Bayesian networks and evolutionary algorithms veteran spanning federated causal discovery and telemetry anomaly detection
4
papers here
4
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
24
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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—
No reliable structured output metric is available yet, so output is left unscored. This axis would measure total and per-year publication volume, venue types, and authorship positions across the researcher's indexed record, none of which can be estimated from the supplied data.
Impact—
Impact cannot be scored yet: citation totals for this profile are not present in the supplied data, beyond the h-index of 24 and i10-index of 45 recorded on OpenAlex. This axis would measure citation uptake, influence on follow-up work, and per-paper citation trajectories.
Focus0/100
Focus scores 0 because the recent record is evenly spread: OpenAlex field weights sit at 0.25 each across causal discovery/federated learning, spacecraft telemetry anomaly detection, probabilistic graphical models/evolutionary computation, and AI broadly. Topic counts fan across five clusters, led by Bayesian Modeling and Causal Inference (59) and Data Mining Algorithms (46). This reads as deliberate breadth: each 2026 line extends the same probabilistic-plus-metaheuristic modeling core.
Collaboration48/100
Collaboration scores 48, reflecting a compact, repeat-partner pattern: 10 distinct coauthors in the recent three-year window. The frequent-coauthor list is anchored by long-standing collaborators Pablo Torrijos and José A. Gámez, with Fabio Stella, Óscar Alcarria, Rafael Sánchez and others recurring. All four 2026 catalog papers are multi-authored, with Puerta in the last position on three and middle on one — senior supervision rather than solo work.
Activity89/100
Activity scores 89 on evidence of a strong late-career surge: 20 works in the last three years, with OpenAlex yearly counts of 8 (2024), 6 (2025) and 6 (2026) after a quieter 2019-2023 run (as low as 1 in 2020). All four 2026 arXiv papers in the Evolve catalog land in a single year, spanning Bayesian network fusion, federated causal discovery, and telemetry anomaly detection.
Reproducibility89/100
Reproducibility scores 89, the mean readiness rating across 4 reviewed papers. Three of the four rate high — the treewidth-limited fusion paper released code, and the federated cumulant study is judged sound and honest — while the spacecraft telemetry paper is medium, with Evolve flagging model choice on full-mission data and a 0.408 M1 recall. This is a consistently open, verifiable output style.
Research summary
José M. Puerta is a computer science researcher at the University of Castilla-La Mancha, Spain, whose OpenAlex-indexed record spans more than three decades, from 1993 to 2026, with an h-index of 24 and an i10-index of 45 on that source. His topical core is Bayesian modeling and causal inference (59 works), flanked by data mining algorithms (46) and metaheuristic and evolutionary optimization (22 and 19 works), a combination long associated with probabilistic graphical model research in Spain. The recent record shows a marked late-career acceleration: 20 works in the last three years, with yearly counts recovering to 8 in 2024 after a quieter 2019-2023 stretch. Four 2026 arXiv papers in the Evolve catalog illustrate the current program: two on tractable structural fusion of Bayesian networks using genetic algorithms under treewidth constraints, one on federated causal discovery via regression-directed cumulants, and one on unsupervised anomaly detection for spacecraft telemetry with adaptive EVT thresholding. Across these, Puerta typically occupies the senior last-author position, working with repeat collaborators including Pablo Torrijos, José A. Gámez, and Fabio Stella. Evolve rates reproducibility highly (mean 89 across 4 papers, with code released on the fusion work), while flagging self-baseline comparisons, a 0.408 M1 recall on telemetry, and cumulant-variance findings that temper causal claims.
Collaboration pattern
The collaboration pattern is compact and durable: 10 distinct coauthors in the recent three-year window, with repeated partners rather than rotating ad-hoc teams. Pablo Torrijos and José A. Gámez anchor the frequent-coauthor list, alongside Fabio Stella, Óscar Alcarria, Rafael Sánchez, Javier Sempere, Juan C. Alfaro, Juan M. Auñón and Juan A. Aledo. In all four 2026 catalog papers Puerta takes a senior authorship position (last on three, middle on one).
Research keywords
Highlights
Treewidth-limited Bayesian network fusion with genetic algorithms
Pre-fusion edge pruning makes consensus BN fusion tractable
Federated causal discovery without moving raw data
Label-free anomaly detection for spacecraft telemetry
Timeline
- 1993Earliest year in the OpenAlex record (3 works), marking the start of a publishing trajectory that is still active in 2026.
- 2010Output climbs to 12 works, opening the most productive stretch on record.
- 201113 works — the highest single-year count in the indexed record.
- 2024Activity re-accelerates: 8 works, after a quieter 2019-2023 run that dipped to 1 in 2020.
- 2026Four arXiv papers enter the Evolve catalog: two Bayesian network fusion studies, a federated causal discovery method, and spacecraft telemetry anomaly detection.
Papers in the Evolve catalog
Structural Fusion of Bayesian Networks with Limited Treewidth Using Genetic Algorithms
Structure-Aware Unsupervised Anomaly Detection for Spacecraft Telemetry with Adaptive EVT Thresholding
Genetic Algorithms for Tractable Bayesian Network Fusion via Pre-Fusion Edge Pruning
Federated Causal Discovery via Regression-Directed Cumulants
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.
- Bibliometrics (h-index 24, i10-index 45, yearly counts, topics) come from OpenAlex only; no other citation source is included.
- Output and impact dimensions are unscored because the underlying data is insufficient; no estimates were made.
- All four catalog papers are 2026 arXiv preprints; Evolve verdicts are independent reviews, not peer-review outcomes.
- Profile is name-resolved from public bibliographic data (OpenAlex identity, ORCID 0000-0002-9164-5191); biography beyond these records is not covered.