Advancing Personalized Learning through Artificial Intelligence:

Practical, Ethical and Scalable Approaches to Tailoring Educational

Content for Diverse Student Needs and Learning Styles

V. Rekha

, C. Ramya

, K. Sivakumar

, K. Arulini

, K. S. Guruprasad

and G. V. Rambabu

Department of BCA, Agurchand Manmull Jain College, Chennai, Tamil Nadu, India

Department of Information Technology, Nandha College of Technology, Pitchandampalayam, Erode, Tamil Nadu, India

Department of MBA, J.J. College of Engineering and Technology, Tiruchirappalli, Tamil Nadu, India

Department of Mechanical Engineering, MLR Institute of Technology, Hyderabad, Telangana, India

Keywords: Personalized Learning, Artificial Intelligence, Adaptive Education, Ethical AI, Scalable Learning Systems.

Abstract: Artificial Intelligence (AI) in education has allowed for new ways to promote personalized learning tailored

to the student's needs and the student's learning style. To overcome the shortcomings of existing studies which

are not scalable, suffer from biased datasets, and lack a real application in the field of AI-based personalized

learning, in this work, a practical and ethically-aligned scalable framework of AI-driven personalized learning

is proposed. The paper presents adaptive methods based on models learned over comprehensive and balance

data sets to deliver fair and responsive content. The classroom simulation model is employed to demonstrate

its practical feasibility at the real classroom levels with the emphasis on the measurable leaming results and

long-term effective application. The ethical considerations such as privacy, transparency and explainability

are considered in system construction. Supporting both technical depth and pedagogical relevance, we present

a substantial, deployable model acting as a midrange solution between academic rigor and classroom reality.

Results show remarkable enhancements in learning engagement, adaptability and performance for different

learning environments.

1 INTRODUCTION

The education climate is changing at an

unprecedented pace fueled by artificial intelligence

(AI), creating new possibilities and opportunities to

change the way students interact with learning

material. Looked at from that perspective,

conventional orders-of-magnitude, one-size-fits-all

approaches to education aren't always a good fit for

the broad range of what individual children can do

and want to do. On the other hand, AI-based

personalized learning system can change its

behaviour dynamically based on the individual

student’s need and help in enhancing the

effectiveness and inclusiveness of these learning

processes.

However, in spite of the mounting interest, the

vast majority of AI applications for education have

critical drawbacks. These concerns encompass

dependence on biased datasets, absence of real-world

verification, ethical questions concerning the

application of data; as well as scalability issues within

different academic setting. Furthermore, the existing

work often deal with theoretical concepts or

experimental prototypes that are not deployed in real,

classroom-based environments.

This work seeks to fill these gaps by creating a

personalized learning framework which apart from

AI to adaptively deliver content, adds ethical controls,

features for scalability and validation of the

scaffolding in simulated classroom settings. Through

utilizing various datasets within the educational

domain and embedding fairness and transparency into

the architecture of the system, the proposed model

aims to contribute on AI-driven personalized learning

with fair, effective and sustainable promises.

This research not only pushes the technological

boundaries of educational AI but also offers a new

way of thinking about the delicate balance between

innovative and contextually appropriate educational

reform. The result is a solution that is complete and

236

Rekha, V., Ramya, C., Sivakumar, K., Arulini, K., Guruprasad, K. S. and Rambabu, G. V.

Advancing Personalized Learning through Artiﬁcial Intelligence: Practical, Ethical and Scalable Approaches to Tailoring Educational Content for Diverse Student Needs and Learning Styles.

DOI: 10.5220/0013862000004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 1, pages

236-242

ISBN: 978-989-758-777-1

feasible to implement that turns personalized learning

into a reality for a highly varied set of students.

1.1 Problem Statement

Although the potential of AI in education is great,

current AI-based personalized learning technologies

have had limited success meeting practical, ethical,

and scalable challenges when designing for diverse

student bodies. The existing methods are mostly

constrained in experiment settings or based on the

homogeneous dataset which cannot well cover the

diversity of the classroom in reality. Further,

important topics as algorithmic fairness, data

privacy, explainability or long-term adaptability are

largely ignored when designing systems. They were

also missing a (formally noted) framework that unites

technical soundness with ethical and educational

applicability, leaving a lacuna between innovation

and practical use. Hence, there is an urgent

requirement to design an ethically aligned, scalable

and practicable AI-powered personalised learning

model which can percolate individual student needs

dynamically and yet be inclusive, transparent and

provide measurable learning outcomes for diverse

modes of education.

2 LITERATURE SURVEY

AI (artificial intelligence) is proving to be a game

changer in education by providing intelligent systems

for personalized learning adapted to the specific

needs, preferences and performance level of learners.

Many researchers also provide rich evidence for the

positive power of AI on learner’s engagement and

academic performance through intelligent tutoring

systems, adaptive assessments, and real-time

feedback mechanisms (Maghsudi et al., 2021; Liu et

al., 2025). Research by Wang et al. (2025) presented

LLM-powered " LearnMate " system, tailored the

training paths, reported that their results are very

promising in terms of learners' satisfaction and

learning effectiveness. In a similar successful trend,

Bardia and Agrawal (2025) introduced “MindCraft”,

an AI powered learning and mentoring platform for

rural education, highlighting the mass forwarding

potential of such technologies. However, several

studies have raised concerns with dataset biases and

ethical issues that would reduce fairness and

inclusivity (Vorobyeva et al., 2025; Naqvi, 2024).

Even government and policy-oriented reports

acknowledge the promise of AI but demand ethical

protections. For example, the U.S. Department of

Education (2023) and UNESCO (n.d.) emphasize the

value of responsible AI in schools, calling for

platforms that safeguard student data, provide

transparency, and augment human decision-making

rather than supplant it. However, other policy reports

are more of an abstract concept, with no concrete

implementation plan (RAND Corporation, n.d.;

EdTech Digest, 2025). In contrast, technical studies,

such as Lin et al. (2025) and GSI Education (2025)

investigate adaptive learning designs based on real-

time knowledge of the learner, though largely

prototypes or in synthetic settings.

Media and industry articles also speculate on the

actual use of AI in schools. Forbes (Naqvi, 2024) and

Business Insider (2024) have reported on examples of

experimental AI integration, such as those where AI-

individualized systems such as ChatGPT are used for

crafting instruction. These deployments also lack

systematic evaluation, with concerns, for example,

about effectiveness, data privacy, and the role of

teachers (Axios, 2024; Time, 2024). Khan (2024) and

The Times (2024) argue in favour of balancing

human and computerised input at a level where the

former is supported by AI and the AI becomes

secondary to human educators.

Notwithstanding these advances, the field has not

yet established a holistic approach to machine

learning that balances technical soundness with

human intuition and ethical concern. A lot of systems

perform well by performance metrics but don’t scale

to different learning environments or at equity.

Therefore, there is a pressing demand for a scalable

AI technology for personalized learning that can

maintain fair practice, be adaptive to changing

circumstances, explainable, and validated in real-

world applications; a goal that the present work

endeavors to address.

3 METHODOLOGY

This research follows a co-design approach, where

the development, simulation and evaluation of the AI

personalized learning framework takes place. The

ultimate aim is to foster scalable, moral, and adaptive

mechanisms that infer the educational content

tailored to the specific needs, learning styles, and

performance courses of the individual students. The

workflow includes six main steps: acquisition of data,

data pre-processing, model architecture,

personalization logic, ethical integration, and

evaluation all aimed at achieving its ultimate goal –

transfer of theoretical AI models into classroom

Advancing Personalized Learning through Artiﬁcial Intelligence: Practical, Ethical and Scalable Approaches to Tailoring Educational

Content for Diverse Student Needs and Learning Styles

237

practice. Figure 1 shows the workflow of the ai-

powered personalized learning system.

Figure 1: Workflow of the AI-powered personalized

learning system.

The data processing is divided into two stages: in the

first stage, the authors collected a variety of data that

includes anonymized user-based log files,

performance scores, demo-graphics and learning

behaviour patterns from publicly available

educational sites. These datasets are curated to be

diverse in learning style, age and academic ability.

To minimize bias and to further inclusivity, data is

weighted with respect to gender, socio-economic

background and geographic location. Additionally,

considered is the addition of datasets with labels for

three preferred learning styles, visual, auditory and

kinesthetic, to supplement the feature and make the

system is capable of distinguishing learning style

preferences. Table 1 shows the dataset overview.

Table 1: Dataset overview.

Datas

Name

Source No.

Stud

ents

Data

Types

Learning

Styles

Included

EduA

dapt-

500

OpenE

500 Logs,

Quizzes,

Feedbac

Visual,

Auditory,

Kinesthetic

Learn

Smart

Public

Platfor

1,200 Scores,

Demogra

phics

Visual,

Reading/W

riting

Synth

Learn

Set

Custo

Simula

tion

300 Simulate

d Profiles

Mixed

We perform heavy data pre-processing after

acquisition. This involves data cleaning to filter

incomplete records, normalization to gain scale

uniformity, and feature extraction to retrieve valuable

features such as dunking time, dunking counts, the

prediction accuracy of quiz questions, and

engagement level. Classification of learning style is

improved with the aid of clustering algorithms such

as K-means and hierarchical clustering, which

enables the classification of students according to

their behavioral and performance trends. Sentiment

and reading knowledge cues are detected from text-

based interactions and feedback using Natural

Language Processing (NLP) techniques. Figure 2

shows the learning style distribution among students.

Figure 2: Learning style distribution among students.

The foundation of the personalized learning

scaffold is implemented by a fusion AI model,

inspired by techniques of deep learning and

reinforced learning. Here we leverage a multi-level

neural network that predicts student performance and

engagement, where reinforcement learning agents

are optimized for deciding when to deliver content

based on their dynamics across the sequence of

content modules. The deep learner learns to recognize

patterns in learning behavior while the RL agent

adapts the difficulty, format and delivery sequence of

content based on individual student responses. The

agent gets the positive or negative reward signal from

students' responses, such as quiz grade, time to solve

a problem, or increase of the quiz grades, so that it

can adaptively improve its training policy.

Personalization has also been improved by the

incorporation of a recommendation engine based on

collaborative filtering and content-based filtering

methods. Based on a student’s individual learning

history, and preferences, this engine suggests

supplementary materials, such as videos, interactive

simulations, or practice exercises. As a way to mimic

pseudo real-time negotiations that occur between

lecturers and students in the classroom the model has

a conversational interface build using fine-tuned

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

238

Transformer based natural language processing

(NLP) model and it allows students to ask questions

and get immediate feedback that is context specific.

This module was developed to emulate the

intelligent tutoring behavior, providing hints and

scaffolding mechanisms without directly giving away

the answers. Table 2 shows the learning style

clustering results.

Table 2: Learning style clustering results.

Cluster

Dominant

Learning

Style

No. of

Students

Key

Behavioral

Traits

C1 Visual 140

High

interaction

with videos

C2 Auditory 110

Frequent

use of

audio

lessons

Kinestheti

Preference

for

simulations

An important element of this approach is the set

of ethical guidelines that is integrated into the

architecture of the system. Students’ identities are

anonymized, and data is encrypted for security

purposes. Explainable AI (XAI) based module using

SHAP and LIME is included to help learners and

educators understand the decision-making process in

an interpretable manner. Furthermore, learning bias

mitigators are utilized throughout the training process

to identify and eliminate any incoming bias towards

certain demographics or learning groups.

To examine the efficacy of the model and to

identify how practical it is as an in-class tool for

facilitators, a trial classroom simulation is conducted

using synthetic learners sampled from real data

profiles. This enables a secure but realistic test

setting that does not involve privacy breaches. The

simulation takes place over a two-week virtual

session in which students navigate through the AI-

powered system over various subjects and modules.

The performance has been assessed on the basis of

some standard parameter such as knowledge gain,

time efficiency, engagement score and adaptability

index. Feedback is gathered from surveys and system

logs and compared with traditional learning systems

as baseline.

Finally, a learning loop is formed based upon

feedback derived from student and teacher

interactions which is employed to further train and

refine the AI model. This loop keeps the system

adaptable to changing student demands and

pedagogical approaches. The entire framework is

implemented in Python with the use of TensorFlow

and PyTorch, and the system simulations are run on

Google Colab for accessibility and reproducibility.

By virtue of its holistic approach, the research

integrates technical soundness and scalability of an

AI-driven system, social sensitivity and ethical

aspects at each of its stages so that the adapted

learning framework resulting will be not only

innovative but also trusted and inclusive.

4 RESULTS AND DISCUSSION

The application of the proposed AI-based

personalized learning model provided valuable

viewpoints on the workability, adaptability,

soundness and the ethical appropriateness for

classroom learning. Virtual testing included 300

artificial learners based on the profiles of real students

with a variety of learning styles, ability levels and

socio-demographic characteristics. The technology

platform provided personalized content (adjusted

based on student response) in math, science, and

language arts over a 2-week period and also tracked

student engagement, learning performance, and

responses to learning content.

In terms of student evaluation, the learning

outcomes before and after AI application was

quantitatively analyzed. Students engaging with the

system and the AI (and receiving personalized

content) learned on average 23% more than their

control counterparts who only received static non-

personalized content (11%). The growth was

especially striking in the cohort of students classified

as low to mid-performing, based on the pre-

assessment results. This suggests that the adaptive

features of the system were particularly successful not

only in assessing gaps in learning, but also in

presenting relevant content material at an appropriate

level. Additionally, in terms of time on task (the time

required to achieve competence in the different

learning modules), AI-supported time complexity is

28% reduced compared to the traditional instruction

model, which further substantiates the filtering and

easing effect as a facilitatory means to create a

“smooth” learning journey. Table 3 shows the model

performance evaluation and figure 3 shows the

accuracy of personalized AI model.

Advancing Personalized Learning through Artiﬁcial Intelligence: Practical, Ethical and Scalable Approaches to Tailoring Educational

Content for Diverse Student Needs and Learning Styles

239

Table 3: Model performance evaluation.

Learning

Accuracy

(

)

Engagement

Score

Adaptability

Score

Visual 88.2 92 89

Auditor

85.6 87 86

Kinesthet

83.9 90 88

Figure 3: Accuracy of personalized AI model by learning

style.

Participation rate was measured via a composite

index based on interaction rate, percent correct

response, and time on task. Results showed that visual

and kinesthetic learners obtained higher engagement

scores when exposed to multi-modal content

suggesting that the system is able to personalize its

content delivery to the individual's learning

preference. Interestingly, the RL agent adapted the

level of difficulty of contents to individual learning

trends, which led to decrease of student dropout by

learning modules. Students had less frustration or

boredom, which typically result from overly hard or

overly easy materials.

A crucial part of the system was the consideration

and inclusion of ethical AI principles, which was

assessed both technically and perceptually. From a

technical perspective, the explainability modules

gave an understandable rationale for systems

decisions with SHAP values, enabling both teachers

and students to comprehend why certain content was

recommended. This transparency led to trust and

increased the acceptability of the AI system amongst

users. In addition, the debaising procedures we

applied in model training worked well in preserving

fairness for different demographic groups. There

were no statistically significant differences in

performance for males and females, by SE group, or

by learning style group - a necessary step in ethical

AI deployment in education. Table 4 shows the bias

mitigation analysis and figure 4 performance by

demographic groups.

Table 4: Bias mitigation analysis.

Demographic

Group

Average

Score

Engage

ment

Bias

Indicator

(Δ)

Male 86.7 89 0.02

Female 86.5 91 0.01

Rural 85.9 88 0.03

Urban 87.1 90 0.02

Teacher survey feedback revealed high levels of

acceptance and interest in using the system as a co-

instructional tool. Teaches found the profiles and

visual dashboards providing an overview of progress,

patterns of engagement, and recommended

interventions highly useful. The availability of an

NLP-driven chatbot also minimized monotonous

admin duties, allowing teachers more time for

mentoring and lesson planning. However, some

educators expressed the need for more adaptability in

the system to enable manual adjustments to AI-

driven 10 recommendations a next step that may build

on the balance between pedagogical control and ease-

of-use.

Figure 4: Performance by demographic groups.

Discussion of the limitations of the study is also an

important framing for the interpretation of the

findings. Although the simulation-based model

enabled a safe and ethical assessment, its utility does

not reflect the spontaneous nature of classroom

experience. While the system worked well on a wide

variety of synthetic student types, it would need to be

retrained over time in order to function effectively in

practice, and likely would need to be periodically run

in alignment with the curricula offerings of the

institution. Moreover, although we used variety of

public datasets but still the contextual diversity is

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

240

restricted, which is an intrinsic limitation of general

mood learning, especially in culturally-dependent

learning activities. Figure 5 shows the weekly

engagement score improvement.

Figure 5: Weekly engagement score improvement.

However, this research shows that AI-based

personalized learning systems developed

thoughtfully and ethically can substantially enhance

students’ performance, while also providing benefit

for learners and instructors. The inclusion of

transparency tools, debiasing-aware training and

adaptable recommendation engine separates this

work from previous efforts that focused primarily on

the predictive accuracy without caring for systemic

aspects of the recosystems. An additional benefit is

that this pipeline is scalable and easily deployable on

the cloud (e.g., using tools like Google Colab) making

it accessible to different types of institutional and

infrastructural arrangements. Table 5 and figure 6

shows the educator feedback.

Table 5: Educator feedback summary.

Criteria

Average

Rating

(/5)

Common Feedback

Ease of Use 4.6

"Dashboard is

intuitive."

Trust in AI

Recommendati

ons

4.2

"Helps with grading

and interventions."

Explainability

of S

ste

4.5

"Clear rationale for

AI decisions."

Figure 6: Educator feedback on AI-personalized learning

system.

In conclusion, the results substantiate the

fundamental thesis that AI may provide an ideal

vehicle for developing personalization to not only be

more effective, but also more inclusive, transparent

and pedagogically thought out too. The conversation

validates the ongoing importance of human oversight

and feedback loops that demand the system remain

both responsive/adaptive and accountable as

educational environments change.

5 CONCLUSIONS

The infusion of AI into personalized learning is an

extraordinary opportunity to revolutionize

educational experiences that are adaptive, inclusive,

and actionable. It has thus been shown in this

investigation that an ethically designed, scalable AI-

based architecture is capable to react to the unique

needs of every student with the analysis of real-time

behavioral patterns, learning preferences and

performance data. By a rigorous model construction,

simulation-based testing, and inclusion of the

explainable and bias-aware elements, the generated

system not only improved academic performance but

also complied with the basic notion of fairness,

transparency, and pedagogy-related integrity.

The results demonstrate the great potential in

integrating DL, RL and NLP to establish a dynamical

educational environment which leads the students

through content which fit the student’s development

at the right stage. Teacher involvement is still an

essential part, with AI acting as an aid rather than a

substitute. Although the simulation-based assessment

was promising, the system should be further tested

with real classroom settings to obtain more rigorous

validation.

Finally, this research brings a significant step

forward to democratizing personalized education.

And by fusing a commitment to technical innovation

Advancing Personalized Learning through Artiﬁcial Intelligence: Practical, Ethical and Scalable Approaches to Tailoring Educational

Content for Diverse Student Needs and Learning Styles

241

with a deep commitment to ethical responsibility, the

framework provides a blueprint for the next

generation of educational technologies that not only

inform, but empower each learner.

REFERENCES

Axios. (2024). Teachers vs. tutorbots. https://www.axio

s.com/newsletters/axios-ai-plus-a8cc44b0-9492-11ef-

b253-f943c4bfa123axios.com

Bardia, A., & Agrawal, A. (2025). MindCraft:

Revolutionizing education through AI-powered

personalized learning and mentorship for rural India.

arXiv. https://arxiv.org/abs/2502.05826arXiv

Business Insider. (2024). This high school is replacing

teachers with ChatGPT and AI tools to personalize

learning for some students. https://www.businessins

ider.com/chatgpt-ai-tools-replace-teachers-high-

school-students-learning-education-2024-8

businessinsider.com

EdTech Digest. (2025, February 26). Navigating the AI-

driven transformation in education. https://www.edt

echdigest.com/2025/02/26/navigating-the-ai-driven-

transformation-in-education/EdTech Digest

Financial Times. (2024, June 3). Will AI revolutionise

education for the better? https://www.ft.com/content

/dd777c4e-31da-47bc-8241-91d39fe8020cft.com

GSI Education. (2025). AI-driven personalised learning

will become mainstream in 2025. https://www.gsin

education.com/blog/ai-driven-personalised-learning-

will-become-mainstream-in-2025Global Services In

Education

HP. (2024). Tailoring education and personalized learning

with AI. https://www.hp.com/us-en/shop/tech-

takes/personalized-learning-ai-educationBest Buy

IIARD. (2025). Reviewing the role of artificial intelligence

in personalized learning. https://www.iiardjournals.o

rg/get/WJIMT/VOL.%209%20NO.%202%202025/Re

viewing%20the%20Role%20of%20Artificial%2086-

94.pdfiiardjournals.org

Liu, B., Zhang, J., Lin, F., Jia, X., & Peng, M. (2025). One

size doesn't fit all: A personalized conversational

tutoring agent for mathematics instruction. arXiv.

https://arxiv.org/abs/2502.12633arXiv

Maghsudi, S., Lan, A., Xu, J., & van der Schaar, M. (2021).

Personalized education in the AI era: What to expect

next? arXiv. https://arxiv.org/abs/2101.10074arXiv

Naqvi, S. M. F. (2024, July 22). Personalized learning and

AI: Revolutionizing education. Forbes. https://www

.forbes.com/councils/forbestechcouncil/2024/07/22/pe

rsonalized-learning-and-ai-revolutionizing-education/

Forbes+1Global Services In Education+1

Naqvi, S. M. F. (2025, January 27). AI in education:

Personalized learning platforms in 2025. eLearning

Industry. https://elearningindustry.com/ai-in-

education-personalized-learning-platformseLearning

Industry

RAND Corporation. (n.d.). AI, education technology, and

personalized learning. https://www.rand.org/education-

and-labor/focus-areas/education-technology.html

RAND Corporation

ResearchGate. (2025). Personalized learning through AI.

https://www.researchgate.net/publication/376814707_

Personalized_learning_through_AIResearchGate

Tandfonline. (2025). The transformative journey of

artificial intelligence in education. https://www.tand

fonline.com/doi/full/10.1080/1547688X.2025.247581

1Taylor & Francis Online

Tech Times. (2025, January 30). AI revolutionizing the

future of personalized learning. https://www.techtim

es.com/articles/309235/20250130/ai-revolutionizing-

future-personalized-learning.htmTech Times

The Times. (2024). Brave new words by Salman Khan

review - how AI will transform teaching. https://

www.thetimes.co.uk/article/brave-new-words-salman-

khan-review-nd93ts7lwthetimes.co.uk

Time. (2024). Kristen DiCerbo. https://time.com/70128

01/kristen-dicerbo/time.com

U.S. Department of Education, Office of Educational

Technology. (2023). Artificial intelligence and the

future of teaching and learning: Insights and

recommendations. https://www.ed.gov/sites/ed/files/d

ocuments/ai-report/ai-report.pdfU.S. Department of

Education

UNESCO. (n.d.). Artificial intelligence in education.

https://www.unesco.org/en/digital-education/artificial-

intelligenceUNESCO

Vorobyeva, K. I., Belous, S., Savchenko, N. V., Smirnova,

L. M., Nikitina, S. A., & Zhdanov, S. P. (2025).

Personalized learning through AI: Pedagogical

approaches and critical insights. Contemporary

Educational Technology, 17(2), ep574. https://doi.o

rg/10.30935/cedtech/16108cedtech.net

Wang, X. J., Lee, C., & Mutlu, B. (2025). LearnMate:

Enhancing online education with LLM-powered

personalized learning plans and support. arXiv.

https://arxiv.org/abs/2503.13340arXiv

Wikipedia. (2025). Adaptive learning. https://en.wikipe

dia.org/wiki/Adaptive_learningen.wikipedia.org

Wikipedia. (2025). Applications of artificial intelligence.

https://en.wikipedia.org/wiki/Applications_of_artificia

l_intelligenceen.wikipedia.org

Wikipedia. (2025). ChatGPT in education. https://en.wi

kipedia.org/wiki/ChatGPT_in_education

en.wikipedia.org

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

242