Explainable-AI researcher at the University of Thessaly applying neuro-fuzzy models to maritime predictive maintenance
2
papers here
2
reviewed by Evolve
2
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-13) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Not yet scored: the snapshot contains no aggregate output metrics (publication counts or venue history) for this researcher. This axis would measure volume and consistency of published work across years and venues once sufficient records are indexed.
Impact—
Not yet scored: citation counts and downstream uptake are absent from the available data. This axis would measure the work's influence through citations and reuse; it is left unscored rather than estimated.
Focus0/100
The computed focus score is 0, derived from a field distribution split exactly evenly — 0.5 for 'predictive maintenance / applied machine learning' and 0.5 for 'machine learning (explainable AI / predictive maintenance)'. In practice the two labeled fields overlap heavily: both catalog papers combine explainable AI methods (neuro-fuzzy systems, fuzzy decision trees, saliency maps) with condition-based maintenance of maritime and naval systems, so the low score reflects labeling breadth more than genuinely scattered research interests.
Collaboration36/100
The collaboration score is 36, computed from a compact author network: 5 distinct coauthors appear in the snapshot. Repeat collaboration is visible in Evolve's frequent-coauthor list, where Dimitris K. Iakovidis and Panagiotis G. Kalozoumis each appear twice and Georgia Sovatzidi once. Kalogeropoulos is first author on both catalog papers, consistent with a small, stable team producing applied XAI work for maritime engineering rather than large multi-institution consortia.
Activity—
Not yet scored: the data lacks a longitudinal publication record, so recent-output cadence cannot be computed. This axis would measure publishing activity over time — papers per year and recency — and is left unscored until more indexed records are available.
Reproducibility20/100
Reproducibility scores 20, a mean taken over 2 papers assessed by Evolve; both carry an explicit 'low' reproducibility-readiness rating. The reviews flag concrete concerns: the maritime neuro-fuzzy paper's 'outperforms SOTA' claim overstates ties in Table I (SVM at 0.99) and lacks variance statistics, while the naval-propulsion paper's advantage claim is weakened by limited baseline comparison and possible data leakage. Reproducibility would likely improve with released code, variance reporting, and stronger baselines.
Research summary
Dionisis Kalogeropoulos is a researcher at the University of Thessaly working at the intersection of explainable artificial intelligence and predictive maintenance, with a focus on maritime and naval applications. This profile is name-resolved from public bibliographic data. Evolve's catalog currently holds two of his papers, both 2026 arXiv preprints with him as first author. The first distils a neuro-fuzzy network into saliency maps and fuzzy decision-tree rules to support trustworthy maritime decision-making; the second presents an explainable condition-based maintenance framework for naval-propulsion systems, combining a residual neural network with a fuzzy head distilled into local fuzzy decision trees and reporting 99.24% accuracy on the public CODLAG dataset. His field distribution splits evenly (0.5/0.5) between predictive maintenance / applied machine learning and machine learning (explainable AI / predictive maintenance), yielding a computed focus score of 0. Collaboration is compact — 5 distinct coauthors, with Dimitris K. Iakovidis and Panagiotis G. Kalozoumis recurring — and scores 36. Both papers carry a reproducibility mean of 20 and a 'low' readiness rating in Evolve's reviews, with flagged baseline, variance, and leakage concerns.
Collaboration pattern
Kalogeropoulos works within a small, stable circle: the snapshot records 5 distinct coauthors and a collaboration score of 36. Recurring partners are Dimitris K. Iakovidis and Panagiotis G. Kalozoumis, each appearing twice in Evolve's frequent-coauthor list, with Georgia Sovatzidi also listed. He is first author on both catalog papers, suggesting a pattern in which he leads the modeling and writing while the same core team recurs across outputs in applied explainable AI for maritime engineering.
Research keywords
Highlights
Neuro-fuzzy explanations for trustworthy maritime decision-making
Explainable condition-based maintenance for naval-propulsion systems
A coherent first-author XAI-for-maintenance agenda
Both Evolve-reviewed entries are 2026 first-author preprints sharing one methodology — fuzzy-logic explainability (fuzzy decision trees, saliency, rule extraction) applied to maritime asset health. The field distribution splits 0.5/0.5 across two adjacent labels, and both papers carry 'low' reproducibility readiness.
Timeline
- 2026Releases first-author arXiv preprint 'Explainable Neuro-Fuzzy Prediction for Trustworthy Decision-Making in Maritime', later reviewed into the Evolve catalog.
- 2026Releases first-author arXiv preprint 'Explainable Predictive Condition-based Maintenance of Naval-Propulsion Systems using Fuzzy Logic', reporting 99.24% accuracy on the public CODLAG dataset.
- 2026Evolve reviews both preprints: the explanation mechanisms are rated promising, but both papers receive a 'low' reproducibility-readiness rating over statistical and baseline concerns.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Output, impact, and activity axes are unscored: the bibliometric snapshot lacks publication counts, citation metrics, and longitudinal records.
- Only the 2 Evolve-reviewed papers (both 2026 arXiv preprints) are in the catalog; no full publication list or indexed citation data was available.
- No ORCID or country is recorded; the profile is name-resolved from public bibliographic data.
- The reproducibility mean (20) covers only 2 reviewed papers, and Evolve verdicts reflect the assessment at review time.