Early-career web-mining ML researcher: behavioral fingerprinting, user re-identification, and next-domain prediction
1
papers here
1
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-20) 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 indexed record currently contains too few outputs to measure reliably. This axis would track total publications, venue types, and first- or corresponding-author share across the record.
Impact—
Not yet scored: citation counts and downstream uptake are unavailable in the current data. This axis would measure scholarly influence through citations, citing works, and adoption of the reviewed methods.
Focus100/100
The focus score is 100: the field distribution in the raw data contains exactly one field — web mining, user modeling, and privacy (machine learning) — and the single catalogued paper is assigned to it. All indexed output therefore sits on one topic line: behavioral fingerprinting, user re-identification, and next-domain prediction from browsing traces. With only one indexed paper, this concentration should be read as a narrow early record, not a confirmed long-term specialization.
Collaboration32/100
The collaboration score is 32, computed from a compact footprint of 4 distinct co-authors in the underlying bibliometric data. Work to date sits within one small recurring team — Omran Berjawi, Walid Fahs, and Rida Khatoun appear as frequent co-authors — a pattern typical of an early, single-group research phase rather than an extended international network. Broader collaboration signals should emerge as more outputs are indexed.
Activity—
Not yet scored: the publication record is too sparse to compute recent-year trends. This axis would track annual output volume and the continuity of research activity over recent years.
Reproducibility55/100
The reproducibility score is 55 (reproMean), computed from one reviewed paper. Evolve's verdict calls the study transparent, which supports reproduction, while reproducibility readiness is rated medium. The main deductions are methodological: sub-point gains attributed to LLM components lack statistical significance tests, and the comparison baselines are described as trivial. Adding significance tests and stronger baselines would be the most direct route to a higher score.
Research summary
Ralph Elsaghbini is an early-stage researcher working at the intersection of web mining, user modeling, and privacy, with machine learning as the methodological core. The current profile rests on a deliberately small evidence base: one paper in the Evolve catalog — the 2026 arXiv preprint 'Behavioral Fingerprinting and Navigation Prediction in Web Browsing' — on which Elsaghbini is first author (non-corresponding). That study models browsing traces and reports that roughly 90% of users can be identified from short sessions, while the next visited domain can be predicted with about 80% accuracy using a GraphSAGE-based approach. Evolve's review characterizes the study as transparent, while flagging that sub-point gains attributed to LLM components lack statistical testing and are compared against trivial baselines; reproducibility readiness is rated medium (score 55). The bibliometric radar reflects a maximally concentrated focus (100) in a single field, a small collaboration footprint (32) built on four distinct co-authors — repeatedly Omran Berjawi, Walid Fahs, and Rida Khatoun — and missing coverage for output, impact, and activity. Taken together, this is a focused, first-author-driven early profile whose headline claims warrant the statistical tightening the review recommends.
Collaboration pattern
The bibliometric snapshot shows a compact collaboration footprint: 4 distinct co-authors, with Omran Berjawi, Walid Fahs, and Rida Khatoun recurring as frequent collaborators. The pattern suggests work within a single small team on the one catalogued paper rather than a broad network. No institution or country data are available to contextualize the affiliations behind these names.
Research keywords
Highlights
~90% user identification from short browsing sessions
~80% next-domain prediction accuracy with GraphSAGE
Transparent study, with statistical caveats
Timeline
- 2026First-author arXiv preprint 'Behavioral Fingerprinting and Navigation Prediction in Web Browsing' appears in the catalog, assigned to web mining, user modeling, and privacy (machine learning).
- 2026The study reports ~90% user identification from short sessions and ~80% next-domain prediction accuracy using a GraphSAGE-based model.
- 2026Evolve's review credits transparency but flags untested LLM gains and trivial baselines; reproducibility readiness is rated medium (55).
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.
- Dimension scores come from a limited bibliometric snapshot and currently rest on a single indexed paper.
- Output, impact, and activity axes are unscored because the underlying data are insufficient.
- No institution, country, or ORCID is on record; the profile is name-resolved from public bibliographic data.
- Quoted findings come from Evolve's review of one arXiv preprint and have not been independently verified.