Computational social scientist at Johns Hopkins working on online hostility, communal knowledge, and multilingual NLP
3
papers here
3
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
11
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 current data lacks sufficient volume metrics for this axis. Output would measure total and per-year publication production relative to peers; only partial per-year counts are available, so no score is given.
Impact—
Not yet scored: citation-based impact data is insufficient here. This dimension would measure influence through citations, citing works, and field-normalized percentiles; no such figures accompany this snapshot, so the axis is left unscored rather than estimated.
Focus0/100
The focus score is low (0) because the field distribution splits evenly into thirds: computational social science with computational sociolinguistics, computational social science with NLP, and NLP itself, at 0.33 each. Topic counts (sentiment analysis 18, text-analysis methods 15, misinformation 11, hate speech 11, social media and politics 10) suggest one coherent social-media research program rather than scattered work, but the axis captures the even spread.
Collaboration44/100
The collaboration score is moderate (44). In the last three years this profile lists 20 works with 8 distinct coauthors, and on the three Evolve-reviewed papers the researcher appears as a middle or last author rather than corresponding author. Recurring collaborators include Sagar Kumar, Brooke Foucault Welles, Nicholas W. Landry, Kristina Lerman, and Patrick Gerard, indicating sustained small-team work rather than large consortia.
Activity89/100
The activity score is high (89). OpenAlex counts by year show growth from 1 work in 2017 and 2 in 2018 to 6 in 2021, 5 in 2022, 3 in 2023, 4 in 2024, and 8 in each of 2025 and 2026, with 20 works in the most recent three years. The three 2026 Evolve-reviewed papers indicate the upward trend is still continuing.
Reproducibility85/100
The reproducibility score is high (85), averaged across 3 Evolve-reviewed papers. Two cataloged papers — the Bluesky Starter Pack common-ground study and the TurEngMix corpus benchmark — carry high reproducibility readiness, while the intergroup hostility unification paper is rated medium. Evolve's verdicts also flag specific limitations, such as GPT-4o-as-filter circularity and small minority classes in TurEngMix.
Research summary
Julia Mendelsohn is a computational social scientist and NLP researcher at Johns Hopkins University whose OpenAlex-indexed work divides almost evenly across computational social science with computational sociolinguistics, computational social science with NLP, and natural language processing — 0.33 each. The bibliometric snapshot reports an h-index of 11, an i10-index of 11, and 20 works in the most recent three years, with annual counts rising from 1 in 2017 and 2 in 2018 to 8 in each of 2025 and 2026, supporting a high activity score of 89. Leading topics include sentiment analysis and opinion mining (18 works), computational and text analysis methods (15), misinformation and its impacts (11), hate speech and cyberbullying detection (11), and social media and politics (10), sketching a research program on hostile and misleading online discourse. Three Evolve-reviewed 2026 papers illustrate its range: a framework unifying six theories of intergroup hostility whose threat pathway outperformed 105 rival structures; evidence that communal common ground in Bluesky Starter Packs is graded by shared community affiliation; and TurEngMix, presented as the first Turkish-English code-mixed benchmark for language identification and named entity recognition. Evolve's reproducibility assessment averages 85 across these three papers. Output and citation-impact scores are not yet available in the current data.
Collaboration pattern
Collaboration is moderate (score 44): 20 works in the last three years involve 8 distinct coauthors, and on the three Evolve-reviewed papers Mendelsohn appears as a middle or last author, not corresponding. Recurring partners include Sagar Kumar, Lawrence Swaminathan Xavier Prince, Brooke Foucault Welles, Nicholas W. Landry, Patrick Gerard, Kristina Lerman, Ilayda Dogan, and Phuong-Anh Nguyen-Le — consistent with sustained small-team projects bridging NLP and computational social science groups.
Research keywords
Highlights
Unifying six theories of intergroup hostility into one empirical system
Graded communal common ground in Bluesky Starter Packs
First Turkish-English code-mixed benchmark for language ID and NER
Timeline
- 2017Earliest indexed works appear in the OpenAlex counts: 1 work in 2017 and 2 in 2018.
- 2021Output accelerates to 6 indexed works in 2021, followed by 5 in 2022, the strongest run before 2025.
- 2025Indexed works reach 8, opening a recent three-year window totaling 20 works.
- 2026Three Evolve-reviewed papers appear: the intergroup hostility unification, the Bluesky Starter Pack common-ground study, and the TurEngMix corpus 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 figures (h-index, yearly counts, topics) come from an OpenAlex name-resolved snapshot; output and impact axes are unscored because citation data is missing.
- 2026 counts reflect a partial year as of October 2026.
- The reproducibility score averages only 3 Evolve-reviewed papers, all 2026 arXiv preprints.
- Author-position and corresponding-author information is limited to the 3 cataloged papers.