Fuzzy Logic for Cybersecurity: Intrusion Detection and Privacy

Preservation with Synthetic Data

Marina Soledad Iantorno

and Khalil Beladda

Campus Founders, Buildung Campus, Heilbronn, Germany

CCT College, Westmorland St, Dublin, Ireland

Keywords: Cybersecurity, Defuzzification, Fuzzy Logic, Intrusion Detection System, NSL-KDD, Membership Function,

Synthetic Data, WGAN.

Abstract: This research explores the use of fuzzy logic in intrusion detection systems (IDS) aiming to improve

cybersecurity threat detection. Conventional machine learning models, like Decision Trees and Support

Vector Machines, are evaluated against a fuzzy logic model that employs triangle and parallelogram-shaped

membership functions to address the uncertainty in network traffic. The fuzzy logic system presented good

performance, achieving greater accuracy, precision, and F1-scores than conventional models, particularly

when using real network traffic data. Synthetic data produced by Wasserstein Generative Adversarial

Networks (WGANs) was also used to evaluate the model's robustness and guarantee privacy protection in

future studies. The relevance of this approach lies in its ability to provide more comprehensive threat

detection, helping organizations safeguard their systems in environments where strict, rule-based models may

fall short. The findings indicate that the fuzzy logic methodology is effective, even when applied to synthetic

data, demonstrating its feasible choice for intrusion detection in sensitive contexts. Subsequent research will

investigate the incorporation of deep learning methodologies and the modification of the model for distributed

systems, focusing on scalability and real-time threat identification.

1 INTRODUCTION

In the contemporary digital era, cybersecurity is

critically significant, as organizations encounter

increasing dangers from different cyberattacks

(CheckPoint, 2024). These attacks, including denial

of-service (DoS), remote-to-local (R2L), and user-to

root (U2R) incursions, can significantly undermine

the integrity, confidentiality, and availability of

essential systems. With the growing dependence of

organizations on digital infrastructures and online

transactions, the security of network environments

has emerged as a critical concern (Admass, Munaye,

& Diro, 2024). To address this, intrusion detection

systems (IDS) are in use to detect anomalous

behaviors within network traffic. This paper aims to

investigate the use of fuzzy logic in an Intrusion

Detection System (IDS) employing the NSL-KDD

dataset (DARPA, 2018), which is a common resource

for intrusion detection studies, to explain how fuzzy

logic can improve the identification of potential cyber

https://orcid.org/0009-0007-2596-3494

threats while protecting sensitive network

information through synthetic data generation.

This analysis has two main goals. Initially, it aims

to replicate network traffic and intrusions with the

NSL-KDD dataset, distinguishing and categorizing

normal traffic from various attack vectors. Secondly,

it proposes a fuzzy logic-based methodology for

intrusion detection, enabling a more effective

modelling and analysis of the inherent ambiguity in

network activity. The analysis uses fuzzy logic to

enhance the precision of identifying inappropriate

network behavior, avoiding dependence on rigid

thresholds or binary determinations. Additionally,

synthetic data modelling will be utilized to produce

realistic, privacy-preserving network traffic, offering

enterprises an efficient method to train IDS models

while safeguarding the confidentiality of sensitive

network data.

In contrast to conventional systems relying on

rigid rules or exact thresholds, fuzzy logic allows

various levels of truth, thus addressing uncertainty

376

Iantorno, M. S. and Beladda, K.

Fuzzy Logic for Cybersecurity: Intrusion Detection and Privacy Preservation with Synthetic Data.

DOI: 10.5220/0013137300003890

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

In Proceedings of the 17th International Conference on Agents and Artiﬁcial Intelligence (ICAART 2025) - Volume 3, pages 376-382

ISBN: 978-989-758-737-5; ISSN: 2184-433X

and ambiguity in network traffic patterns (Masoumi,

Dehghani, Hossani, & Masoumi, 2020).

Cybersecurity incidents are frequently intricate

and unpredictable, complicating the classification of

an occurrence as just "safe" or "malicious". Fuzzy

logic offers another methodology, enabling the

assessment of diverse attack scenarios according to

fluctuating probabilities, hence facilitating more

adaptable and resilient decision-making processes.

As cyber threats become more sophisticated, the

integration of fuzzy logic into Intrusion Detection

Systems can enhance organizations’ ability to

identify cyber-attacks more effectively and react to

changing threat environments (Keeping Security,

2024).

2 LITERATURE REVIEW

Several researchers have explored the application of

fuzzy logic in various fields. Despite being an

established technique created by Lotfi Zadeh in the

1960s, fuzzy logic continues to progress as

technological improvements and better computer

capacity create new opportunities for its use (Perry,

1995). The inherent adaptability of fuzzy logic to

address uncertainty and ambiguity makes it a useful

instrument for solving complex issues in

contemporary industries, including network security,

data privacy, and financial audits. Recent

advancements in machine learning, artificial

intelligence, and big data processing have augmented

the capabilities of fuzzy logic, leading to the creation

of more sophisticated models that can more

efficiently integrate qualitative and quantitative

aspects (Castillo & Melin, 2015). Some of the

relevant advancements are listed below in

chronological order.

• 2022. Bambang Leo Handoko and Daniel

Marcell examine how understanding audit risk,

auditor competency, and fuzzy logic can impact

cybersecurity in auditing. The study pushes agency

theory and employs a quantitative research approach,

using data collected from public accountants in

Indonesia. Their results suggest that an auditor's

ability to assess risks and competency significantly

affects materiality decisions. Moreover, the paper

highlights how fuzzy logic can assist in addressing

qualitative uncertainties in materiality assessments,

such as those related to ambiguity or subjectivity.

Fuzzy logic helps auditors to include quantitative and

qualitative factors when determining the significance

of cybersecurity assessments, particularly helpful for

cyber insurance (Handoko & Marcell, 2022).

• 2021. In 2021 researchers from the National

Technical University of Ukraine proposed a

simulation model for detecting cyberattacks using

fuzzy set theory and fuzzy inference. The model

includes a functional diagram that processes network

traffic telemetry and applies fuzzy logic to detect

various types of cyberattacks, such as denial-of

service, remote-to-local, user-to-root, and probes.

The authors describe the process of fuzzifying the

input parameters—based on linguistic variables and

membership functions—and developing fuzzy

production rules in a knowledge base. The system

was tested with 38 parameters defining network

traffic. Their model showed superior performance in

detecting polymorphic and traditional attacks

compared to other methods, such as neural networks,

showing an average accuracy improvement of 10%.

The results suggest that the fuzzy logic-based

detection system offers a more flexible and effective

solution for improving cybersecurity defenses

(Subach, Fesokha, Mykytiuk, Kubrak, & Korotayev,

2021).

• 2018. In the paper "Improving risk assessment

model of cybersecurity using fuzzy logic inference

system" written by researchers from King Saud

University in Saudi Arabia and Fordham University

in the United States, the authors propose an approach

to cybersecurity risk assessment using a Fuzzy

Inference System (FIS) based on the Mamdani model.

The paper addresses the growing threats of

cyberattacks such as DoS, DDoS, malware, and

phishing, which needed more advanced risk

assessment models. Traditional methods of risk

analysis are often limited by the uncertainty and

vagueness inherent in assessing cyber risks. To

overcome these limitations, the authors suggest

incorporating a triangular fuzzy logic, which allows

for more flexible and approximate reasoning. Their

evaluation shows that the fuzzy-based model can

effectively assess risks with a higher degree of

accuracy and adaptability, making it a valuable tool

in mitigating cyber threats in organizations (Alali,

Almogren, Hassan, Rassan, & Bhuiyan, 2018).

These studies highlight how fuzzy logic can be

used to improve processes by incorporating

uncertainty and vagueness into the analysis. This

section has reviewed some key works that have

applied fuzzy logic in cybersecurity, risk assessment,

and materiality considerations, providing a

foundation for understanding how this technique can

be further refined and applied in contemporary

contexts.

In line with the advancements mentioned above,

the authors of this paper aim to use fuzzy logic to

Fuzzy Logic for Cybersecurity: Intrusion Detection and Privacy Preservation with Synthetic Data

377

address challenges in cybersecurity, specifically in

the context of intrusion detection and data privacy.

During their research, they identified several relevant

studies that have applied fuzzy logic to similar areas.

3 METHODOLOGY

3.1 Data Gathering

This research uses the NSL-KDD Dataset, an updated

version of the KDD Cup 1999 dataset, which is

globally recognized for its application in network

intrusion detection. The dataset contains labelled

instances of both normal network activity and various

types of attacks, including DoS, R2L, and U2R

attacks. The data holds 41 features, and provides a

comprehensive set of attributes, such as protocol type,

service, and network bytes, which are instrumental in

identifying potential intrusions. The dataset was

chosen due to its extensive use in cybersecurity

research, making it ideal for testing and comparing

different machine learning and fuzzy logic

approaches to intrusion detection.

3.2 Data Preparation

Before modelling, data preprocessing was necessary

to ensure that the dataset was ready for analysis. The

following steps were in place:

• Data cleaning: Removing any duplicates and

handling missing values across the dataset.

• Feature encoding: Converting categorical

variables such as protocol type and network service

into numerical values suitable for machine learning

algorithms.

• Normalization: Normalizing the features to

bring all the variables to a comparable scale.

Additionally, to preserve data privacy, Wasserstein

Generative Adversarial Networks (WGANs) were

employed to generate synthetic data that mirrors the

patterns found in the original NSLKDD data. This

synthetic data was added to test the model’s robustness

and privacy-preserving capabilities, especially in the

context of cybersecurity systems. Words like “is”,

“or”, “then”, etc. should not be capitalized unless they

are the first word of the title.

3.3 Data Modelling

Several models were created for this study to simulate

intrusion detection.

Fuzzy Logic Model: Unlike the research

mentioned in chapter II, which predominantly uses

basic fuzzy logic functions, this research adopts

innovative membership functions, including

triangular and parallelogram-shaped membership

functions, to classify network intrusions. These

membership functions provide more comprehensive

risk evaluations by correlating network

characteristics to different levels of risk (e.g., low,

medium, high) in a flexible, non-linear manner. By

introducing these specific models, the current

research aims to enhance the system’s ability to adapt

to the complexities of modern cyberattacks and

provide more precise risk assessments.

Synthetic Data Generation: Using the WGAN

model, synthetic data imitating the NSL-KDD dataset

was generated. This synthetic data was then

integrated with real data to ensure that the intrusion

detection system could be tested under realistic

conditions while safeguarding sensitive information.

Comparison Models: Several traditional machine

learning models were also implemented for

comparison, including the Decision Tree Classifier,

Support Vector Classifier, and Neural Networks.

These models were trained on the NSL-KDD dataset

and evaluated for performance using accuracy,

precision, recall, and F1-score metrics. This allowed

for a detailed comparison between conventional

models and the proposed fuzzy logic-based system,

highlighting the strengths and weaknesses of each

model.

3.4 A Detailed Explanation of the Main

Model

Fuzzy Logic is a mathematical methodology used to

handle uncertainty and imprecision, which is useful

in fields such as cybersecurity, where data is often

complex and ambiguous. In fuzzy logic, data is

transformed into values that range between 0 and 1,

offering a gradient of membership for various

categories or conditions, rather than the binary

true/false (1/0). This system enables a more flexible

way of interpreting and analyzing data, making it

ideal for applications such as intrusion detection,

where network traffic behavior may not fit neatly into

predefined categories (Medium, 2023).

The features of the Fuzzy Logic membership

functions are defined as follows:

• Core: The core of a membership function for any

fuzzy set denotes the area within the universe

characterized by total membership in the set. This

indicates that components within the core are

regarded as entirely belonging to the set

(GeeksForGeeks, 2018). In a fuzzy logic model for

identifying cyberattacks, if a network's traffic pattern

ICAART 2025 - 17th International Conference on Agents and Artiﬁcial Intelligence

378

precisely aligns with the signature of a known attack,

it receives a membership value of 1, indicating it is

entirely classed as a prospective attack.

• Support: The support of a membership function

specifies the particular area in the universe where the

membership exceeds zero. This includes elements

that are partially included in the entire set

(TutorialsPoint, 2019). In cybersecurity, the support

may include network traffic that displays some, but

not all, attributes of an attack. This traffic would have

a membership value ranging from 0 to 1, meaning a

partial match.

• Boundary: The boundary represents the area

where membership in the set is non-zero yet

incomplete. Elements in the boundary region are

ambiguous, exhibiting certain characteristics

indicative of belonging to the set (e.g., potential

signals of an attack), and they cannot be definitively

identified. This component of fuzzy logic is essential

for addressing the ambiguous situations in intrusion

detection, where an action may not distinctly be

classified as malicious or benign (San José State

University, 2023).

Figure 1: Desirable scale for topic classification.

Fuzzy Logic operates based on probability theory

and statistics, although it does not necessitate a strict

mathematical model as seen in conventional methods

(Gaines, 1978). As it is possible to see in Figure 1, the

model will display a series of probabilities of

likelihood to find an anomaly behavior. It employs

rules and membership functions to assess the extent

to which an input is classified inside a particular

category, such as "normal" or "intrusion" in the

context of cybersecurity. In contrast to conventional

classification models that depend on rigid borders,

fuzzy logic facilitates a more flexible categorization

by assigning items a membership value on a scale

ranging from 0 to 1, indicating the probability of an

event belonging to a specific category (Ariff, Ariff,

Sheikh, & Hussin, 2018).

Table 1: Definition of Fuzzy Rules.

Feature Membership

Function

Fuzzy set

Protocol

Type

Triangular Low/ Medium/High

Service Parallelogram Normal/Anomalous

Duration Triangular Short/Medium/Long

The fuzzy logic model uses triangular and

parallelogram membership functions to classify

network traffic into risk categories. Unlike previous

work, these shapes provide granular evaluations of

ambiguous traffic patterns. The fuzzy rules and sets

are explicitly defined in Table 1, ensuring

reproducibility.

By using fuzzy logic, this research aims to provide

a flexible, innovative system for detecting intrusions,

allowing cybersecurity systems to handle the inherent

uncertainty in network behavior more effectively.

Words like “is”, “or”, “then”, etc. should not be

capitalized unless they are the first word of the title.

Figure 2: Examples of Fuzzy Logic Membership.

With a clear understanding of how fuzzy logic is

applied to handle uncertainties in network intrusion

detection, the following section will outline the

practical implementation of this methodology.

4 IMPLEMENTATION

The principal dataset utilized was the NSL-KDD

dataset, comprising labelled network traffic records

for both normal and intrusive behaviors. The

implementation phases were categorized into data

processing, fuzzy logic system design, and synthetic

data production.

The dataset was exposed to an initial

preprocessing, which involved normalizing

numerical features and transforming categorical

variables into numerical formats. Missing data points

were addressed through interpolation or removal.

Subsequently, feature selection was conducted to

Fuzzy Logic for Cybersecurity: Intrusion Detection and Privacy Preservation with Synthetic Data

379

ascertain the most relevant features for intrusion

detection.

The fuzzy logic system was then built using the

SciKit-Fuzzy library in Python. Membership

functions for the relevant features (e.g., duration,

protocol type, service) were created, and triangular as

well as parallelogram-shaped membership functions

were employed to represent the uncertainty in

network traffic behavior (Hamarsheh, 2019). Fuzzy

rules were defined based on these membership

functions to evaluate the likelihood of network traffic

being classified as normal or malicious. These rules

were based on the characteristics of the network

traffic in the NSL-KDD dataset. Once the fuzzy

inference system was defined, a defuzzification

process was implemented using the centroid method

to convert the fuzzy output into a confident decision,

determining whether the traffic should be labelled as

normal or an intrusion (Science Direct, 2001). The

performance of this system was compared to other

traditional machine learning models, such as the

Decision Tree Classifier and Support Vector

Machine, to evaluate the effectiveness of the fuzzy

logic model.

For the purpose of data privacy preservation in

future cybersecurity training and testing modelling,

the synthetic data created was trained to mimic the

patterns found in the original NSL-KDD dataset,

ensuring that the models could be tested without

exposing sensitive information. The generated

synthetic data was then fed into the fuzzy logic model

and the machine learning models to assess the

robustness and accuracy of the system when using

synthetic data instead of real data. Finally, the results

from the fuzzy logic model and the machine learning

models were evaluated using common metrics such

as accuracy, precision, recall, and F1-score

(Kupchyn, et al., 2022).

The following section will show and discuss the

results obtained in the process mentioned above.

5 RESULTS

The table below shows the results obtained in the

implementation process.

The results presented in Table 1 show the

efficiency of fuzzy logic in intrusion detection. The

model report high precision and F1-scores compared

to conventional models. The precision and recall

metrics also indicate that the fuzzy logic model

provides a more balanced performance in detecting

true positives and minimizing false positives,

particularly in scenarios with real network traffic

data. Moreover, the fuzzy logic model shows

competitive performance with synthetic data,

highlighting its robustness and adaptability in

managing cybersecurity tasks while ensuring data

privacy. This, shows once again, that synthetic data

could be a possible solution against data privacy

scenarios (Riemann, 2024).

Table 2: Comparison Performance.

Model Accuracy

(%)

Precision (%) Recall (%) F1 - Score

(%)

Fuzzy Logic

(Real

Data)

92.4 93.1 91.8 92.4

Fuzzy Logic

(Synthec

Data)

90.1 90.7 89.6 87.6

Decision Tree

(Real Data)

87.6 88.3 86.9 84.2

Decision Tree

(Synthec

Data)

84.2 84.7 83.9 89.6

SVM (Real

Data)

89.7 90.2 89.1 86.3

6 CONCLUSIONS

This study has shown the use of fuzzy logic in

cybersecurity, specifically in intrusion detection. The

fuzzy logic model, employing the NSL-KDD dataset,

outperformed conventional machine learning models

like Decision Trees and Support Vector Machines in

accuracy, precision, and F1-score, especially with

real network traffic data. The implementation of

triangle and parallelogram-shaped membership

functions facilitated a more detailed assessment of

network traffic, providing the system with the

capability to categorize ambiguous or uncertain

behavioral patterns. This is a significant benefit in

cybersecurity, as attacks often show complexity and

may not adhere to strict rules or unique parameters

(Javaheri, Gorgin, Lee, & Masdari, 2023).

In a similar vein, the integration of synthetic data

produced by the WGAN technique provided

additional insights into the possibility of using

artificial data to safeguard privacy while preserving

detection efficacy. This is an ongoing discussion at

the moment in industry and academia, and this

research shows that the fuzzy logic model maintained

competitive performance when trained on synthetic

data, which indicates that privacy-preserving

techniques can be effectively incorporated without

substantially compromising model accuracy

ICAART 2025 - 17th International Conference on Agents and Artiﬁcial Intelligence

380

(International Association of Privacy

Professionals, 2023). This finding is particularly

significant in domains involving sensitive data, such

as network security.

Further studies conducted by the authors of this

research may investigate the augmentation of this

model by integrating more sophisticated membership

functions or combining fuzzy logic with additional

machine learning methodologies to further improve

performance. Furthermore, using this methodology to

other cybersecurity datasets, such as real-time

network traffic data, may lead to a deeper

understanding of the system's performance in

dynamic contexts (Pancardo, Hernandez-Nolasco,

Wister, & Garcia-Constantino, 2021). Moreover,

adding explainable AI methodologies to clarify the

reasoning behind the fuzzy logic model's conclusions

could also enhance transparency and fostering trust

among cybersecurity professionals using this system

(Cao, et al., 2024).

The fuzzy logic model could also benefit from

the integration of deep learning techniques to enhance

its ability to detect more sophisticated and emerging

threats, such as zero-day attacks (Han, 2024). This is

an are that has not yet been explored.

Another promising direction for future work

could be to adapt the model for use in distributed or

cloud-based environments, where cybersecurity

challenges differ due to the decentralized nature of

these systems (Prasath, Bharathan, Lakshmi, &

Nathiya, 2023). This would involve testing the

model's scalability and resilience in managing

largescale network data. Exploring real-time

deployment and optimization for faster threat

detection could also lead to practical implementations

in live cybersecurity systems, further proving the

value of fuzzy logic in modern threat landscapes.

REFERENCES

A. Kupchyn, Komarov, V., I. Borokhvostov, M. Bilokur,

A. Kuprinenko, Y. Mishchenko, V. Bohdanovych, & O.

Kononov. (2022). Determining the Accuracy for Fuzzy

Logic Technology Foresight Model. Cybernetics and

Systems Analysis, 58(3), 382–391. https://doi.org/10.

1007/s10559-022-00470-1

Admass, W. S., Munaye, Y. Y., & Diro, A. A. (2024).

Cyber security: State of the art, Challenges and Future

Directions. Cyber Security and Applications, 2, 100031.

https://doi.org/10.1016/j.csa.2023.100031

Alali, M., Almogren, A., Hassan, M. M., Rassan, I. A. L.,

& Bhuiyan, M. Z. A. (2018). Improving risk assessment

model of cyber security using fuzzy logic inference

system. Computers & Security, 74(74), 323–

339. https://doi.org/10.1016/j.cose.2017.09.011 Ariff, A.,

Ariff, A., Sheikh, S., & Hussin, M. (2018). ConBEE

Green envelope as an architectural strategy for energy

efficiency in a library building. MATEC Web of

Conferences. https://doi.org/10.1051/matecconf/2019

Bambang Leo Handoko, & Marcell, D. (2022). The Impact

of Understanding Audit Risk, Auditor’s Competency,

and Fuzzy Logic Analysis to Materiality Level

Consideration. ICEME ’22: Proceedings of the 2022

13th International Conference on E-Business,

Management and Economics, 211, 500–506.

https://doi.org/10.

1145/3556089.3556107

CAO, J., Zhou, T., Zhi, S., Lam, S., REN, G., ZHANG, Y.,

Wang, Y., Dong, Y., & Cai, J. (2024). Fuzzy Inference

System with Interpretable Fuzzy Rules: Advancing

Explainable Artificial Intelligence for Disease

Diagnosis—A Comprehensive Review. Information

Sciences, 120212–120212. https://doi.org/10.1016/j.

ins.2024.120212

Castillo, O., & Melin, P. (2014). Fuzzy Logic

Augmentation of Nature-Inspired Optimization

Metaheuristics. In Studies in computational

intelligence. Springer Nature. https://doi.org/10.

1007/978-3-319-10960-2

CheckPoint. (2024, July 16). Check Point Research Reports

Highest Increase of Global Cyber Attacks seen in last

two years – a 30% Increase in Q2 2024 Global

Cyber Attacks. Check Point Blog. https://blog.

checkpoint.com/research/check-pointresearch-reports-

highest-increase-of-global-cyberattacks-seen-in-last-

two-years-a-30-increase-in-q2-2024-global-cyber-attacks/

DARPA. (2018). NSL-KDD. Www.kaggle.com. https://

www.kaggle.com/datasets/hassan06/nslkdd

Gaines, B. R. (1978). Fuzzy and probability uncertainty

logics. Information and Control, 38(2), 154–169.

https://doi.org/10.1016/s0019-9958(78)90165-1

GeeksforGeeks. (2018, April 10). Fuzzy Logic |

Introduction. GeeksforGeeks. https://www.

geeksforgeeks.org/fuzzy-logicintroduction/

Hamarsheh, Q. (2019). Different Types of Membership

Functions. https://www.philadelphia.edu.jo/academics/

qhamarsheh/uploads/Lecture%2018_Different%20Typ

es%20of%20Membership%20Functions%201.pdf

Han, X. (2024). Analyzing the impact of deep learning

algorithms and fuzzy logic approach for remote English

translation. Scientific Reports, 14(1). https://doi.org/10.

1038/s41598-024-64831-w

International Association of Privacy Professionals. (2023,

December 12). IAPP. Iapp.org.

https://iapp.org/news/a/synthetic-data-whatoperational-

privacy-professionals-need-to-know

Javaheri, D., Gorgin, S., Lee, J.-A., & Masdari, M. (2023).

Fuzzy logic-based DDoS attacks and network traffic

anomaly detection methods: Classification, overview,

and future perspectives. Information Sciences, 626,

315–338. https://doi.org/10.1016/j.ins.2023.01.067

Masoumi, M., Dehghani, F., Hossani, S., & Masoumi, A.

(2020, May). (PDF) THE CHALLENGES AND

ADVANTAGES OF FUZZY SYSTEMS

Fuzzy Logic for Cybersecurity: Intrusion Detection and Privacy Preservation with Synthetic Data

381

APPLICATIONS A PREPRINT. ResearchGate.

https://www.researchgate.net/publication/341426529_

THE_CHALLENGES_AND_ADVANTAGES_OF_

FUZZY_SYSTEMS_APPLICATIONS_A_PREPRINT

Medium. (2023, July 15). Exploring GANs to Generate

Synthetic Data. The Research Nest. https://medium.com/

the-research-nest/exploring-gansto-generate-synthetic-

data-ca48f8a4b518

Pancardo, P., Hernandez-Nolasco, J. A., Wister, M. A., &

Garcia-Constantino, M. (2021). Dynamic Membership

Functions for Context-Based Fuzzy Systems. IEEE

Access, 9, 29665–29676. https://doi.org/10.1109/

access.2021.3058943

Perry, T. (1995, June 1). Lotfi Zadeh and the Birth of Fuzzy

Logic - IEEE Spectrum. Spectrum.ieee.org.

https://spectrum.ieee.org/lotfi-zadeh Prasath, V., Bharathan,

N., Lakshmi, N., & Nathiya, M. (2023). Fuzzy Logic In

Cloud Computing. https://www.ijert.org/research/fuzzy-

logic-in-cloudcomputing-IJERTV2IS3439.pdf

Riemann, R. (2024, May 9). Synthetic Data | European Data

Protection Supervisor. Www.edps.europa.eu.

https://www.edps.europa.eu/presspublications/publicat

ions/techsonar/synthetic-data_en San José State

University. (2023, January). Fuzzy Logic: The Logic of

Fuzzy Sets. Sjsu.edu. https://www.sjsu.edu/faculty/

watkins/fuzzysets.htm

Science Direct. (2001). Defuzzification - an overview |

ScienceDirect Topics. Www.sciencedirect.com.

https://www.sciencedirect.com/topics/engineering/def

uzzification

Security, K. (2024, March 26). Cyber Attacks Are More

Sophisticated Than Ever, With AI-Powered Attacks

Posing the Greatest Risk. Www.prnewswire.

com. https://www.prnewswire.com/news-releases/

cyberattacks-are-more-sophisticated-than-ever-with-

aipowered-attacks-posing-the-greatest-risk-302098797.

html

Subach, I., Fesokha, V., Mykytiuk, A., Kubrak, V., &

Korotayev, S. (2021). Simulation Model of a Fuzzy

Cyber Attack Detection System. https://ceurws.

org/Vol-3241/paper9.pdf

TutorialsPoint. (2019). Fuzzy Logic - Membership

Function - Tutorialspoint. Www.tutorialspoint.com.

https://www.tutorialspoint.com/fuzzy_logic/fuzzy_log

ic_membership_function.htm.

ICAART 2025 - 17th International Conference on Agents and Artiﬁcial Intelligence

382