Computational text and network scientist advancing scientometrics and NLP for scholarly data at UIUC.
3
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
23
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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—
Publication-volume metrics are not available in this snapshot, so this axis is unscored. It would measure cumulative and per-year output across venues. Annual counts exist (20 works in 2024; 19 in 2025), but the full output record is incomplete.
Impact—
Citation-impact aggregates are not yet available, so no score is assigned. This axis would measure influence via citations normalized by field and year. The h-index (23) and i10-index (45) are reported, but underlying citation totals are missing.
Focus100/100
This axis scores 100: her OpenAlex topic profile is tightly concentrated — Computational and Text Analysis Methods (65), Topic Modeling (47), Sentiment Analysis and Opinion Mining (44), Complex Network Analysis Techniques (42), and scientometrics and bibliometrics research (35) — and the field distribution maps to a single field, scientometrics / NLP for scholarly data. These clusters form one coherent computational social-science agenda.
Collaboration54/100
Scored 54, mid-range. In the recent three-year window she has 52 works with 14 distinct co-authors, indicating steady but not large-team collaboration. Recurring partners include Stephen Meisenbacher, Andreea-Elena Bodea, Florian Matthes, Kanyao Han, Zhiwen You, and Jinseok Kim; on the recent catalog papers she appears as last or middle author rather than corresponding author.
Activity100/100
Scored 100. Her recent three-year record lists 52 works, and OpenAlex countsByYear shows sustained momentum: 14 works in 2023, 20 in 2024 (the record's highest), 19 in 2025, and 13 already counted in 2026. Output has been continuous since 2003, with three additional 2026 arXiv preprints in the Evolve catalog.
Reproducibility55/100
Scored 55, from limited coverage: only one catalog paper has been assessed, with a mean repro-readiness of 55 and a medium rating. The 2026 funder-disambiguation paper ships a usable auto-labeling pipeline, but Evolve notes its labels lack manual verification; broader evidence across her portfolio is not yet available.
Research summary
Jana Diesner is a researcher at the University of Illinois Urbana-Champaign whose profile is name-resolved from OpenAlex bibliographic data. Her topic footprint is tightly concentrated in computational and text analysis methods (65), topic modeling (47), sentiment analysis and opinion mining (44), complex network analysis techniques (42), and scientometrics and bibliometrics research (35). The snapshot records an h-index of 23 and an i10-index of 45, with 52 works in the most recent three-year window produced alongside 14 distinct co-authors. Annual output has been continuous since 2003 and remains high: 14 works in 2023, a record-high 20 in 2024, 19 in 2025, and 13 already counted in 2026. Recent Evolve catalog entries show an active line in NLP for scholarly data: she is last author on a 2026 arXiv paper on multi-functional embedding models for funder name disambiguation, where fine-tuned sentence-transformer embeddings beat GPT-5.2-class models on Web of Science records at 0.91 accuracy, though the auto-labels lack manual verification. Two further 2026 preprints — ContextBias on bias persistence in text-to-image models and PrivBench on text-to-text privatization benchmarking — extend her reach into generative-model evaluation and privacy. Collaboration is steady rather than large-team, with recurring partners such as Stephen Meisenbacher, Florian Matthes, Kanyao Han, and Jinseok Kim.
Collaboration pattern
Collaboration is steady but compact: 14 distinct co-authors across 52 works in the most recent three-year window. A recurring cluster includes Stephen Meisenbacher, Andreea-Elena Bodea, Ahmet Bilal Akın, Alexandra Klymenko, and Florian Matthes, alongside collaborators such as Kanyao Han, Zhiwen You, Jinseok Kim, Shaghayegh Kolli, Sina Emami, Moreno D'Incà, and Pouyan Nejadi. On the recent catalog papers she sits in last-author or middle positions without being corresponding author.
Research keywords
Highlights
Fine-tuned embeddings outperform GPT-5.2-class models on funder name disambiguation
ContextBias: controlled evaluation of bias persistence under context shift in text-to-image models
A 2026 arXiv preprint with Diesner in the last-author position, offering a controlled evaluation of how biases in text-to-image models persist when prompts shift context. No Evolve verdict or repro-readiness rating is available yet.
PrivBench: a holistic, modular benchmarking platform for text-to-text privatization
A 2026 arXiv preprint listing Diesner among middle authors, presenting a holistic, modular platform for evaluating text privatization methods. It has not yet received an Evolve verdict, keywords, or repro-readiness assessment.
Timeline
- 2003Earliest year in the OpenAlex countsByYear record, with 1 work; output builds steadily through the 2000s (6 works in 2005, 8 in 2008).
- 2014A productivity peak with 19 recorded works, the highest annual count until 2024.
- 2024Highest annual output in the record: 20 works counted by OpenAlex.
- 202519 works recorded, keeping output near its peak for a second consecutive year.
- 2026Three arXiv preprints in the Evolve catalog — the funder name disambiguation paper (last author), ContextBias, and PrivBench — while OpenAlex already counts 13 works for the 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.
- Bibliometric indicators (h-index, i10-index, countsByYear, topics, co-author counts) come from OpenAlex only; the profile is name-resolved from public bibliographic data.
- Output and impact dimensions are unscored because the underlying publication and citation aggregates are absent from this snapshot.
- The reproducibility score rests on a single assessed paper (reproPapers = 1) and may not represent the full portfolio.
- ContextBias and PrivBench carry no Evolve verdict, keywords, or repro-readiness rating yet.