progressive neurodegenerative
changes within the brain, making early detection critical for effective
treatment and minimizing potential damage. Given the challenges of predicting AD
in its initial stages, this research explores various Machine Learning (ML)
models, including Support Vector Machine (SVM), Random Forest (RF), and Gradient
Boosting (XGBoost), to develop accurate prediction models. Utilizing datasets
from Kaggle, this study employs two distinct feature extraction methods: Local
Binary Patterns (LBP) and Discrete Wavelet Transform (DWT). Both feature sets
are fed into ML models, and the performance of these models is evaluated using
essential metrics, including accuracy, precision, F1 score, True Positive Rate
(TPR), True Negative Rate (TNR), False Positive Rate (FPR), and False Negative
Rate (FNR).Among the six evaluated models, the combination of the XGBOOST model
with DWT features stood out, proving to be the most effective in predicting
Alzheimer's Disease emerging as the standout performer, achieving the highest
accuracy rate of 97.88%. This research underscores the potential of ML in early
AD detection, offering a promising avenue for improving patient outcomes and
alleviating the societal, financial, and economic burdens associated with this
devastating condition. |
|
Keywords: |
Alzheimer, Kaggle, Local Binary Pattern, Discrete Wavelet Transform, Machine
Learning, XGBoost, Accuracy |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
BACK POSE BASED PERSON RECOGNITION WITH PIX2PIX GENERATIVE ADVERSARIAL NETWORK |
|
Author: |
MOHANKRISHNA SAMANTULA, SRINIVAS GORLA, D. LALITHA BHASKARI |
|
Abstract: |
Face recognition has attracted greater attention during the past few decades,
there were various algorithms that were proposed for the task and many generate
accurate results. Our work initiated with a question – can we identify a person
when seen from behind? This led to the idea of back pose human recognition
(BPGAN). In our work, we introduced a novel approach (BPGAN) which involves
generating a person’s frontal image when the system is feed with images that
were captured from behind and images taken partially like when the person is
positioned at different angles like facing right and facing left. This approach
enables the AI based system to perform the task of recognition of a person even
when seen from behind, where it lacks the data to recognize the individual from
frontal data and has some back pose images of the individual. In this paper we
are trying to mimic the human behavior of identifying the person when seen from
behind relying on the body shape and his historical knowledge of that particular
person’s facial information. This algorithm helps in recognizing the individuals
in the crime scenes where the images are blurred or images from behind are
available. |
|
Keywords: |
Backpose, Pix2pix, GAN, CNN, Human recognition. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
SKIN ABRASIONS IDENTIFICATION WITH VISUAL GEOMETRY GROUP CNN |
|
Author: |
SURYA PRASADA RAO BORRA, M. JEEVAN SUJITHA, A. GEETHA DEVI, D. HARI KRISHNA,
LAKSHMI RAMANI BURRA, VUTUKURI SARVANI DUTI REKHA, ATCHUTHA BHAVANI SIRIKI |
|
Abstract: |
Perhaps of the most predominant malignant growth, skin disease, has filled in
prominence as of late. It is important to exactly analyze skin sores and
recognize harmless and threatening sores, which is really difficult, to furnish
patients with the consideration they require. Our review's goal is to classify
the skin injury photographs that we have gathered from different patients. The
Kaggle dataset fills in as the wellspring of the information for this
undertaking. The informational collection is parted into a preparation
informational index and a test informational collection in the following stage
after the information photographs are contracted utilizing the picture handling
library. The pictures ought to be developed to zero in additional on the sore
district. 20% of the informational collection is used for testing and 80% is
utilized for preparing in this system. Here, we recommended the VGG model, which
is learned through move learning and prepared for up to 50 ages utilizing the
preparation informational index, for ordering skin sores. On the test
informational collection, the prepared VGG model is scrutinized, and its
precision is estimated and evaluated. Our exploratory investigations show that
precise skin sore order is conceivable. |
|
Keywords: |
VGG,CNN, Classification, Skin Abrasion |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
FROM TEXTBOOKS TO CHATBOTS: THE INTEGRATION OF CHATGPT IN MODERN UNIVERSITY
PEDAGOGY |
|
Author: |
ALI ATEEQ, MOHANNAD MOUFEED AYYASH, MARWAN MILHEM, MOHAMMED ALZORAIKI, QADRI
KAMAL ALZAGHAL |
|
Abstract: |
In the present research, the relationship between academic performance, ChatGPT
chatbot, and traditional textbooks in the context of higher education is
empirically investigated. Understanding these interrelationships is essential
for developing successful educational methods in light of the quickly changing
educational environments characterized by the fusion of conventional pedagogy
and technology. A sample size of 210 participants was used in the research to
investigate the educational influence of ChatGPT chatbot and traditional
textbooks using Structural Equation Modelling (SEM). These were chosen using
convenience sampling, although only 173 respondents were actively involved,
providing a reasonably representative demographic. The empirical data produced
exciting results. A path coefficient of 0.204, a statistically significant
T-value of 2.705, a p-value of 0.003, and a strong positive correlation between
ChatGPT chatbot and academic performance were specifically found. Like
traditional textbooks, academic performance also showed a strong association, as
seen by the strong path coefficient of 0.423, the T-value of 2.663, and the
p-value of 0.004. The aforementioned results align with other scholarly
investigations that emphasize the significance of corporate governance in
promoting sustainable development and the impact of traditional textbooks on
improving organizational performance. Organizations aiming to enhance academic
performance should prioritize optimizing ChatGPT chatbots and traditional
textbooks as viable means of progress. |
|
Keywords: |
Artificial Intelligence, ChatGPT, Academic Performance, Chatbots, Higher
Education, Traditional Textbooks, Student Engagement, Bahrain |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
UNRAVELING THE INFLUENCE OF ARTIFICIAL INTELLIGENCE, ORGANIZATIONAL, AND
ENVIRONMENTAL FACTORS IN STRATEGIC PLANNING: IMPLICATIONS AND PRACTICAL INSIGHTS |
|
Author: |
IMAD AL MUALA, ABDALLAH MISHAEL OBEIDAT, ALI RATIB ALAWAMREH, BADAR ALHATMI,
AHMAD ABO EISHEH, ZAEDH HASSAN FAHAD ALRHABA |
|
Abstract: |
This study explores how artificial intelligence (AI) affects organizational
strategy planning. We discovered through thorough empirical investigation that
the strategic planning process is not greatly impacted by AI in our particular
corporate environment. This finding has important implications for resource
allocation, indicating that businesses should carefully consider how much money,
time, and talent they devote to AI projects based on their sector and
operational context. Our study promotes a balanced strategy, emphasizing the
significance of integrating AI with conventional strategic planning techniques
to improve decision-making. We additionally emphasize the importance of
enterprises determining their level of AI maturity and providing flexible
guidelines for AI integration. For enterprises attempting to traverse the
complex interaction between AI and strategic planning, real-world case studies,
decision support tools, change management techniques, and a proposed research
agenda all contribute to the provision of thorough insights and practical
recommendations. In the end, this research aims to improve strategic planning
procedures by providing useful advice or the adoption and integration of AI. |
|
Keywords: |
Unraveling, Artificial Intelligence, Organizational, Environmental Factors,
Strategic Planning. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
SYSTEMATIC REVIEW AND SIMULATIVE COMPARISON OF VIDEO WATERMARKING SCHEMES |
|
Author: |
HAWEEZ SHOWKAT, ROHUN NISA, Dr. ASIFA BABA |
|
Abstract: |
These days, one of the most sought-after data security study areas is
watermarking. Innumerable techniques have been developed to achieve better
results in terms of robustness and perceptual quality which are generally in a
tradeoff mode. In order to get insight into current trends in the watermarking
area, a non-exhaustive assessment of research publications was done in order to
weigh the advantages and disadvantages of each one. This paper's primary
contribution is a comparison of simulation results for video watermarking using
SVD, DWT-SVD, LWT-SVD, RWDT-SVD, and SVD-APDBCT, which provides an understanding
of different assessment parameters. Watermarked videos with various noise attack
variations have been subjected to simulations, and comparison analysis for
fidelity and resilience in terms of PSNR and CC is obtained. |
|
Keywords: |
SVD ,DWT,RDWT,LWT,APDBCT |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
LEARNING OBJECT FROM EMERGENCE TO NOWADAYS: SYSTEMATICS LITERATURE REVIEW |
|
Author: |
AHMED SALEM MOHAMEDHEN, NOUHA ARFAOUI, RIDHA EJBALI, MOHAMEDADE FAROUK NANNE |
|
Abstract: |
The e-learning domain is one of the richest fields of scientific research. It
has witnessed a rapid development and has become a popular topic since the
1990s. In this article, we present a systematic literature review of the
evolution and development of the learning object concept. Firstly, we present
the different proposed definitions, characteristics, and types of learning
objects. Secondly, we introduce learning object metadata standards, and learning
object models. Finally, the different types of recommender systems are
presented, as well as the different recommender systems and models developed to
recommend learning objects to learners. Analyzing and summarizing studies on
learning objects published in the 21st century is the objective of this study.
Through the use of appropriate keywords and the application of inclusion and
exclusion criteria, a total of 205 papers were identified during the search
process. |
|
Keywords: |
Learning Object, E-learning, Recommender system, Learning style, LOM |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
BLOCKCHAIN BASED AGRICULTURAL FRANCHISE IN YARUMORI COOPERATIVE |
|
Author: |
ANANDA DWI , GUNAWAN WANG |
|
Abstract: |
This study rigorously investigates the cutting-edge advancements in franchise
business models, aiming to rectify prevalent deficiencies. Through an exhaustive
literature review, this research proposes integrating blockchain technology and
Web3 frameworks into conventional franchise systems. Specifically, the study
spotlights Yarumori, an agricultural enterprise that adheres to a traditional
franchise model. The innovation initiative centralizes on two main goals:
firstly, to overcome operational hurdles inherent in the existing system, and
secondly, to revitalize the Yarumori franchise framework. This rejuvenation aims
to enhance the efficiency and effectiveness of its business operations
significantly. The proposed system employs Solidity and Ethereum for its
blockchain network, capitalizing on their widespread community support and
robust infrastructure. This research contributes substantially to the evolution
of franchise systems, positioning blockchain technology as a critical and
transformative element in modernizing business processes. |
|
Keywords: |
Blockchain, Waterfall, Etherium, Web3, Unified Modeling Language |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
NOVEL CONGESTION CONTROL MECHANISM TO IMPROVE PERFORMANCE OF MOBILE ADHOC
NETWORK WITH QUEUE MODEL |
|
Author: |
DR. S. HEMALATHA, KOMALA C R, A. AJINA, SURESH BALAKRISHNAN THANGASAMY, G.
VIKRAM, RAMU KUCHIPUDI, PRAMODKUMAR H KULKARNI, DR. HARIKUMAR PALLATHADKA |
|
Abstract: |
Self organized communication network without relay on any infrastructure is
named MANET. Transmission of packet in reliable communication will be done with
the help of the transport layer which is having the responsibility of flow
control, congestion control etc. If the packet transmission is fails and not
reaching to the destination node says that the entire network forming is vain.
This article proposed the novel algorithm and better buffer management technique
to resolve the Congestion problem in MANET .The novel algorithm does the role of
congestion avoidance technique eliminating unwanted packets, avoid multiple
flooding, detecting the attackers and dropout the packet. Buffer management
technique support for maintain the buffer in a threshold level , if reaches the
threshold level and refine the buffer space by keeping only data packets and
removing remaining all other packets .This buffer management and novel
congestion algorithm help to the MANET to improve the performance of the
communication network with the support of queue model |
|
Keywords: |
MANET, Congestion Control, Routing Protocol, Queue Management, NBM Algorithm |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
INVESTIGATION OF CASCADED 2DOF-PID CONTROLLER WITH IMPROVED INVASIVE WEED (IIW)
TECHNIQUE |
|
Author: |
GUDAVALLI MADHAVI, VEMULAPALLI HARIKA, VEERANNA, MAJAHAR HUSSAIN, AZAHARAHMED,
MUZEEB KHAN |
|
Abstract: |
A vast increment in power utilization aims towards the increment in power
generation, leading to complex advance in the control. The control approach for
frequency change with load is mainly exhibited in four main groups. Initially,
the Type of controller-Classical control approach involved the design of
different controllers in reducing divergence in frequency due to load
perturbations and Tie-line power flow due to power exchange between areas. In
this work a novel cascaded 2-Degree of freedom PID controller is proposed, and
comparative analysis is done with 2-Degree of freedom PID controller and PID is
applied in a secondary regulatory framework for multi-area interconnected power
systems, and investigations are carried out during random load disturbances,
plant generation participation changes, and parameter variations. Finally, Soft
computing approaches- optimizable techniques for tuning the controller
parameters. Due to utilization primary controller schemes will makes the
response to a very large time. Hence to obtain fast response a fast controller
scheme is required which has been proposed in this work. In view of various
metaheuristic techniques available can be utilized for gain optimization
parameters and better solution. |
|
Keywords: |
Cascaded two-degree freedom controller, Improved Invasive weed technique. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
IMPACT OF INFORMATION AND COMMUNICATION TECHNOLOGIES (ICT) AND THE KNOWLEDGE
ECONOMY ON ECONOMIC GROWTH: AN ANALYTICAL APPROACH BASED ON THE ARDL MODEL |
|
Author: |
YOUSSEF JOUALI, AIT EL MOUMEN TARIK, JAMILA JOUALI |
|
Abstract: |
This article delved into the profound impact of emerging Information and
Communication Technologies (ICT) and the knowledge economy on economic growth,
with a primary focus on the unique dynamics within the Moroccan context.
Leveraging the ARDL model and World Bank datasets, our survey systematically
analyzed various variables, revealing a significant association between several
factors and economic growth. Surprisingly, internet access showed no substantial
correlation in the Moroccan context. The core objective of this study was to
elucidate the intricate relationship among ICT, the knowledge economy, and
economic performance, contextualized within the distinctive features of the
Moroccan economic landscape. In this era which is defined by a rapid
technological evolution, our research provided valuable insights, advocating for
innovation and knowledge creation as pivotal strategies in the Moroccan context.
Additionally, the study uncovered a crucial link between investments in
research, development, higher education, and overall economic growth. It is
imperative to note that while this article focuses specifically on aspects
related to ICT and the knowledge economy in Morocco, it does not encompass all
contributing factors to economic growth. Nonetheless, it serves as a substantive
contribution, shedding light on trends that intertwine ICT, the knowledge
economy, and economic growth in the Moroccan context. The implications extend
beyond national boundaries, underlining the global significance of this research
field. Despite the valuable insights provided, we acknowledge inherent
limitations within the scope of this study. |
|
Keywords: |
Knowledge Economy, ICT, Economic Growth, GDP, ARDL. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
NEW ENCRYPTION ALGORITHM BASED ON ARTIFICIAL INTELLIGENCE’S FACE RECOGNITION,
SYMMETRIC ENCRYPTION, AND STEGANOGRAPHY |
|
Author: |
MIAD SALEM ALBALAWI , NUHA KAMAL HUWAYKIM, WAAD AHMED ALBRAIQI, MOHAMMED
ALWAKEEL |
|
Abstract: |
Encryption is critical to maintain the privacy, accuracy, and integrity of
sensitive data. It protects the confidentiality of data by making it
inaccessible to unauthorized parties. There are mainly two types of encryptions
used widely, symmetric encryption and asymmetric encryption, in addition to
encryption steganography is also used to protect data, where the data is hidden
within another object to avoid being accessed by unauthorized user.
In this
research we propose a new symmetric encryption technique, that uses artificial
intelligence’s face recognition, encryption, and steganography to encrypt and
protect the data. Firstly, to generate a unique encryption key, the face
attributes are extracted from a face photo using face recognition technique, and
then the encryption algorithm process these attributes to generate the
encryption key. Once the key is generated it is used by the encryption algorithm
to encrypt the original data and generate the cyphered data, finally
steganography technique is used to hide the cyphered data in a cover photo and
hide the face photo that was used to generate the key in another cover photo,
and then both cover photos will be sent to the receiver. At the receiver side,
the face photo that was originally used to generate the key, and the cyphered
data are retrieved from the two received cover images using steganography
technique, then the artificial intelligence’s face recognition technique is used
to extract the face attributes from the retrieved face photo, finally the
decryption algorithm process these attributes to generate the encryption key and
used it to decrypt the cyphered data and retrieve the original data. In the
final section of this research the encryption strength of the proposed technique
is discussed. |
|
Keywords: |
Symmetric Encryption, Private Key Generation, Artificial Intelligence,
Steganography, Face Attributes |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
SIGN LANGUAGE RECOGNITION OF WORDS AND SENTENCE PREDICTION USING LSTM AND NLP |
|
Author: |
RASHMI GAIKWAD, LALITA ADMUTHE |
|
Abstract: |
Sign language enables the speech and hearing impaired people with a way of
communication. These people understand sign language very well and can
communicate with each other easily. Problem arises when they want to communicate
with other people who do not understand sign language. To bridge this gap of
communication a real-time sign language recognition system is proposed in this
paper where meaningful sentences are generated from a few recognized signs of
words. In the first part of this research, Long and Short Term Memory (LSTM)
network is used for the recognition of signs of words. The network gives a
maximum accuracy of 97.53%. In the second part the recognized words are fed as
input to the natural language processing module for prediction of meaningful
sentences. The system can detect the signs performed by different signers even
if it is trained on dataset recorded on a single signer. |
|
Keywords: |
Sign Language Recognition (SLR), American Sign Language (ASL), MediaPipe
Holistic, Long Short Term Memory (LSTM), Natural Language Processing (NLP). |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
ADVANCING MALWARE ARTIFACT DETECTION AND ANALYSIS THROUGH MEMORY FORENSICS: A
COMPREHENSIVE LITERATURE REVIEW |
|
Author: |
ISHRAG HAMID, Razan Alajlan, Khaled Riad |
|
Abstract: |
This research paper conducts a thorough literature review on the role of memory
forensics in identifying and analyzing malware artifacts. With malware becoming
increasingly sophisticated, traditional detection techniques often fall short.
The paper traces the evolution of malware detection methods, from initial
signature-based approaches to contemporary techniques utilizing machine learning
and AI. It underscores memory forensics' critical role in identifying elusive
malware, thereby strengthening cybersecurity efforts. The paper examines various
memory forensic techniques, such as process and string analysis, and anomaly
detection. It also discusses the challenges posed by complex malware evasion
strategies and the necessity for specialized forensic tools and expertise. The
paper concludes by suggesting future research directions for improving memory
forensic methods to combat the ever-changing malware threat landscape, making it
a valuable resource for cybersecurity researchers. |
|
Keywords: |
Memory Forensics; Malware Analysis; Artifacts; Memory-based Analysis. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
ATHLETE SELECTION MODEL USING SPORTS STATISTICS DATA FABRIC TECHNIQUES FOR
NATIONAL SPORTS EXCELLENCE |
|
Author: |
CHANITA SATTABURUTH, PALLOP PIRIYASURAWONG, PRACHYANUN NILSOOK |
|
Abstract: |
Physical fitness factors that significantly impact athletes' performance. This
research objective is to analyze and develop a model for the successful
selection of athletes based on physical fitness factors. The sample comprises
data from high-potential athletes who attend sports schools affiliated with the
Thailand National Sports University. The research methodology combines
Multi-Layer Perceptron and Multiple Linear Regression as data analysis
techniques to identify suitable models for successful competitive athletes. The
results of the model evaluation indicate that the accuracy is 72.73% and the
R-squared value is 0.665. The experiment shows that analyzing the athlete
selection model could reveal the factors that influence the selection of
athletes. These factors will be described in this article. |
|
Keywords: |
Data Mining, Neural Network, Data Fabric Technique, Aquatic Athletes, Physical
Fitness |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
A NOVEL TECHNIQUE TO DETECT THE HOTSPOTS IN INFLUENZA EFFECTED REGIONS |
|
Author: |
P. NAGARAJ, KALLEPALLI ROHIT KUMAR, V.BIKSHAM, CHUNDURI KIRAN KUMAR, MANDADI
VASAVI |
|
Abstract: |
In modern days, healthcare is very important in human life. The World Health
Organization (WHO) publishes the latest articles on the humans suffering from
different types of flues. Machine Learning (ML) plays a major role in predicting
the diseases. The report of patients sometimes fails to provide the needed
information if the data is not recorded properly. So recently we have made an
endeavour to find the most globally spreading influenza virus in all over the
world called as Swine flu or the Influenza (H1N1) flu. Firstly we filter the
number of Patients died by swine flu using Artificial Neural Network (ANN) and
detected the location using Dynamic Boundary Location Algorithm (DBLA) and
geographic information system (GIS). The next finding is location where more
number of patients died termed as hotspots, applying Dynamic Hotspots Detection
Algorithm (DHDA), then using High Ranking Frequency Prioritization (HRFP)
algorithm high risk hotspots known as prioritized hotspots are detected. |
|
Keywords: |
Machine Learning, Ann, Gis, Dbla, Dhda, Hrfp. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
TECHNICAL ANALYSIS OF INTERNET SHUTDOWNS: ECONOMIC AND CYBERSECURITY DIMENSIONS
IN INDIA AND INTERNATIONAL CONTEXT |
|
Author: |
HARISH CHOWDHARY, DR NAVEEN KUMAR CHAUDHARY, FARAJ ABDULLAH HAZZA HARAHSHEH,
MOHAMMED AHMED MUSTAFA, DR MANINDRA RAJAK, RAJ KUMAR TOMAR |
|
Abstract: |
This article delves into the multifaceted repercussions of internet shutdowns,
with a specific lens on their economic and cybersecurity implications in India
and globally. The concept of internet shutdowns is explored as intentional
disruptions of internet and mobile services by authorities, marked by varying
degrees of scope and severity. The research underscores the profound impact of
these shutdowns, particularly in regions where mobile-based internet access
predominates, and broadband is less accessible. The paper includes an exhaustive
literature review, encompassing the themes of digital authoritarianism, control
over information flow, and the relationship between state authority, cyber
power, and public dissent. It extends the discourse to the realm of
cybersecurity, highlighting its critical role in sectors such as healthcare,
education, and public policy. The paper also emphasizes the importance of
incorporating cybersecurity policy and the development of organizational
cybersecurity capabilities. Methodologically, the paper outlines a rigorous
approach to standardizing diverse datasets on internet shutdowns from 2016 to
2022, utilizing data from the KeepItOn Shutdown Tracker Optimization Project.
This process entails data acquisition, harmonization, and cleaning, resulting in
a comprehensive, analyzable dataset. The empirical core of the study quantifies
the economic impacts of internet shutdowns through an analysis of shutdown
durations and their correlated economic consequences. A novel economic impact
per hour calculation is presented, alongside annual economic impact estimations
for India from 2016 to 2022. The paper also contemplates employing regression
analysis for more nuanced economic impact projections. Concluding with
cybersecurity considerations, the paper examines how internet shutdowns foster
environments conducive to cyber threats and vulnerabilities. The study calls for
robust strategies to mitigate these impacts, underscoring the criticality of
understanding and addressing the broad spectrum of effects engendered by
internet shutdowns. |
|
Keywords: |
Internet Shutdowns, Cybersecurity, Economic Impact, Internet Governance, Data,
Standardization, Digital Safety, Policy Analysis, India. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
ENHANCING MANET BATTERY LIFE USING MANET PERFORMANCE FACTORS AND CLUSTER HEAD |
|
Author: |
DR. S. HEMALATHA,RAJALKSHMY P, HARIHARAN N, BHARATHI GP, RAMU KUCHIPUDI,
VIJAYKUMAR KISAN JAVANJAL, NANDA ASHWIN, PROF. RAJESH P CHINCHEWADI |
|
Abstract: |
Power management in a wireless network is a time-consuming process, especially
in a Mobile Adhoc Network, because each node runs on its own node power. When
the internal battery dies, the entire communication system fails. Several
strategies are proposed to increase the performance of the MANET battery
management, and this might be used to evaluate the MANET's performance metrics.
This article presented new strategies that use internal node parameter
adjustments such as muting ideal nodes, beacon signal utilization, and changes
in MANET. This is achieved by forming a cluster head based on the mobile region
which does the role of forwarding packets by a single node, amid the clustered
to extend battery life, which is accomplished by a collaborative route
management mechanism among the nodes. The proposed research was simulated using
the Network simulator3 , and the produced result parameters were compared with
the existing related research work in AODV protocol , with the result concluding
that the unique method works best and saves 10 to 30 percent of power rather
than existing AODV protocol . |
|
Keywords: |
Sleep And Awake, Battery Life Time, Route Management, Forwarded Packet, Cluster
Head. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
LEARNING DISTRIBUTED REPRESENTATION OF DRUG SEQUENCES FROM ADVERSE EVENT
REPORTING DATA |
|
Author: |
RASHA ASSAF, AMJAD RATROUT, MOHAMMED KHALILIA, RASHID JAYOUSI |
|
Abstract: |
Predicting adverse event reactions (ADR) is challenging as it depends on the
patient’s condition, pre-existing conditions, and the different medications
being administered. One source of data for ADR that we believe is under-utilized
is the FDA Adverse Event Reporting System (FAERS) electronic submissions. FARES
contains a large number of ADRs including drugs and their attributes (timestamp,
dosage, route, duration, etc.), in addition to reactions and outcomes. In this
paper, we utilize FARES data to model each ADR as a sequence of medications to
train a model that learns the similarity between the drug sequence and the ADRs.
As a by-product of this work, we also learn the drug sequence representations,
which can be used for other down streaming tasks. Our model is based on a
transformer to encode drug sequences and adverse events, the model then outputs
the likelihood of the ADR given the drug sequence. Our best model achieved 0.87
F1 score, showing efficient representations for the drug sequences. We also
performed qualitative analysis to validate the drug sequence representations. To
our knowledge, we are the first to utilize drug sequences from FARES data to
learn drug embeddings. |
|
Keywords: |
Adverse Events, Embeddings, Sequence Modeling, LSTM, Transformers, RoBERTa |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
AN IMPROVED MUTUAL EXCLUSION MAC PROTOCOL FOR MAC LAYER IN MANET TO OVERCOME
HIDDEN AND EXPOSED TERMINAL PROBLEM |
|
Author: |
DR.S.HEMALATHA, K SREE DIVYA, VENKATAGURUNATHAM NAIDU KOLLU, SHAIK
SHAFI,CHETTIYAR VANI VIVEKANAND, G.VIKRAM6, RAMU KUCHIPUDI, PROF. RAJESH
PCHINCHEWADI |
|
Abstract: |
The Hidden and Exposed Terminal problems are the most difficult in Mobile Adhoc
Network, because node collisions reduce the performance of the Mobile Adhoc
Network. Many protocols were addressed to tackle the issues in hidden and
exposed protocols, but none of them were able to provide a permanent solution,
and the Hidden Exposed Nodes issues remain in the MANET. This article suggested
a novel approach that supports the MAC layer by creating Hidden and Exposed
Tables and sending MERT/MER signals. The proposed work, known as the ME-MAC
protocol, was implemented with the NS2.34 and the results were compared with the
traditional WiMARK protocol, CAD-CW protocol, and CFC-MAC protocol. Furthermore,
the proposed work achieved the maximum Packet Delivery Ratio of 60%, less End to
End Delay from 0 to 75ms, and higher Throughput. |
|
Keywords: |
MANET, ME-MAC protocol, MERT/MER, Hidden and Exposed node, MAC layer |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
CONSENSUS SCHMIDT-SAMOA JACCARD EXTREME LEARNED BLOCKCHAIN FOR ENHANCED SECURITY
WITH SIDE CHANNEL ATTACK DETECTION IN CLOUD |
|
Author: |
RAMAKRISHNA SUBBAREDDY, Dr.P. TAMIL SELVAN |
|
Abstract: |
Side Channel Analysis (SCA) is among the newly emerged threats to affect
confidential information by examining non-functional computation aspects of
program implementation such as elapsed time, amount of memory allocation, or
network packet size. The ability to automatically find out the amount of
information that a malicious user expands through side-channel observations
allows evaluation of the security of an application. During the access the
confidential information from the cloud, some amount of information is
discharged through side channels attacks. Therefore, it is important to specify
the amount of discharging in order to detect vulnerabilities in recovering
confidential information. In order to improve the side channel attack detection
accuracy, a new Cryptographic Consensus Schmidt-Samoa Jaccard Extreme Learned
Blockchain (CCSJELB) is introduced. The proposed CCSJELB technique performs two
different processing steps such as block generation and validation. In the first
step, the cloud server request to user about the registration. Then the server
generates the blocks for each user with the help of the HAVAL cryptographic hash
function to minimize communication overhead. Then the Wilcoxon signed-rank test
consensus mechanism is applied for active or inactive user-generated blocks.
After that, an improved Schmidt-Samoa Jaccard Extreme Learning Machine is
employed for block validation. The Extreme Learning Machine consists of many
layers for side-channel attack detection and secure data transmission. The
number of generated blocks for each cloud user is given to the input layer. For
each generated block, pair of keys is generated by applying improved
schmidt-Samoa cryptography. After that, encryption is performed with a public
key and stored on the cloud server. Whenever the user accesses the block, the
validation is performed based on the Jaccard coefficient. Then the sigmoid
activation function is applied to provide the attack classification results.
Finally, the decryption is performed to obtain original data. The attack
classification results, and secured communication results are obtained at the
output layer with higher accuracy. An experimental assessment of the proposed
CCSJELB is carried out with respect to communication overhead, throughput,
attack detection accuracy, false positive rate, attack detection time, and
confidentiality rate with a different number of traces. The analyzed performance
results designate that the CCSJELB increases data communication security with a
higher channel attack detection accuracy, throughput, confidentiality rate, and
minimum overhead and time than the conventional methods. |
|
Keywords: |
Cloud Computing, Side Channel Attacks, HAVAL Cryptographic Hash, Wilcoxon
Signed-Rank Test Consensus Mechanism, Improved Schmidt-Samoa Cryptography,
Extreme Learning Machine, Extreme Learning Machine, Extreme Learning Machine. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
DATA-DRIVEN PREDICTIVE ANALYSIS FOR CUTTING MACHINE FAILURES: A TECHNICAL REPORT
ON RELIABILITY OPTIMIZATION |
|
Author: |
Salma MAATAOUI, Ghita BENCHEIKH, Maria EZZIANI, Ghizlane BENCHEIKH |
|
Abstract: |
The prevention of recurring failures in modern manufacturing systems is of
paramount importance for minimizing costs and downtime. Despite the potential
for real-time data analysis using sensors offered by Industry 4.0 technologies,
their widespread adoption, particularly among smaller manufacturing systems,
remains a challenge. In response, this paper introduces an alternative approach
to predictive maintenance planning, utilizing historical maintenance
intervention data in the absence of sensor-based information. The study
investigates the pivotal role of Artificial Intelligence (AI), specifically
Machine Learning (ML) and Prognostics and Health Management (PHM), in augmenting
the efficiency of predictive maintenance. Utilizing a comprehensive dataset from
Schleuniger cutting machines spanning June 2021 to June 2023, our research
evaluates two predictive maintenance approaches: Precision-Based Maintenance
Prediction (PBMP) and Occurrence-Driven Maintenance Prediction (ODMP). The
objective is to extract valuable insights from historical maintenance data,
enabling proactive decision-making and preventing future failures. The
deployment results presented in this study demonstrate the effectiveness of
predicting the number of failures, providing valuable information that can
enhance maintenance planning and reduce total downtime. By addressing the
practical challenges faced by smaller companies in adopting sensor technologies,
this research contributes valuable insights to the broader landscape of
predictive maintenance in manufacturing. |
|
Keywords: |
Industry 4.0, Predictive maintenance, Artificial Intelligence, Prognostics and
Health Management, Schleuniger Cutting Machines, Data-Driven Decision-Making |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
CONSISTENCY AND CORRECTNESS OF REQUIREMENTS FOR ARTIFICIAL INTELLIGENCE SYSTEMS |
|
Author: |
DR.M.R RAJA RAMESH, DEVAKIVADA GANESH, NALLABARIKI PRAVEEN KUMAR, PROF. M. JAMES
STEPHEN |
|
Abstract: |
Artificial intelligence has become the part of our life with its advancement and
involvement in our day to day activities. Due to the trust we place on the
artificial intelligence systems, they must work as per the human expectations
with zero deviation. This can be achieved if the requirements are specified
correctly and all the requirements stated for the system must be consistent with
one another. Measuring consistency and correctness is sacrosanct for attaining
quality in the system that is to be developed. This paper proposes a set metrics
for measuring the correctness and consistency of the requirements stated in the
software requirements specification document. Initially the requirements
document was restructured according the format that is suitable for applying the
metrics. The factors consistency and correctness are divided into sub factors
and sub metrics are developed for measuring the same. Finally, the sub metrics
are combined to calculate the final metrics for consistency and correctness. The
proposed metrics are applied on the SRS document of Navigation and Maps
artificial intelligence system and this approach is compared with the existing
model and the results are satisfactory. |
|
Keywords: |
Consistency, Correctness, SRS, Metrics And Artificial Intelligence. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
MACHINE LEARNING ALGORITHMS IN QUALITY CONTROL OF TEXTILE FIBER MANUFACTURING |
|
Author: |
MIROSLAV TRANKOV, EMIL HADZHIKOLEV, STANKA HADZHIKOLEVA |
|
Abstract: |
The paper presents a study on the use of machine learning methods in performing
quality control in the production of textile fabrics. The main objectives,
problems and parameters in yarn quality control are discussed. The results of
using the Linear Regression, Logistic Regression, Decision Tree and Random
Forest machine learning algorithms on production data for various yarn
parameters are presented. Based on production data for 20 days, possible
deviations from the research parameters for the 21st day are predicted. The
approach used can also be applied for other aspects of the production process -
to optimize planning, analyze the performance of different machines, analyze the
raw materials used, identify poorly performing system components, etc. |
|
Keywords: |
Machine Learning, Textile Manufacturing, Textile Fiber Manufacturing, Yarn
Quality |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
A NOVEL FEDERATED-LEARNING BASED ADVERSARIAL FRAMEWORK FOR AUDIO-VISUAL SPEECH
ENHANCEMENT |
|
Author: |
MOHAMMED AMIN ALMAIAH, AITIZAZ ALI, RIMA SHISHAKLY, TAYSEER ALKHDOUR, ABDALWALI
LUTFI, MAHMAOD ALRAWAD |
|
Abstract: |
Current speech enhancement (SE) techniques operate in the spectral domain,
utilizing either edge computing or the cloud. Most existing frameworks offer
solutions for a limited number of noise conditions and rely on first-order
statistics. To address these limitations, researchers have explored machine
learning approaches to learn complex functions and train large datasets.
However, these models typically rely on centralized servers like the cloud,
which raises security concerns. Furthermore, running such training models on
edge devices is challenging due to their limited battery power and privacy
issues. In this study, we propose a federated learning-based SE framework for
multiple clients, using two speakers, to overcome these challenges. Our proposed
framework offers a decentralized model that allows for both local and global
training of data. Moreover, it is well-suited for adversarial networks and
private clinics as it preserves privacy on edge devices and in the cloud,
facilitating SE in a distributed fashion. The proposed model enables multiple
clients to train their data independently and send the aggregated training model
to the cloud. In contrast to existing approaches, our method operates at the
waveform level, training the model end-to-end and incorporating two speakers
with different noise conditions into a single model. This allows for sharing
model parameters with multiple clients using federated learning. Our approach
provides improved security, speed, and reduced battery usage for various clients
using hearing aids, resulting in enhanced robustness and other speech-centric
design choices to improve speech quality securely. |
|
Keywords: |
Speech Enhancement; Federated Learning; Cloud computing; Deep learning AV
dataset; SDG. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
REVOLUTIONIZING HEALTHCARE: UNLEASHING BLOCKCHAIN BRILLIANCE THROUGH FUZZY LOGIC
AUTHENTICATION |
|
Author: |
TAYSEER ALKHDOUR, MOHAMMED AMIN ALMAIAH, AITIZAZ ALI, ABDALWALI LUTFI, MAHMAOD
ALRAWAD, TING TIN TIN |
|
Abstract: |
In the ever-evolving landscape of digital healthcare, security and
authentication emerge as paramount concerns. This paper introduces an innovative
strategy aimed at tackling these challenges by melding the robustness of
blockchain technology with the precision of advanced fuzzy logic authentication.
We embark on an exploration of this fusion of cutting-edge technologies, diving
deep into how they collaboratively enhance the security, privacy, and efficiency
of healthcare systems. This pioneering approach has the potential to
revolutionize the management, accessibility, and protection of healthcare data,
ushering in a new era characterized by secure and patient-centric digital
healthcare. Digital healthcare systems play a crucial role in delivering
efficient and accessible healthcare services. Nevertheless, ensuring the
existence of secure authentication and key agreement mechanisms is imperative to
safeguard sensitive patient data and uphold the system’s integrity. Current
methodologies grapple with constraints related to vulnerability to cyberattacks,
scalability challenges, and optimizing resource allocation. Furthermore, the
integration of blockchain technology introduces additional layers of complexity
that necessitate careful consideration. This research advocates for an optimized
approach that combines fuzzy logic with blockchain technology to address
authentication and key agreement challenges within digital healthcare systems.
The proposed solution harnesses the adaptability and versatility of fuzzy logic
algorithms to navigate the realm of uncertainty and imprecision inherent in
authentication decisions. By leveraging fuzzy logic, the system can effectively
reduce false positives and false negatives, thereby reinforcing its resilience
against adversarial attacks. Moreover, the integration of blockchain technology
provides a decentralized and tamper-resistant infrastructure tailored for
securely storing and managing authentication and key agreement data. This
promotes transparency and trust within the system, mitigating the risks
associated with unauthorized access and data tampering. Additionally, the
blockchain-based architecture lends itself to efficient resource allocation and
scalability, enabling the system to promptly process authentication requests,
even in expansive digital healthcare environments. The effectiveness of the
proposed method is evaluated using the NIST Special Database 302, with results
demonstrating superior performance compared to existing approaches. It achieves
minimal False Rejection Rate (FRR), False Acceptance Rate (FAR), and response
time. Furthermore, the proposed method minimizes communication overhead during
authentication processes and exhibits resilience against a spectrum of
cyberattacks, including Replay attacks, Man-in-the-middle attacks, Denial of
Service (DoS) attacks, and Impersonation attacks. The combination of exceptional
security, efficiency, and resilience against diverse cyber threats positions
this solution as a promising choice for secure data sharing within peer-to-peer
(P2P) cloud environments. |
|
Keywords: |
Fuzzy Logic; Blockchain; Smart-contract, Lizard Search Algorithm, Homomorphic
Encryption, Cyber-attacks and Sustainable Development Goals (SDG). |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
DEEP LEARNING FOR MALWARE DETECTION: LITERATURE REVIEW |
|
Author: |
JAWHARA BOODAI, AMINAH ALQAHTANI, AND KHALED RIAD |
|
Abstract: |
Malware is among the biggest cybersecurity threats, that are changing all the
time to dodge traditional signature-based detection. In particular, machine
learning, especially deep learning, is a promising method for malware detection.
This paper provides an SLR of deep learning approaches for malware detection on
Windows, Android, IoT, and other platforms. In all, we searched five major
digital libraries and found 107 highly relevant studies published in 2015-2023.
The SLR methodology consisted of well-formulated search queries,
inclusion/exclusion criteria, and stringent full-text evaluation. Convolutional
neural networks (CNNs) are most popular, learning spatial patterns from raw
binaries. Malware sequential behaviors are modeled using LSTM networks. Spatial
and temporal learning are combined in ensemble models such as CNN-LSTM which
achieve high accuracy. But essential challenges persist, such as the
generalization problem under obfuscation, lack of transparency, and lack of
labeled real-world data. Although deep learning makes the malware detection more
accurate than traditional methods, evasion attacks, interpretability, and data
limitations need to be addressed. This SLR offers important insights into the
strengths, tendencies, datasets, and weaknesses of deep learning for strong
malware defense. With persistent threats, the use of effective AI-based
approaches will only further grow in importance. |
|
Keywords: |
Deep Learning; Malware Detection; Convolutional Neural Networks; Long Short-Term
Memory Networks |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
SENTIMENT ANALYSIS BASED ON PUBGM PLAYER ASPECTS FROM APP STORE REVIEWS USING
BIDIRECTIONAL ENCODER REPRESENTATION FROM TRANSFORMER (BERT) |
|
Author: |
HANDRIZAL, FUZY YUSIKA MANIK, HANIF MISBAH A |
|
Abstract: |
Players Unknown's Battlegrounds Mobile (PUBGM) stands as one of the most popular
mobile games among gamers, where players engage in online battles to survive
until only one player remains. With its growing popularity, many PUBGM players
share their experiences through reviews on the App Store. Therefore, the
analysis of player reviews on the App Store is crucial for understanding their
perspectives on the game and enhancing the gaming experience. This research
employs sentiment analysis using the Bidirectional Encoder Representations from
Transformers (BERT) model, achieving accuracies of 84%, 82%, and 83% across
three experiments with varying hyperparameter settings. Testing the number of
epochs reveals that epoch 3 yields favorable results and is consequently adopted
for sentiment analysis. The findings of this study suggest that increasing the
number of epochs does not necessarily lead to higher accuracy. The accuracy of
sentiment analysis is also influenced by the quality and quantity of the dataset
employed. High-quality datasets can enhance sentiment analysis accuracy, and an
abundance of high-quality datasets can further improve accuracy. |
|
Keywords: |
Bidirectional Encoder Representations from Transformers, Sentiment Analysis,
Game mobile, Players Unknown's Battlegrounds Mobile (PUBGM), User Experience,
App Store, Deep Learning, Transformers
|
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
INTRUSION DETECTION SYSTEM FOR (IOT) NETWORKS USING CONVOLUTIONAL NEURAL NETWORK
(CNN) AND XGBOOST ALGORITHM |
|
Author: |
FIRAS H. ZAWAIDEH, GHAYTH AL-ASAD, GHAITH SWANEH, SARA BATAINAH, HUSSAIN BAKKAR |
|
Abstract: |
Recently, the Internet of Things systems has seen a significant growth, as it
becomes a part of our daily activities. However, some related security issues of
rapidly evolving IoT threats need to be solved regarding the traditional
Intrusion Detection Systems (IDS). This work proposes a novel mechanism to
overcome the challenges of the traditional IDS including the computational
power, storage, and high rates of false alarms. Our mechanism uses Convolutional
Neural Network (CNN) to extract features. The CNN architecture is effective at
extracting significant features necessary for anomaly detection from the raw
data collected by IoT devices. These features are fed to the XGBoost, which is
superior at identifying intricate associations in the data, increasing the
accuracy of intrusion detection. The combination of Deep learning and ensemble
methods provides a robust solution to protect IoT environments against various
attacks. An IoT NetFlow- based dataset was used, named NF-bot-IoT dataset, it
contains four types of attacks within IoT network. The experiment is conducted
using python programming language, and the Results indicate how effective this
mechanism is, showing how it can detect a variety of attacks with high accuracy
equals 98.76. This study helps to strengthen the IoT's security framework and
increases its resistance to cyber threats. |
|
Keywords: |
CNN; IOT; Internet of things; security; deep learning; XGBoost algorithm |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
EDUCATIONAL SIMULATIONS IN HEALTH SECTOR: A NEW DIMENSION USING MACHINE LEARNING |
|
Author: |
D. BENAMMI, S. JAOUHAR, S. BOUREKKADI, B. OUMOKHTAR |
|
Abstract: |
The use of machine learning (ML) as an instrument in educational health games
represents a major advancement in the educational environment. This article
explores how ML, a branch of artificial intelligence, is integrated in
innovative ways to enrich the healthcare learning experience. The article
highlights the selection of suitable development platforms, such as Unity or
Unreal Engine, which enable smooth integration of ML-based features. It also
highlights the importance of rich medical databases and ML-based simulation
models to create realistic scenarios, thereby providing a faithful
representation of in-game healthcare situations. The development of machine
learning models is explored in detail, with an emphasis on adaptive
personalization of content based on user performance and needs. The use of
classification and recommendation algorithms to dynamically adjust the
difficulty level and provide intelligent feedback is also highlighted.
Integration with connected health devices is explored, showing how this linkage
allows real-time data to be collected, thereby enriching the gaming experience
and providing contextual information. Collaboration with healthcare
professionals is also emphasized, ensuring the clinical accuracy of information
and scenarios integrated into the simulations. The article highlights ML-based
continuous assessment to track user progress, as well as the integration of
real-time adaptation algorithms to adjust game parameters based on individual
performance. Pilot studies and the integration of user feedback are presented as
crucial aspects to evaluate educational effectiveness and continually improve
the game. The article explores the emergence of educational games in the field
of health, highlighting their innovative integration with machine learning. It
highlights the importance of this combination to improve health education,
providing interactive, adaptive and stimulating experiences. The article
discusses key aspects such as choosing the development platform, using medical
databases and simulation models, developing machine learning models, integrating
intuitive user interfaces, data security, connection with connected health
devices, collaboration with health professionals, pilot studies, and user
feedback. He highlighted the transformative potential of this approach,
highlighting how machine learning enriches health education through interactive
and personalized educational simulation. |
|
Keywords: |
Machine learning , Healthcare , Health education , Educational simulation. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
REVOLUTIONIZING LOGISTICS THROUGH DEEP LEARNING: INNOVATIVE SOLUTIONS TO
OPTIMIZE DATA SECURITY |
|
Author: |
M. BOUJARRA, A. LECHHAB, A. AL KARKOURI, I. ZRIGUI, Y. FAKHRI, S. BOUREKKADI |
|
Abstract: |
This article explores the impact of deep learning on logistics, an industry
facing increasing challenges such as rapid demand, efficiency and accuracy. Deep
learning is emerging as a revolutionary technology capable of transforming
supply chain management. The article examines the basics of deep learning,
machine learning and artificial intelligence before diving into real-world
applications in logistics. Successful business examples illustrate the benefits
in terms of inventory optimization, route planning and process automation.
Challenges, including data security, are analyzed, while highlighting promising
prospects and technological developments. The article is thought provoking about
the critical importance of deep learning in the digital transformation of
logistics, providing readers with an informed view of the opportunities and
challenges of adopting these innovative technologies. Our article makes a
significant contribution to research by exploring innovative applications of
deep learning in the field of logistics. As the logistics sector faces
increasing challenges in data security, our research aims to provide novel
solutions by leveraging the capabilities of deep learning. Our article makes a
significant contribution to research by exploring innovative applications of
deep learning in the field of logistics. As the logistics sector faces
increasing challenges in data security, our research aims to provide novel
solutions by leveraging the capabilities of deep learning. This abstract will
highlight the innovative aspects of our contribution, highlighting how our
approach can revolutionize logistics practices while ensuring robust data
security. |
|
Keywords: |
Logistics, Deep Learning, Data Security, Optimization, Inventory Management,
Digital Transformation. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
A NOVEL METHODOLOGY FOR SECURE DEDUPLICATION OF IMAGE DATA IN CLOUD COMPUTING
USING COMPRESSIVE SENSING AND RANDOM PIXEL EXCHANGING |
|
Author: |
PRATHAP ABBAREDDY, SREEDHAR BHUKYA, CHANDRAMOULI NARSINGOJU, B NARSIMHULU |
|
Abstract: |
With cloud computing technology, managing multimedia content became easier for
organizations in general and commercial content owners. However, duplication of
data causes unnecessary burden over cloud resources. At the same time, there is
security to be provided to multimedia content so as to ensure data integrity. In
this paper we considered these two concerns while proposing a novel methodology
for secure deduplication of image data in cloud computing. We proposed a
framework known as Image Security and Deduplication Framework (ISDF) which
exploits compressive sensing and random pixel exchanging for security and a
deduplication mechanism for getting rid of duplicate images while storing in
cloud resources. Compressive sensing is a signal processing technique used to
leverage image processing and image security. We proposed an algorithm named
Deduplication and Secure Image Storage and Retrieval in Cloud (DSISRC). This
algorithm exploits deduplication mechanisms and security mechanisms for
efficient management of image content in the cloud. Besides the deduplication
process benefits Cloud Service Provider (CSP) with optimal storage and
processing leading to conservation of resources. A benchmark dataset is used for
our empirical study. Experimental results revealed that the proposed algorithm
performs well in terms of image security and deduplication. |
|
Keywords: |
Secure Deduplication, Compressive Sensing, Data Anonymization, Cloud Computing,
Image Security |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
AN INNOVATIVE ENSEMBLE LEARNING METHODOLOGY FOR THE IDENTIFICATION OF MALARIA
USING MICROSCOPIC RED BLOOD CELL IMAGES |
|
Author: |
SYED AZAR ALI, DR. SINGAMSETTY PHANI KUMAR |
|
Abstract: |
Malaria is a potentially fatal parasite illness transmitted by female Anopheles
mosquitoes that are infected. Microscopists are able to identify this disease by
studying the sample of microscopic pictures of red blood cells. The detection
technique necessitates the expertise of a professional microscopist, which may
take less time and yield subpar results when used to large-scale diagnosis.
Infectious and noninfectious erythrocyte images are collected and processed into
models of transfer learning like as CNN, densnet201, Nasnet large, InceptionNet,
Xception, Hybrid (CNN + DenseNet201 + NasNet Large + InceptionNet + Xception),
KNN, SVM, Mobilenet, VGG16, Resnet50, InceptionV3, Densenet169, Resnet101,
Lenet, efficientnetV2S which are all trained on the same dataset which is taken
form Kaggle and augmented. The methodologies involving transfer acquisition and
fine-tuning are utilized, and the results are compared. but the efficiency is
very less, with individual method. Therefore, in this work we introduce an
innovative methodology, utilizing ensemble learning in combination with deep
learning techniques to accurately detect malaria parasites in red blood cells
Images. Resnet50(Residual Network), Inception (googlenet), and DenseNet201 are
three techniques utilized in employing an weighted average ensemble method. To
decrease the variability in estimations, a technique called max polling cluster
is employed together with weighted average collaborative models. Diverse image
processing approaches, such as the data expansion technique like augmentation,
boost etc. are employed to improve dataset and address the issue of overfitting
in the model. Additionally, other methods such as tradition CNN, Transfer
Knowledge, and CNN-ML classifier procedures are utilized to assess their
effectiveness in comparison to ensemble learning pattern. The model that has
been suggested in this paper achieves better efficiency compared to other
techniques discussed in literature survey, With a precision of 96.87%, it
successfully distinguishes between parasitic and healthy cells. Hence, the
method employed for deep learning possesses the capability to precisely and
autonomously detect malaria |
|
Keywords: |
Analysis ,Deep Learning, Malaria Disease, Machine Learning and Classification |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
SEMI BEAM SECTOR ANTENNA FOR MANET HET PROBLEM |
|
Author: |
DR.S.HEMALATHA, T KAVITHA, B. RAMESH, P JESU JAYARIN, NANDA ASHWIN, D. ROSY
SALOMI VICTORIA, RAMU KUCHIPUDI,HARIKUMAR PALLATHADK |
|
Abstract: |
The Mobile Adhoc Network is the most widely self-organizing and simply
established network for instant communication; nevertheless, addressing the
Hidden and Exposed node terminals limits the preference for quick communication
via MANET. Incorporating the directional antenna within the MAC layer
functionality can help with the hidden and exposed nodes problem. Initially,
Omni antennas were employed in MANET for transmission; however they did not
handle extended range, power optimization, or interference. This research
article discusses the incorporation of a semi-beam sector directional antenna
technology into the physical layer to address hidden and exposed node
difficulties in the MAC layer. This proposed antenna is divided into four
sectors based on the location of the neighbor node. The respective sector
transmitter or receiver will transmit or receive the packet. This antenna also
determines the receiver direction based on the location of the next hope
received and focuses the packet floating. This technique supports the Hidden and
Exposed node problem in MANET while also improving routing efficiency and power
optimization. This study was simulated using Network Simulation, and the results
showed a 20% to 30% improvement in total MANET Network performance and metric
value, as well as an overall antenna gain of 10 dBi in the Semi Beam sector
antenna. |
|
Keywords: |
MANET, Semi Beam Antenna, Physical Layer, Hidden Terminal, Exposed Terminal |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
ENHANCING CLOUD SECURITY BASED ON THE KYBER KEY ENCAPSULATION MECHANISM |
|
Author: |
IBRAHIM ALTARAWNI, MOHAMMED AMIN ALMAIAH, ANAS ALBADAREEN, KHALID ALTARAWNEH,
TAYSEER ALKHDOU, ABDALWALI LUTFI, AND MAHMAOD ALRAWAD |
|
Abstract: |
Due to its ability to provide flexible data processing and storage, cloud
computing has become a crucial component of today's technological
infrastructure. However, there is still serious concern about the security of
data transferred and kept on the cloud. By integrating the Kyber Key
Encapsulation Mechanism (KEM) into the cloud architecture, this article offers a
revolutionary method of improving cloud security. A post-quantum cryptography
method renowned for its strong defenses against quantum assaults is the Kyber
KEM. The possibility of quantum threats to conventional encryption techniques is
becoming more and more significant in cloud environments where sensitive data is
exchanged and stored. Cloud service providers can strengthen their encryption
protocols and guarantee that data is secure and private even in the face of
developing quantum computing capabilities by including Kyber KEM. The
application of Kyber KEM in a cloud security architecture is covered in this
study, with a focus on how it may offer secure key exchange, data secrecy, and
defense against quantum assaults. To further illustrate the usefulness and
efficacy of Kyber KEM, its performance and efficiency in a cloud environment are
assessed. Securing sensitive data processed and stored on the cloud requires the
integration of cutting-edge cryptographic algorithms like Kyber KEM, especially
in this day and age when data security is of utmost importance. This study is an
important step in guaranteeing the integrity and privacy of cloud-based data
since it clarifies how Kyber KEM can improve cloud security and defend against
changing cyberthreats. |
|
Keywords: |
Cloud Security; Encapsulation Mechanism; Kyber Key; cryptographic
algorithms. |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
Title: |
DOES IT-AFFORDANCE HAS AN IMPACT ON BUILDING CONSUMERS ATTITUDE TOWARD FASHION
LIVE STREAMING |
|
Author: |
WANDA WANDOKO, SITI RUQAYYAH KHALISHAH, SOLA GRACIA PUSPITA M S S |
|
Abstract: |
In recent years, fashion live streaming has become popular, but there is still
not much academic research on fashion live streaming. The main purpose of this
research is to examine the influence of IT affordance on consumers attitudes
toward fashion live streaming based on affordance theory and theory reasoned
action. in total, 375 usable questionnaires were used through online survey.
Structured equation modelling partial least square was used for data analyses in
this research. The result shows that meta voicing and guidance shopping have
significant effects on consumers attitude toward fashion live streaming, except
for visibility. The findings also suggest that consumers attitude toward fashion
live streaming had a significant impact on their intention to purchase. There
are several research limitations such as the approach used and the context of
the research location. This research has several implications for the
information systems literature, especially the IT-affordance literature, theory
of reasoned action. This research also has managerial implications for live
streaming application developers. |
|
Keywords: |
Affordance, Customers, Live Streaming, Attitude |
|
Source: |
Journal of Theoretical and Applied Information Technology
29th February 2024 -- Vol. 101. No. 4-- 2024 |
|
Full
Text |
|
|
|
