arxiv: 2604.23725 · v1 · submitted 2026-04-26 · 💻 cs.SI

Recognition: unknown

Uncertainty-Aware Fuzzy Centrality Measures for Influential Node Identification: A Structural Modeling Approach Toward E-Commerce Applications

Shima Esfandiari , Seyed Mostafa Fakhrahmad

Authors on Pith no claims yet

Pith reviewed 2026-05-08 04:59 UTC · model grok-4.3

classification 💻 cs.SI

keywords e-commerce networksfuzzy centralityinfluential nodesuncertainty modelingnetwork analysisstructural modelingrecommendation systemsinfluential node identification

0 comments

The pith

Uncertainty-aware fuzzy centrality measures improve influential node identification in e-commerce networks by handling implicit interaction uncertainty.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

E-commerce platforms create large-scale networks where interactions between users, products, and entities are often uncertain and derived from noisy implicit signals. Deterministic network models cannot adequately capture this uncertainty, resulting in unreliable identification of influential nodes crucial for marketing, recommendations, and behavior analysis. The paper proposes uncertainty-aware fuzzy centrality measures through a structural modeling approach to address this limitation. A sympathetic reader would care if these measures lead to more accurate influence rankings, enabling better decision-making in commercial applications. This shifts the focus from assuming clear relationships to modeling gradations of uncertainty in network structures.

Core claim

The paper establishes that by incorporating fuzzy logic to represent uncertain relationships in e-commerce interaction networks, new centrality measures can be defined that better identify influential nodes than traditional deterministic approaches, providing a structural framework tailored to applications like online marketing and customer analysis.

What carries the argument

Uncertainty-aware fuzzy centrality measures, which use fuzzy sets to model and propagate uncertainty in network edges for calculating node influence.

If this is right

More accurate targeting in online marketing campaigns based on reliable influence rankings.
Enhanced recommendation systems that account for uncertain user interactions.
Improved customer behavior analysis in noisy e-commerce data environments.
Extension of network analysis techniques to other domains with implicit relationships.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the measures prove effective, they could be adapted to real-time dynamic networks in e-commerce platforms.
This approach connects to broader challenges in handling uncertainty in graph-based machine learning for commercial data.
Testing on diverse e-commerce datasets could reveal domain-specific adjustments needed for the fuzzy parameters.
Potential to reduce false positives in identifying influencers who appear central only due to noisy data.

Load-bearing premise

That fuzzy logic can meaningfully represent and propagate the uncertainty present in implicit e-commerce interaction signals without introducing artifacts that distort node influence rankings.

What would settle it

Running the proposed fuzzy centrality measures and standard deterministic ones on a labeled e-commerce dataset with known influential nodes or outcomes, and checking if the fuzzy versions consistently produce rankings that better predict real influence metrics such as sales impact or engagement levels; mismatch in performance would challenge the claim.

Figures

Figures reproduced from arXiv: 2604.23725 by Seyed Mostafa Fakhrahmad, Shima Esfandiari.

**Figure 3.** Figure 3: The run time of the five methods in the dataset. VI. CONCLUSION This paper proposed novel uncertainty-aware fuzzy centrality measures designed to identify influential nodes in fuzzy networks, with a specific emphasis on applications in ecommerce scenarios. Recognizing the challenges of accessing real e-commerce data due to privacy and proprietary constraints, we introduced the concept of using structurall… view at source ↗

read the original abstract

In recent years, e-commerce platforms have become one of the most prominent examples of large-scale interaction networks, where understanding influence dynamics among users, products, and digital entities is essential for applications such as online marketing, recommendation systems, and customer behavior analysis. A key challenge in these platforms is that interactions are often uncertain, noisy, and inferred from implicit signals rather than explicitly defined relationships. This uncertainty cannot be effectively captured using deterministic network models...

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Fuzzy centrality for uncertain e-commerce graphs is internally consistent and reduces cleanly to the standard case, but the work stays mostly methodological with thin validation.

read the letter

This paper introduces uncertainty-aware fuzzy centrality measures for identifying influential nodes in e-commerce interaction networks. It models uncertainty in interaction weights through membership functions and uses fuzzy min/max operations to compute the scores and rankings. The main technical move is showing that the fuzzy versions collapse back to ordinary centrality when uncertainty is zero, and the propagation rules are written out explicitly. That part is clean and avoids internal contradictions or hidden fitting assumptions. The derivations stay consistent across the formulas they present. The e-commerce motivation is handled as context rather than a formal claim that needs proof, which keeps the focus on the structural modeling. The approach is general enough that it could slot into existing network tools without new data requirements. The soft spot is the lack of reported experiments. There are no head-to-head tests on real platform data, no accuracy metrics against probabilistic alternatives, and no checks on whether the fuzzy rankings actually improve downstream tasks like marketing or recommendations. The practical advantage is asserted from the motivation rather than demonstrated. This is the sort of paper that would interest researchers working on noisy interaction graphs in applied network analysis or fuzzy systems. A reader already familiar with centrality extensions would pick up the specific propagation rules quickly. I would send it to peer review. The core math holds up and the application area is relevant, so referees can focus on adding the missing validation and clarifying how much is new versus prior fuzzy network work.

Referee Report

0 major / 2 minor

Summary. The manuscript proposes uncertainty-aware fuzzy centrality measures for identifying influential nodes in large-scale e-commerce interaction networks. It argues that deterministic network models cannot capture uncertainty arising from implicit, noisy interaction signals, and instead defines fuzzy centrality via membership functions on interaction weights, derives ranking formulas that reduce to standard centrality measures when uncertainty is zero, and provides explicit propagation rules using fuzzy min/max operations for application to e-commerce structural modeling.

Significance. If the derivations hold under empirical scrutiny, the work offers a principled extension of centrality analysis to uncertain networks, with direct relevance to recommendation systems and marketing on e-commerce platforms. Strengths include the internal consistency of the fuzzy construction, the explicit reduction to classical centrality under zero uncertainty, and the use of standard fuzzy operations without introducing hidden parameters or circularity.

minor comments (2)

Abstract: the phrasing 'structural modeling approach' is used without defining the specific structural properties (e.g., graph type, weight semantics) being modeled beyond the centrality definitions themselves.
The manuscript would benefit from an explicit statement in the methods section of how membership functions are chosen or calibrated for implicit e-commerce signals, as this choice directly affects ranking stability.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our manuscript and the recommendation for minor revision. The referee's summary accurately captures the core contributions of the uncertainty-aware fuzzy centrality measures, including the handling of noisy interaction signals in e-commerce networks via membership functions, the explicit reduction to classical centrality when uncertainty is zero, and the use of standard fuzzy min/max operations. As no specific major comments appear in the report, we have no points requiring detailed rebuttal or clarification at this time.

Circularity Check

0 steps flagged

No significant circularity; fuzzy extension remains independent of inputs

full rationale

The manuscript defines fuzzy centrality via membership functions on interaction weights and derives ranking formulas that reduce to standard centrality when uncertainty is zero, as noted in the skeptic analysis. This is an internally consistent structural extension of existing network concepts rather than a self-referential fit or self-citation chain. No equations or steps in the abstract or described derivations reduce by construction to the paper's own fitted parameters, prior self-citations, or renamed empirical patterns. The e-commerce motivation serves as application context without load-bearing formal claims that loop back to the definitions.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no identifiable free parameters, axioms, or invented entities; the proposal rests on unstated assumptions about how fuzzy membership functions map to e-commerce uncertainty.

pith-pipeline@v0.9.0 · 5372 in / 883 out tokens · 52626 ms · 2026-05-08T04:59:03.414384+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

37 extracted references · 8 canonical work pages

[1]

Costa, O.N

L.d.F. Costa, O.N. Oliveira Jr, G. Travieso, F.A. Rodrigues, P.R. Villas Boas, L. Antiqueira, M.P. Viana, L.E. Correa Rocha, Analyzing and modeling realworld phenomena with complex networks: a survey of applications. Advances in Physics 60(3), 329–412 (2011)

2011
[2]

Gurrapu, N

N. Gurrapu, N. Baydeti, Rumor detection from online social media through feature selection and machine learning. Multimedia Tools and Applications pp. 1–25 (2025)

2025
[3]

R. Wang, J. Li, H. Sun, S. Duan, G. Zhang, Y. Xiao, Rumor spreading model based on emotional characteristics and influence. Expert Systems with Applications 274, 126952 (2025)

2025
[4]

Harris, C.M

E.E. Harris, C.M. Petrovits, M.H. Yetman, Spreading the news: Donor response to disclosures about nonprofit fraud. Nonprofit and Voluntary Sector Quarterly 53(3), 638–665 (2024)

2024
[5]

Amiri, H

M. Amiri, H. Sotudeh, Comparative opinion mining of tweets on retracted papers and their valid peers: a semi -experimental follow -up. Scientometrics pp. 1–21 (2025)

2025
[6]

Esfandiari, S.M

S. Esfandiari, S.M. Fakhrahmad, Identifying influential users in social networks with an extended hybrid global structure model, in 2024 10th International Conference on Web Research (ICWR) (IEEE, 2024), pp. 1– 5

2024
[7]

Esfandiari, S.M

S. Esfandiari, S.M. Fakhrahmad, Mining Influential Spreaders in Complex Networks by an Effective Combination of the Degree and K - Shell, in 20 24 20th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP) (IEEE, 2024), pp. 1–6

2024
[8]

L. Li, L. Yuan, and J. J. Tian, Influence of online E-commerce interaction on consumer satisfaction based on big data algorithm, Heliyon, vol. 9, no. 8, p. e18322, 2023. doi: 10.1016/j.heliyon.2023.e18322

work page doi:10.1016/j.heliyon.2023.e18322 2023
[9]

C. Chen, M. Chih, and D. T. T. Nguyen, How e -commerce platform features impact post -adoption consumer behaviors: evidence from Vietnam's online cosmetic sector, Asia Pacific Journal of Market ing and Logistics, vol. ahead -of-print, no. ahead -of-print, 2025. doi: 10.1108/APJML-04-2025-0625

work page doi:10.1108/apjml-04-2025-0625 2025
[10]

Shang, Y.C

D.Chen, L.Lu, M.S. Shang, Y.C. Zhang, T.Zhou, Identifying influential nodes in complex networks. Physica A: Statistical Mechanics and its Applications 391(4) , 1777–1787 (2012). https://doi.org/10.1016/j.physa.2011.09.017

work page doi:10.1016/j.physa.2011.09.017 2012
[11]

The European Physical Journal B , author =

M.Barthelemy, Betweenness centrality in large complex networks. The European Physical Journal B 38(2), 163 –168 (2004). https://doi.org/10.1140/epjb/e2004-00111-4

work page doi:10.1140/epjb/e2004-00111-4 2004
[12]

Cohen, D

E. Cohen, D. Delling, T. Pajor, R.F. Werneck, Computing classic closeness centrality, at scale, in Proceedings of the Second ACM Conference on Online Social Networks (ACM, 2014), pp. 37 –50. https://doi.org/10.1145/2660460.2660465

work page doi:10.1145/2660460.2660465 2014
[13]

Ruhnau, Eigenvector -centrality a node -centrality Social Networks 22(4), 357–365 (2000)

B. Ruhnau, Eigenvector -centrality a node -centrality Social Networks 22(4), 357–365 (2000). https://doi.org/10.1016/S0378-8733(00)00016-0

work page doi:10.1016/s0378-8733(00)00016-0 2000
[14]

Zhang, H.E

L.Lu, T.Zhou, Q.M. Zhang, H.E. Stanley, The h-index of a network node and its relation to degree and coreness. Nature Communications 7, 10168 (2016). https://doi.org/10.1038/ncomms10168

work page doi:10.1038/ncomms10168 2016
[15]

Haveliwala, Efficient computation of pagerank

T. Haveliwala, Efficient computation of pagerank. Tech. rep., Stanford (1999)

1999
[16]

Esfandiari, S.M

S. Esfandiari, S.M. Fakhrahmad, The collaborative role of k -shell and pagerank for identifying influential nodes in complex networks. Physica A: Statistical Mechanics and its Applications 658, 130256 (2025)

2025
[17]

Esfandiari, M.R

S. Esfandiari, M.R. Moosavi, Identifying influential nodes in complex networks through the k -shell index and neighborhood information. Journal of Computational Science 84, 102473 (2025)

2025
[18]

Zareie, R

A. Zareie, R. Sakellariou, Centrality measures in fuzzy social networks. Information Systems 114, 102179 (2023)

2023
[19]

Z. Lu, Q. Zhang, X. Du, D.Wu, F. Gao, A fuzzy social network centrality analysis model for interpersonal spatial relations. Knowledge -Based Systems105,206–213(2016)

2016
[20]

R.j. Hu, G.y. Zhang, L.p. Liao, in Fuzzy Information Engineering and Operations Research Management (Springer, 2013), pp. 457–464

2013
[21]

Cooper, A

I. Cooper, A. Mondal, C.G. Antonopoulos, A sir model assumptio n for the spread of covid -19 in different communities. Chaos, Solitons & Fractals 139, 110057 (2020)

2020
[22]

Zadeh, Fuzzy sets

L.A. Zadeh, Fuzzy sets. Information and control 8(3), 338 –353 (1965) L.A. Zadeh, A fuzzy-algorithmic approach to the definition of complex or imprecise concepts. International Journal of Man -Machine Studies 8(3), 249–291(1976)

1965
[23]

D. Qin, Z. Peng, L. Wu, Fchg: Fuzzy cognitive hypergraph for interpretable fault detection. Expert Systems with Applications 255, 124700 (2024)

2024
[24]

Esfandiari , M

S. Esfandiari , M. Fakhrahmad, Predic ting node influence in complex networks by the k -shell entropy and degree centrality, in Companion Proceedings of the ACM Web Conference 2024, May 2024, pp. 629–632

2024
[25]

Esfandiari, S

S. Esfandiari, S. M. Fakhrahmad, Identifying influential nodes in complex networks by adjusted feature contributions and neighborhood impact, The Journal of Supercomputing, vol. 81, no. 3, p. 503, 2025

2025
[26]

S. Esfandiari, A multi -attribute model to identify influential nodes in complex networks: Addressing the k -shell limitations, in Companion Proceedings of the ACM on Web Conference 2025, May 2025, pp. 3092– 3095

2025
[27]

S. Esfandiari, Enhancing data quality by identifying influential nodes: Integrating complementary features with CoCoFISo, in Companion Proceedings of the ACM on Web Conference 2025, May 2025, pp. 2116– 2119

2025
[28]

Esfandiari, M

S. Esfandiari, M. Fakhrahmad, A novel K -shell-based feature for detecting influential nodes in directed complex networks: Addressing resolution and core -like groups problems, Information Processing & Management, vol. 63, no. 6, p. 104719, 2026. doi: 10.1016/j.ipm.2026.104719

work page doi:10.1016/j.ipm.2026.104719 2026
[29]

On certain integrals of Lipschitz- Hankel type involving products of Bessel functions,

G. Eason, B. Noble, and I.N. Sneddon, “On certain integrals of Lipschitz- Hankel type involving products of Bessel functions, ” Phil. Trans. Roy. Soc. London, vol. A247, pp. 529-551, April 1955

1955
[30]

Apostolopoulos, N.I

I.D. Apostolopoulos, N.I. Papandrianos, N.D. Papathanasiou, E.I. Papageorgiou, Fuzzy cognitive map applications in medicine over the last two decades: a review study. Bioengineering 11(2), 139 (2024)

2024
[31]

Ahmadi, A

H. Ahmadi, A. Kamal, P. Kumar, k -infnode: ranking top -k influential nodes in complex networks with random walk. Soft Computing 29, 1677– 1690 (2025)

2025
[32]

X. Tang, Y. Zhang, Z. Liu, Fuzzy-weighted model for influence measure in social and biological networks. Journal of Network Science (2024)

2024
[33]

Y.Wang, Z. Chen, H. Zhang, Fuzzy transfer function for node influence estimation in complex networks. Applied Mathematics and Computation (2024)

2024
[34]

X. Li, L. Zhang, J. Gao, Fuzzy weighted information divergence in multilayer networks. Computers & Operations Research (2024)

2024
[35]

Rossi, N

R. Rossi, N. Ahmed, The network data repository with interactive graph analytics and visualization, in Proceedings of the AAAI conference on artificial intelligence, vol. 29 (2015)

2015
[36]

Y. Lou, L. Wang, G. Chen, Structural robustness of complex networks: A survey of a posteriori measu res [feature]. IEEE Circuits and Systems Magazine 23(1), 12–35 (2023)

2023
[37]

Costa, S.C

G.S. Costa, S.C. Ferreira, Nonmassive immunization to contain spreading on complex networks. Physical Review E 101(2), 022311 (2020)

2020