=Paper=
{{Paper
|id=Vol-3333/paper8
|storemode=property
|title= Fuzzy Logic for Emergence Verification
|pdfUrl=https://ceur-ws.org/Vol-3333/Paper8.pdf
|volume=Vol-3333
|authors=Ali Boukehila
|dblpUrl=https://dblp.org/rec/conf/tacc/Boukehil22
}}
== Fuzzy Logic for Emergence Verification ==
https://ceur-ws.org/Vol-3333/Paper8.pdf
Fuzzy Logic for Emergence Verification
Ali Boukehila 1,2
1
LISCO Laboratory, Badji Mokhtar Annaba University, Annaba 23000, Algeria
2
Amar Telidji Laghouat University, Laghouat 03000, Algeria
Abstract
Actions of terrorist organizations, flu viruses or natural disasters can be considered emergent
behaviours in our environment. In recent decades, emergent phenomena have been the subject
of multiple research efforts in the field of complex adaptive systems, however, it is still hard
to predict, track and supervise such phenomena. This highlights the urgency to better
understand the dynamics of these behaviours in order to timely detect critical phase transitions
that might form a risk for software or human environments. This paper introduces an emergent
verification system that integrates a data retriever from agent-based simulations and a
verification module based on fuzzy classification. We follow the classification of emergent
behaviours according to Fromm’s taxonomy. In addition, the paper presents a scenario
implementation using swarms of birds (Boids model) to demonstrate the applicability of the
proposed approach. The results show that the framework is able to verify weak emergence
occurred during simulations. Since this work is a part of ongoing research, the future direction
is also discussed.
Keywords 1
Complex adaptive systems, Emergence, Agent-based simulation, Fuzzy rule-based
classification, Swarming.
1. Introduction
Computer systems have grown extremely complex with the expansion of storage and information
processing technologies as well as the evolution of networks. As a result, models and approaches for
decentralized and large-scale application have been developed. It facilitates the creation of multi-agent
systems and other models and simulation tools. The versatility of agent-based simulation (ABS) enables
for the investigation of complicated systems' behavior. In this study, we look at how agents interact in
order to identify emergence. Emergent behaviors may be observed practically anywhere in real life, yet
their research is limited since they are difficult to identify [1]. Interactions between system components
are unquestionably important in the formation of such events. The system may be described with Agent-
Based Simulation (ABS), where agents represent these components, due to the interaction among the
components and their capacity to decide separately following a given logic [2]. The goal of this study
is to fully use emergence's potential. Variable-based [2][3][4] and event-based [4] research efforts are
primarily separated into two categories. The variable-based technique measures emergence
quantitatively using a specified variable, such as the mass center of an animal population, and detects
emergence by assessing changes in that variable. Applying variable-based approaches in continuous
uncertain systems is tough, and overcoming the computing cost, as well as the necessity for ongoing
human involvement, are additional challenges. The event-based approach focuses on system state
changes [5], and emergent behavior is regarded as a result of events that cause shifting points, either at
the global level (Marco) or at the level of system components (Micro). One of the most common
methods for identifying emergence is to look for changing points in the system; consequently, in this
research, we employ an inflexion predictor to do so. Emergence is a global state in a system.
Tunisian Algerian Conference on Applied Computing (TACC 2022), December 13 - 14, Constantine, Algeria
Envelope-Open al.boukehila@lagh-univ.dz ORCID: 0000-0002-5011-8838
© 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
Data mining is the act of sorting through enormous data sets to find relationships, forecast outcomes,
and solve issues. The suggested solution leverages big data using data mining with an inflection
predictor. Data mining employs a variety of sophisticated and clever approaches, including
classification, clustering, time series analysis, and so on. To detect the increase of emergence, we used
interactions as metrics, and the technique employs a fuzzy classification with simulation data as inputs.
Our technique is designed and tested using an ABS model called the Boids model [6].
This paper is set up as follows. In the next section, theoretical foundations are described. Sec. III
introduces the related works that inspired this study, the proposed approach will be presented in Sec.
IV, Sec. V shows experiments conducted and their analysis. Finally, in Sec. VI some concluding
remarks and future work lines are presented.
2. Theoretical foundations
2.1. Emergence in complex systems
Complex systems have properties that are difficult to predict by studying the behaviour of their parts
[7]. Exchanges in human societies, as well as the flocking in a group of birds, are emerging behaviours
[6]. Emerging phenomena can be beneficial, for example, if the new unknown proprieties are considered
as "positive" or "useful", these behaviours can otherwise be "negative" or "dangerous". Identifying these
properties can prevent potential danger.
Many attempts to define the meaning of emergence have been documented. Originating from
philosophy [3], emergence became useful in ABS especially for studying complex systems. Emergence
provides a great opportunity for understanding interactions in a complex environment [4]. In this work,
we follow Fromm's [8] classification. He proposed a taxonomy that classifies emergence based on
feedbacks and causality.
Table 1
Fromm's emergence classification [8]
Type Name Roles Frequency Predictability System
I Nominal or fixed abundant predictable closed, with
Intentional passive entities
II Weak Flexible frequent predictable in open, with
principle active entities
III Multiple Fluctuating common - not open, with
unusual predictable multiple levels
(or chaotic)
IV Strong New World Rare Not New or many
of roles predictable in systems
principle
In this work, we only consider type II (weak emergence) and type IV (Strong emergence), Strong
emergence is the notion of emergence that is most common in philosophical literature about emergence,
and is the notion invoked by the British emergentists of the 1920s. Weak emergence is the notion of
emergence that is most common in scientific discussions of emergence, and is the notion that is typically
invoked by proponents of emergence in complex systems theory. Weak emergence describes new
properties arising in systems as a result of the interactions at a micro level. However, Bedau says that
the properties can be determined only by computer simulation.
Strong emergence describes the direct causal action of a macro-level system upon its components;
qualities produced are irreducible to the system's constituent parts [23]. The whole is not equal to the
sum of its parts [9].
2.2. The Boids Model
Craig Reynolds [6] created Boids, which is an artificial life simulation. The simulation's goal is to
mimic the behavior of flocks of birds. The Boids simulation, on the other hand, instead of directing the
interactions of a complete flock, merely defines the behavior of each individual bird. The program
generates a result that is sophisticated and realistic enough to be utilized as a framework for computer
graphics applications such as computer-generated behavioral animation in motion picture films using
only a few simple principles, Figure 1: Boids’ rules shows Boids’ rules.
Figure 1: Boids’ rules
The three rules of the Boids model are:
1. Separation: Collision avoidance.
2. Alignment: Heading in the same direction as the near neighbors.
3. Cohesion: Each bird flock with near neighbors to form groups.
Within the simulation, emergent properties in the Boids model could appear. When a packing
behavior occurs, emergent behavior is verified. To make visual recognition of emergence easier, we
chose to analyze a small number of agents. The goal of this paper is to understand these unexpected
behaviours.
2.3. Fuzzy classification
Engineering fuzzy systems, based on fuzzy logic proposed by Zadeh [24] [25], date back to the
1970s, when Mamdami [26], [27] developed the first fuzzy controller. Fuzzy systems have now
acquired popularity in a variety of domains, including control and automation, pattern recognition,
medical diagnosis, and forecasting. Because fuzzy systems are frequently considered as black boxes,
an analytical theory for fuzzy systems is required to resolve misunderstandings and disagreements.
Investigation and optimization of developed fuzzy models, as well as comparison study of different
methodologies, play an essential role [28], [29], [30] The objective of this work is to build a fuzzy rule-
based system (FRS) to verify the raise of emergent proprieties in an ABS, results. Due to the complexity
of the study, several techniques to automatically generate FRS were used i.e., ad-hoc data driven models
and genetic fuzzy systems (GFS). The first technique is based on learning from the input-output dataset
resulting in the simulation, the second group processes learning of a fuzzy system, precisely parameters
of that system, as an optimization problem and uses a genetic algorithm (GA) to proceed that task. The
system modeling process is divided into three parts. First step encloses a data retrieval from the ABS,
the second phase is the emergence classification via the fuzzy system.
3. Related Works
The fast-growing complexity of modern systems is challenging for the research community, and the
need for techniques that can cope with these types of systems is crucial in numerous domains, including
communication, learning, industry, and engineering. Agent-based simulation (ABS) is the most widely
used simulation approach because it is versatile and capable of simulating large, interconnected
systems. (ABS) techniques are commonly used to explore emergent behavior as a result of simulation.
Emergence is prevalent in systems with a large number of pieces, and it is usual to see and manipulate
the emergence of unanticipated behaviors using ABS simulation platforms.
It is critical to provide the theoretical background for constructing and modeling the approach by
initiating the emergence type and taxonomy of the emergent behaviors we are interested in. There are
many other emergence definitions, however for the sake of this study, we simulated the procedure using
Bedau's [9] weak emergence definition. A weak emergence may be proven by simulation, and it is
predicted and perhaps regulated in particular systems. The modeling of flocking in the Boids model,
for example, is controlled by three rules: separation, cohesion, and alignment. The application of these
criteria will result in grouping behavior (emergence). Experts may either increase or eliminate flocking
by changing a few settings.
To define the technique after identifying the emergence’s type we're interested in, we used Fromm's
type II class of emergence [10]. Fromm provided a taxonomy that categorizes four types of emergence
based on distinct feedback patterns. Simple feedback (Negative or Positive) is the major characteristic
of (Type II), which is defined as a top-down interaction from the macro to micro level. Positive
feedbacks are preventative orders that ensure the system does not diverge into detrimental behavior,
according to Fromm. Negative feedbacks are restricting to the behaviors of the agents (e.g. swarm
intelligence) (e.g. Avoiding financial bubbles), To exemplify this class, in the Boids model, the
Separation rule is deemed Positive if the expert does not want the birds to flock; on the other hand,
removing this rule will cause the birds to swarm quickly.
Researchers have been attempting to measure emergence for a long time. As an example. [11]
described emergent behavior as state-changing points and stated that algorithms may be effective in
verifying emergent behavior. [12] employed interaction statistics as a measure to examine the
emergence of emergent behaviors using Agent-based simulation (ABS). [13,14] proposed an ontology-
based system for semantically validating emergence, which employed a semantic state distance metric
to quantify semantic differences between component attribute values. [15] provides a collection of
metrics-based strategies for analyzing vision-based vehicle behavior. [16] uses an age metric to identify
and characterize emergence utilizing swarms of Unmanned Aerial Vehicles (UAVs). [17] proposed a
statistic meter for detecting emergence and demonstrated how communication in disputed contexts is
impacted.
Traditional models (mathematical, statistical) when the data is represented by equations to create the
model, see Niharika et al., [18], have been used to detect shifting points in weather forecasting. Big data
and learning-based models are being used in recent methods. Read [19] for further information on Big
Data and Learning Models. Yu Zheng [20] employed an inflection predictor in his model, and Yu Zheng
used an inflection module to capture unexpected changes in air grade in a noteworthy paper. In general,
emergence verification research has progressed, however, there is still a lack in using interaction as
metrics for the emergence detection. Although it is not debatable that interaction is a critical aspect in
emergence, few existing approaches have addressed the analysis of massive simulation data using well-
known statistical techniques. to deal with the mentioned limitations, we present our approach in the
next section.
4. Proposed Approach
In this section we present the multi-agent simulation framework (Figure 2) that consists of two
components, agent-based simulation engine and fuzzy-based classification engine. These two
components communicate with each other to implement their functionality. The main functionality of
the simulation engine during the simulation is to retrieve data which will be passed to the fuzzy system
to verify emergence.
Figure 2: Proposed framework showing fuzzy and simulation-based engines.
4.1. Simulation engine
The simulation system is based on Boids model which simulates a flight of birds. Emergence is
detected when a flocking situation take place, as mentioned in section II, in Boids model, an emergence
behaviour i.e (packing), happens every time. Simulations were conducted using NetLogo [33], which
is a well-known ABS plat-from. Figure 3 presents a simulation in which, a normal behaviour i.e., no
packing behaviour, only the Separation rule is on. The agent number is 50, Ticks = 300 steps.
Figure 3: Simulation of 50 agents in Boids.
Figure 4 shows an emergence behaviour example, multiple groups of birds are visible, in this case,
the simulation data is retrieved and passed for the verification module.
Figure 4: Emergent behaviour
The emergence behaviour in the Boid model is the presence of an unexpected grouping or packing
behaviour, this behaviour is often observed in this model. This is the reason why the boid model is one
of the most used ABS model to study weak emergence [14][2][35][36][37].
4.2. Data retriever
Figure 4 shows several grouping behaviours. During the simulation, data is retrieved automatically
into an excel file using a Netlogo Spreadsheet extension, (see Figure 5). The dataset represents the
agents’ interactions, in the context of Boids model, there are three type of interaction, all physical, no
messages.
• Cohesion: Each time an agent does an action of cohesion with a neighbor
• Alignment: Each time an agent does an action of alignment with a neighbor
• Separation: Each time an agent does an action of separation to avoid collision with a neighbor
Every time an interaction is detected, a proper counter is incremented.
Figure 5: Data retrieval process
Spreadsheet extension does work only in Netlogo version 5, so we simulate the Boid model in this
version of Netlogo.
Figure 6: Simulation data extraction
In order to get a variety of simulation data, we have run several scenarios using 5 agents, Boids'
three rules are Separation (S), Alignment (A), and Cohesion (C), the results are: Number of agents: 5
and number of steps = 1000. We notice that, for a regular interaction triplet (S, A, C), we always get:
Number of A = Number of C.
We use the notation:
• Groups: Multiple packs of birds in the simulation.
• None: No packing behaviour, no groups.
• Full: All birds gathering in one pack.
The following presents some of the simulation results: Results with agent population = 5.
Table 2
Fromm's emergence classification [8]
Interaction Steps Separation Alignment Cohesion Emergence
I= (S, A, C) 1000 300 1054 1054 Groups
I = (S, A, C) at 3000 5234 6816 6816 Full
step 3000
I = (S, A, C, A) 1000 14 6876 3438 Full
I= (S, C) 1000 16 0 0 None
I= (S, A) 1000 12 2893 0 Full
I= (S, A, A) 1000 0 8142 0 Full
I= (S, C, C) 1000 4 0 128 None
4.3. Verification module
In this work, the fuzzy system is a rule-based classification system (FRBCS) using Chi’s method.
This method was introduced by Chi et al. (1996), which is an extension of Wang and Mendel’s method,
for treating classification problems. The Knowledge Base: it is composed of both the Rule Base (RB)
and the Data Base, where the rules and the membership functions are stored respectively.
To build the rule base, we use an extension of the specification given by Singh [5],
TypeII ≡ EmergentBehavior ⊓ ∃hasParticipant.(System ⊓ sendNegativeFeedbackTo.Component).
We transform every rule into the following fuzzy rule: Rule Rj
𝑖𝑓 𝑥1 𝑖𝑠 𝐴𝑗1 𝑎𝑛𝑑 … 𝑎𝑛𝑑 𝑥𝑛 𝑖𝑠 𝐴𝑗𝑛 𝑡ℎ𝑒𝑛 𝐶𝑙𝑎𝑠𝑠 = 𝐶𝑗 𝑤𝑖ℎ 𝑅𝑊𝑗 (1)
∑𝑥𝑝 ∈𝐶𝑙𝑎𝑠𝑠 𝐶𝑗 𝑈𝐴𝑗(𝑋 )
𝑝
𝑅𝑊𝑗 = 𝐶𝐹 = 𝑝
∑𝑝=1 𝑈𝐴𝑗(𝑋 )
(2)
𝑝
We obtain: IF E=1 and NfB = 1 THEN C=1 when:
• E is the emergent beaviour, equal to 1 i.e verifier (in Boids thats always the case)
• NfB stands for Neative Feed Back (Fromm's condition for class number 1 to be verified)
• C is the fuzzy class, in FBRCS, it is the consequence part of the rules, in this case for this rule,
C=1 so type II emergence is verified (weak emergence).
The membership function contains three fuzzy sets for each interaction (Cohesion, Separation,
Alignment) variable (Low, Medium, High).
Figure 6: Membership function
Sample of the membership function is given below:
For simplifying the calculation,
𝑥 = 𝑥/10 (3)
1, 𝑥 < 30
𝑢𝑙𝑜𝑤(𝑥) = { (40−𝑥) (4)
, 30 ≤ 𝑥 < 40
10
Table 3
Sample fuzzy rule
If Then
Rule number Cohesion Separation Alignment Emergence Type
1 L H L N /
2 H L H Y 1
3 L M L N /
4 H M H Y 1
5 L H M N /
(N= None) for emergence, for the type, as mentioned before, the result will be 1 for weak emergence
and 2 for Strong emergence. After learning fuzzy if-then rules by training patterns, different weights
and constants values were used. Different weights were assigned to the rules to decrease correct
classification rates. The two classes get the highest classification rate for wp=0.25. Table 4. presents
the results of the correct classification.
Table 4
Sample of preliminary Results of correct classification rates with different weighting rules.
𝜂 𝑤𝑝 for all classes Cost(𝑆) Classification rate
0.1 𝑤 𝑝 = 0.25 for all lasses 8 91.3%
0.3 𝑤𝑝= 0.25 for all classes 9.3 90.4%
0.1 𝑥𝑝 ∈ class 1 𝑤𝑝= 0.5, 11 89.1%
𝑥𝑝∈ class 2 𝑤𝑝= 0.25
0.3 𝑥𝑝 ∈ class 1 𝑤𝑝= 0.5, 15 80.3%
𝑥𝑝∈ class 2 𝑤𝑝= 0.25
5. Conclusion
In recent years, emergence has become an important research focus. Emergence can be positive or
negative and appears in a variety of systems. Therefore, we need a mechanism that provides a structured
approach for analysis and control of such behaviors. In this paper, we investigated the use of fuzzy rule-
based classification system combined with data retrieved from an ABS to verify and classify emergence.
Emergent behaviour is verified before-hand in many ABS models such as the Boids model, however,
detecting and classifying that phenomena is challenging. To address this issue, we propose a method to
classify the flocking behaviour in multi-agent system with a fuzzy system. At this moment, we are
extending the rules data base and we are testing it. The first results are promising and we are aiming to
validate this method with other ABS systems using more complex models.
6. Acknowledgements
The author would like to thank the DGRSDT (General Directorate of Scientific Research and
Technological Development) - MESRS (Ministry of Higher Education and Scientific Research),
ALGERIA, for the financial support of LISCO Laboratory.
7. References
[1] M. Randles, H. Zhu, and M. Taleb-Bendiab., ―A Formal Approach to the Engineering of
Emergence and its Recurrence‖, Proc of 2nd International Workshop on Engineering Emergence
in Decentralized Autonomic Systems, 2007.
[2] P., Jing, H., Hu, F., Zhan, Y., Chen, & Y., Shi. Agent-based simulation of autonomous vehicles: A
systematic literature review. IEEE Access, 8, 79089-79103, 2020
[3] J. Goldstein, ―Emergence as a construct: History and issues.‖ Emergence: Complexity and
Organization, 1(1):49-72, 1999cxc
[4] C. Johnson, ―What are emergent properties and how do they affect the engineering of complex
systems?‖ Reliability Engineering and System Safety, 91 12:1475-1481, 2006
[5] S. Singh, et al., ―Detection and classification of emergent behaviors using multi-agent simulation
framework (WIP),‖ In Proceedings of the Symposium on Modeling and Simulation of Complexity
in Intelligent, Adaptive and Autonomous Systems (MSCIAAS '17), Saurabh Mittal and Jose Luis
Risco Martin (Eds.). Society for Computer Simulation International, San Diego, CA, USA, Article
3, 8 pages, 2017
[6] C. W. Reynolds, ―Flocks, herds and schools: A distributed behavioral model‖. ACM SIGGRAPH
computer graphics vol. 21 (4), pp. 25–34. 1987.
[7] S. Kalantari, E. Nazemi, & B. Masoumi. Emergence phenomena in self-organizing systems: a
systematic literature review of concepts, researches, and future prospects. Journal of
Organizational Computing and Electronic Commerce, 30(3), 224-265, 2020
[8] J. Fromm, Types and Forms of Emergence‖, complexity Digest, vol. 25(3), 2005.
[9] M. Bedau, Weak emergence,‖ In Philosophical perspectives: Mind, causation, and world. ed. J.
Tomberlin. 11:375-399: Blackwell Publishers. 1997
[10] J. Fromm, The emergence of complexity,‖ Kassel: Kassel University Press, 2004.
[11] R. Grossman, et al., Discovering emergent behaviour from network packet data: Lessons from the
angle project.', 2009.
[12] R. Gore, and , P Jr. Reynolds., An exploration-based taxonomy for emergent behaviour analysis
in simulations,‖ In Proceedings of the 2007 Winter Simulation Conference. eds. S. G. Henderson,
B. Biller, M.-H. Hsieh, J. Shortle, J. D. Tew, and R. R. Barton. 1232-1240. Piscataway, New Jersy:
Institute of Electrical and Electronics Engineers, 2007.
[13] C. Szabo, and Y.M. Teo, An Objective-based Approach for Semantic Validation of Emergence in
Component-based Simulation Models,‖ ACM/IEEE/SCS 26th Workshop on Principles of
Advanced and Distributed Simulation. (2012)
[14] C. Szabo, and Y.M. Teo , Formalization of weak emergence in multiagent systems, ‖ ACM Trans.
Model. Comput. Simul. 26, 1, Article 6 (September 2015), 25 pages. DOI:
http://dx.doi.org/10.1145/2815502, 2015.
[15] S. Sivaraman, and M.M. Trivedi, , Looking at Vehicles on the Road: A Survey of Vision-Based
Vehicle Detection, Tracking, and Behavior Analysis,‖ IEEE Transactions on Intelligent
Transportation Systems ( Volume: 14 , Issue: 4), 2013.
[16] C. Hewitt, Viewing control structures as patterns of message passing', Artificial Intelligence, vol.
8, no 3, p. 323–374, 1977.
[17] B Fraser, et al., Simulating the effect of degraded wireless communications on emergent behavior,‖
Winter Simulation Conference (WSC), 2017.
[18] S.M., Cabaneros, J.K., Calautit, & B.R., Hughes. A review of artificial neural network models for
ambient air pollution prediction. Environmental Modelling & Software, 119, 285-304, 2019
[19] Gaganjot Kaur Kang, Jerry Zeyu Gao, Sen Chiao, Shengqiang Lu, and Gang Xie, ―Air Quality
Prediction: Big Data and Machine Learning Approaches,‖ International Journal of Environmental
Science and Development vol. 9, no. 1, pp. 8-16, 2018.
[20] Y. Zheng, Y. Xiuwen, L. Ming, L. Ruiyuan, S. Zhangqing, E. Chang, and L. Tianrui,. Forecasting
Fine-Grained Air Quality Based on Big Data‖. In Proceedings of the 21th ACM SIGKDD
International Conference on Knowledge Discovery and Data Mining (KDD '15). ACM, New York,
NY, USA, 2267-2276. DOI: https://doi.org/10.1145/2783258.2788573. 2015.
[21] W. k V. Chan, Interaction metric of emergent behaviours in agent-based simulation,‖ Proceedings
of the 2011 Winter Simulation Conference S. Jain, R.R. Creasey, J. Himmelspach, K.P. White,
and M. Fu, eds. (2011).
[22] P. J. Rousseeuw. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis.
Journal of Computational and Applied Mathematics, 20:53–65, 1987.
[23] Laughlin, Robert (2005), A Different Universe: Reinventing Physics from the Bottom Down, Basic
Books, ISBN 978-0-465-03828-2.
[24] L. Zadeh, Fuzzy sets,‖ Information Control, vol. 8, pp. 338–353, 1965.
[25] Y. Wang, P. Jia, L. Liu, C. Huang, & Z. Liu. A systematic review of fuzzing based on machine
learning techniques. PloS one, 15(8), e0237749, 2020
[26] E. H. Mamdani and S. Assilian, ―An experiment in linguistic synthesis with a fuzzy logic
controller,‖ International Journal of Man-Machine Studies, vol. 7, no. 1, pp. 1–13, 1975.
[27] E. H. Mamdani, Application of fuzzy logic to approximate reasoning using linguistic synthesis,‖
in Proceedings of the sixth international symposium on Multiple-valued logic. Los Alamitos, CA,
USA: IEEE Computer Society Press, 1976, pp. 196–202.
[28] V. Ojha, A. Abraham, & V. Snášel. (2019). Heuristic design of fuzzy inference systems: A review
of three decades of research. Engineering Applications of Artificial Intelligence, 85, 845-864,
2019.
[29] K. Mittal, A. Jain, K.S. Vaisla, O. Castillo, & J. Kacprzyk. (2020). A comprehensive review on
type 2 fuzzy logic applications: Past, present and future. Engineering Applications of Artificial
Intelligence, 95, 103916, 2020.
[30] X. Xiang, C. Yu, L. Lapierre, J. Zhang, & Q. Zhang. Survey on fuzzy-logic-based guidance and
control of marine surface vehicles and underwater vehicles. International Journal of Fuzzy
Systems, 20(2), 572-586, 2018.
[31] P. Kotak, & H. Modi,. Enhancing the data mining tool WEKA. 5th International Conference on
Computing, Communication and Security (ICCCS) (pp. 1-6). IEEE. 2020.
[32] I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques, 2nd
ed. San Francisco: Morgan Kaufmann, 2005.
[33] Wilensky, U. (1999) Netlogo. Evanston, IL: Center for Connected Learning and Computer-Based
Modeling, Northwestern University. Available via
[accessed October 10, 2010].
[34] Soltani, A., Battikh, T., Jabri, I., & Lakhoua, N. (2018). A new expert system based on fuzzy logic
and image processing algorithms for early glaucoma diagnosis. Biomedical Signal Processing and
Control, 40, 366-377. 2018.
[35] A. Boukehila., N. Taleb, & Y. Benazzouz. (2021). Interactions-based method to detect emergent
behavior in ongoing simulations. International Journal of Modeling, Simulation, and Scientific
Computing, 12(04), 2150022.
[36] A. Boukehila, & N. Taleb. (2019, November). Case-Based Approach to Detect Emergence.
In Proceedings of the 2019 3rd International Conference on Big Data Research (pp. 98-102).
[37] A. Boukehila, & N. Taleb. (2020, February). Statistical Study To Detect Emergent Behaviours.
In 2020 2nd International Conference on Mathematics and Information Technology (ICMIT) (pp.
164-168). IEEE.