Computational finance researcher building friction-aware LLM trading evaluation and Ethereum DeFi benchmarks.
2
papers here
2
reviewed by Evolve
2
first-author
1
corresponding
—
works · OpenAlex
5
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—
Not yet scorable: no cumulative publication total is available beyond the recent three-year window of 13 works. This axis would measure lifetime publication volume, yearly consistency, and venue spread.
Impact—
Not yet scorable: per-paper and aggregate citation data are unavailable beyond h-index 5 and i10-index 2. This axis would measure citation accumulation, highly-cited outputs, and downstream influence on the field.
Focus0/100
The focus score of 0 reflects an evenly split agenda: field distribution is 50% machine learning for blockchain/DeFi analytics and 50% financial NLP/computational finance. Topic counts span five clusters — stock market forecasting (12), financial markets and investment strategies (11), sentiment analysis (7), corporate finance and governance (2), and blockchain applications (2) — indicating breadth across two adjacent but distinct areas rather than a single narrow specialization.
Collaboration14/100
Collaboration scores 14, the weakest axis: the recent three-year window records only 1 distinct co-author across 13 works, with Carsten Maple the recurring partner. Both catalog papers are first-authored, one also corresponding. This solo-led pattern is typical of early independent work but limits external validation and network reach relative to peers with multi-institution teams.
Activity77/100
Activity is the strongest axis at 77: OpenAlex records 13 works in the recent three-year window, with yearly counts rising from 3 in 2024 and 3 in 2025 to 7 in 2026 — more than doubling in the most recent year. Both Evolve-reviewed papers appeared in 2026, confirming the acceleration is driven by first-authored arXiv output in financial NLP and DeFi analytics.
Reproducibility55/100
Reproducibility scores 55 — the reproMean across 2 reviewed papers, each rated medium readiness. ETH-TraceBench publishes a 1.35-billion-event benchmark with quantified shift drops of 0.16–0.19 macro-F1 on unseen protocols, aiding replication; the MFAST paper applies strong contamination controls but its headline AUC 0.846 and net Sharpe 2.85 have not been independently validated.
Research summary
Kemal Kirtac is an early-career researcher at the University of Warwick working at the intersection of computational finance and machine learning for blockchain analytics. OpenAlex bibliometric data shows a focused but fast-moving output profile: 13 works in the last three years, accelerating to 7 in 2026, with an h-index of 5 and i10-index of 2. Topic modeling places the core in stock market forecasting methods (12 publications), financial markets and investment strategies (11), and sentiment analysis and opinion mining (7), with smaller threads in corporate finance and governance and blockchain applications. Evolve has reviewed two 2026 first-authored arXiv papers. The first introduces MFAST, a friction-aware evaluation framework for financial language models used in news-based trading, reporting AUC 0.846 and a net Sharpe ratio of 2.85 — headline numbers that Evolve flags as straining credibility. The second, ETH-TraceBench, contributes a 1.35-billion-event Ethereum benchmark showing that simple models reach 0.95–0.96 macro-F1 but lose 0.16–0.19 macro-F1 under protocol shift. Field distribution is evenly split between financial NLP and blockchain/DeFi analytics. Current collaboration is narrow — one distinct co-author in the recent window — and reproducibility readiness is moderate, with medium ratings on both reviewed papers.
Collaboration pattern
Collaboration is currently narrow: OpenAlex records just 1 distinct co-author in the recent three-year window, with Carsten Maple the recurring partner. Both catalog papers are first-authored, and the MFAST paper also lists this author as corresponding. The collaboration axis score of 14 reflects this solo-led profile; broadening the co-author network is the clearest growth opportunity as the research program expands.
Research keywords
Highlights
MFAST: friction-aware evaluation of financial language models for news-based trading
ETH-TraceBench: a 1.35-billion-event Ethereum DeFi benchmark
Timeline
- 20243 works recorded in OpenAlex countsByYear — the earliest activity in the available window.
- 20253 works recorded, matching the prior year's output level.
- 20267 works recorded — output more than doubles; both Evolve-reviewed first-authored arXiv papers appear: the MFAST financial language model paper (also corresponding author) and the ETH-TraceBench Ethereum DeFi benchmark.
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 indicators (h-index 5, i10-index 2, countsByYear, topics) come from an OpenAlex-sourced snapshot; per-paper citation impact and cumulative output totals are not available.
- Qualitative assessments draw only on 2 Evolve-reviewed papers, both 2026 arXiv preprints; peer-reviewed venue publications are not reflected in this profile.
- Output and impact radar axes are unscored because the underlying data is insufficient.
- Identity is name-resolved from public bibliographic data (OpenAlex, ORCID 0009-0007-2044-1830); the name may not be unique.