Institut Mines-Télécom researcher bridging a semantic-web-era agenda to current temporal IR and knowledge-tracing work
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
18
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) 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—
Output is unscored in this snapshot: the composite data pipeline had insufficient coverage for this axis. It would measure cumulative publication volume and annual productivity across the record; no output score or estimate is provided here.
Impact—
Impact is unscored: citation-influence data was insufficient for this axis in the snapshot. It would measure field-normalized citation influence. Raw h-index (18) and i10-index (35) exist in the record but no composite impact score.
Focus0/100
Focus scores 0/100, the lowest concentration band: the record spans five sizable topics — Semantic Web and Ontologies (46), Context-Aware Activity Recognition (34), Recommender Systems (26), Information Retrieval and Search Behavior (20), and Intelligent Tutoring Systems and Adaptive Learning (20) — while recent indexed works split evenly (0.5/0.5) between NLP-driven information retrieval and educational knowledge tracing. Breadth, not a single specialty, characterizes this profile.
Collaboration41/100
Collaboration scores 41/100, a moderate band. The recent window lists 14 works with 7 distinct coauthors, and repeat collaborators include Mourad Hassani, Julien Romero, Christian Jacquelinet, Samuel Girard, Juan D. Pinto, and Jill-Jênn Vie. In the two cataloged 2026 preprints, Bouzeghoub appears in middle position on TEMPS and last position on RegKT — a senior, team-embedded pattern rather than solo first-author work.
Activity79/100
Activity scores 79/100. The record spans 2003–2026 with 14 works in the most recent three-year window, and 2026 already shows 6 indexed works. Output peaked at 17 works in both 2016 and 2017, dipped to 3–4 annually during 2019–2021, and has held at 4–6 per year since 2022 — sustained, current engagement rather than a wind-down.
Reproducibility78/100
Reproducibility scores 78/100, drawn from Evolve's review of 2 cataloged papers with a mean readiness of 78. TEMPS (temporal sentence embeddings) is rated high readiness, while RegKT (IRT-regularized knowledge tracing) is medium: Evolve notes its gains are consistent but small, untested, and the study preliminary. The axis reflects assessed openness and replicability of recent work only.
Research summary
Amel Bouzeghoub is a France-based researcher at Institut Mines-Télécom whose bibliometric record, resolved from OpenAlex (ORCID 0000-0003-4890-9005), spans 2003–2026. The record shows an h-index of 18 and an i10-index of 35, with 14 works in the most recent three-year window and 7 distinct coauthors. Topic tagging places the historical core in Semantic Web and Ontologies (46 tagged works), Context-Aware Activity Recognition Systems (34), Recommender Systems and Techniques (26), Information Retrieval and Search Behavior (20), and Intelligent Tutoring Systems and Adaptive Learning (20). Evolve has reviewed two 2026 arXiv preprints: TEMPS, on weakly-supervised Gaussian temporal sentence embeddings for dense retrieval, where Bouzeghoub is a middle author, and RegKT, which adds an item-response-theory regularizer to deep knowledge tracing for interpretability and overfitting resistance, where Bouzeghoub is the last author. Evolve judges TEMPS's gains real on TimeQA/TS-Retriever though mixed on TempReason, and rates its reproducibility readiness high; RegKT's gains are consistent but small and the study preliminary, with medium readiness. Recent indexed output splits evenly (0.5/0.5) between NLP-oriented information retrieval and educational knowledge tracing, marking a contemporary pivot from the earlier semantic-web and context-aware agenda. Output and impact composite scores are unavailable in this snapshot; the remaining radar dimensions — activity 79, reproducibility 78, collaboration 41, focus 0 — are computed from the raw record.
Collaboration pattern
The collaboration radar reads 41/100. The recent three-year window holds 14 works with 7 distinct coauthors. Repeat collaborators named in the record include Mourad Hassani, Julien Romero, Christian Jacquelinet, Samuel Girard, Juan D. Pinto, and Jill-Jênn Vie. Authorship positions on the two cataloged preprints — middle on TEMPS, last on RegKT — point to a senior contributor embedded in small teams rather than a first-author driver; neither 2026 paper lists Bouzeghoub as corresponding author.
Research keywords
Highlights
TEMPS: weakly-supervised temporal sentence embeddings for dense retrieval
RegKT: IRT-regularized deep knowledge tracing for interpretability
Timeline
- 2003Earliest year in the bibliometric record, with 3 indexed works.
- 2012Annual output reaches 14 indexed works, a high point ahead of the later peak.
- 2016Output peaks at 17 indexed works, matched again in 2017.
- 2019Output contracts to 4 works, holding at 4 then 3 through 2021.
- 2026Two Evolve-reviewed arXiv preprints appear (TEMPS, RegKT); the record shows 6 works this year.
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.
- Bibliometrics are drawn from an OpenAlex name-resolved record (ORCID 0000-0003-4890-9005); counts, topics, and indices reflect that record only.
- Output and impact dimensions are unscored (dimMissing); raw h-index 18 and i10-index 35 are present, but no field-normalized citation benchmarks were available.
- Only two papers (both 2026 arXiv preprints) have Evolve reviews; the wider publication record is not individually assessed.
- The frequent-coauthor list includes 'Amel Bouzeghoub' itself, likely a self-name artifact in the source data.