Seunghwan Park (Kangwon National University): robust regression, variable selection, and imbalanced classification
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-09) 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 publication-count data is available in the current bibliometric snapshot, so this axis is unscored. It would measure the volume and cadence of indexed research outputs — preprints, journal articles, and other publications — across the researcher's career.
Impact—
Citation and influence metrics are not yet available for this profile, so the axis cannot be scored. It would measure cumulative citations, citations per paper, and the uptake of this researcher's work by subsequent studies.
Focus0/100
The recorded field distribution splits exactly evenly — 0.5 for statistical machine learning on imbalanced classification and 0.5 for statistics/statistical machine learning on robust regression and variable selection — producing a focus score of 0. With only two reviewed 2026 preprints, this reflects two parallel methodological threads rather than confirmed long-term fragmentation, and the score may change as more output is indexed.
Collaboration22/100
The snapshot records only 2 distinct coauthors across the reviewed work, yielding a low collaboration score of 22. The recurring partners are Dongha Kim and Wooyoung Shin, and both cataloged 2026 preprints list Park as last author without corresponding status — a compact, theory-oriented team pattern rather than a broad collaboration network.
Activity—
Temporal activity data is insufficient: the snapshot covers only two 2026 arXiv preprints. This axis would measure yearly output volume and the continuity of research activity across the career timeline, both currently unknown.
Reproducibility55/100
The recorded reproducibility mean across 2 assessed papers is 55, and both cataloged preprints carry medium readiness. The SCAD-distillation paper is judged rigorous but fixed-p only, with code not yet public; the density-ratio rescoring paper reports thorough, honest benchmarks whose gains over the strongest baselines are not significant. Public code release would likely lift this score.
Research summary
Seunghwan Park is a researcher at Kangwon National University whose reviewed work sits squarely in statistics and statistical machine learning. The Evolve catalog currently holds two 2026 arXiv preprints on which Park is the last-listed author, and the recorded field profile splits evenly at 50/50 between robust regression with variable selection and imbalanced classification. The first, Sparse Regression Distilled from a Single Robust Fit, distills the fitted surface of a robust fit under a SCAD penalty, with conditional breakdown transfer and oracle-branch theory; the verdict calls it rigorous but fixed-p only, with code not yet public. The second, Density-Ratio Rescoring for Imbalanced Classification Using Raking Duals and Classifier Scores, fuses a raking dual score with a classifier score to improve ranking under class imbalance; it is judged thorough with honest benchmarks, though gains over the strongest baselines are not significant. Both preprints carry medium reproducibility readiness, with a recorded reproducibility mean of 55 across 2 assessed papers. The collaboration footprint is small — 2 distinct coauthors, Dongha Kim and Wooyoung Shin — yielding a collaboration score of 22, consistent with a compact, theory-driven working group.
Collaboration pattern
The collaboration footprint in the snapshot is compact: only 2 distinct coauthors are recorded, and the recurring partners are Dongha Kim and Wooyoung Shin. Both cataloged 2026 preprints place Park in the last author position without corresponding-author status, a pattern typical of a senior or supervisory contributor within a small theory-oriented group rather than a large distributed collaboration network.
Research keywords
Highlights
SCAD distillation turns a single robust fit into sparse regression
Density-ratio rescoring for imbalanced ranking
Timeline
- 2026Posted the arXiv preprint "Sparse Regression Distilled from a Single Robust Fit" as last author.
- 2026Posted the arXiv preprint "Density-Ratio Rescoring for Imbalanced Classification Using Raking Duals and Classifier Scores" as last author.
- 2026Both preprints entered the Evolve review catalog with medium reproducibility readiness (reproducibility mean 55 across 2 papers).
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.
- Bibliometric dimensions are computed from a limited OpenAlex-based snapshot; the output, impact, and activity axes lack underlying data and are unscored.
- The Evolve catalog covers only two 2026 arXiv preprints for this researcher; the full publication record may be broader than what is reviewed here.
- "Seunghwan Park" is a common Korean name; this profile is name-resolved from public bibliographic data anchored to ORCID 0000-0001-7753-085X.
- Neither reviewed paper had publicly confirmed code at review time; both are rated medium reproducibility readiness.