Oshani Seneviratne

Rensselaer Polytechnic Institute · United States

Unclaimed · AI-structured profileORCID OpenAlex

Rensselaer semantic-web researcher working on federated knowledge graphs, KGQA, and streaming entity integration

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) 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 score yet: the underlying output measure — publication volume normalized against field and career-stage baselines — could not be computed from this OpenAlex snapshot, even though raw annual counts exist. This axis would measure sustained, field-adjusted research production.

Impact—

No score yet: citation-based impact could not be computed in this snapshot, so the axis is left unscored rather than estimated. It would measure field-normalized citation influence — how often this researcher's works are cited relative to venue and year expectations.

Focus0/100

Focus scores 0 because OpenAlex topic counts spread almost evenly across Semantic Web and Ontologies (42 works) and Blockchain Technology Applications and Security (41), with Topic Modeling, Privacy-Preserving Technologies in Data, and Ethics and Social Impacts of AI at 15 each; the field distribution likewise splits three ways at about one third each.

Collaboration22/100

Collaboration scores 22 in this snapshot: only 2 distinct recent coauthors are recorded, with Md Saikat Islam Khan Bappy and Jui-Chien Lin the most frequent partners. All three cataloged 2026 papers place Seneviratne in the last author position, consistent with a compact, supervisory-led group rather than large consortia.

Activity100/100

Activity scores a full 100: 51 works in the most recent three-year window, with countsByYear of 16 in 2019, 12 in 2020, 14 in 2021, 10 in 2022, 21 in 2023, and 15 in 2024, plus 19 in 2025 and 17 indexed so far for 2026 — sustained high output across eight consecutive years.

Reproducibility70/100

Reproducibility scores 70 (reproMean across 3 assessed papers). Evolve reviewers rated the vertical-partitioning cost study as high readiness and both FedV-KGQA and the SNAP-KG homophily study as medium, repeatedly crediting honest scoping while noting missing significance tests and small relation vocabularies.

Research summary

Oshani Seneviratne is a computer scientist at Rensselaer Polytechnic Institute (US) whose OpenAlex profile rests on two long-running threads: Semantic Web and Ontologies (42 works) and Blockchain Technology Applications and Security (41), supported by secondary clusters in Topic Modeling, Privacy-Preserving Technologies in Data, and Ethics and Social Impacts of AI (15 works each). The current Evolve catalog draws from her recent semantic-web line: three 2026 arXiv papers, each with Seneviratne in the last author position, covering cost characterization of vertically partitioned federated knowledge graphs, the FedV-KGQA federated question-answering prototype presented as an ISWC poster, and a homophily-boundary study of SNAP-KG for streaming entity integration. Activity scores 100, with 51 works in the most recent three-year window; countsByYear climbs from 16 works in 2019 to a 21-work peak in 2023 and holds at 15-19 through 2024-2026. Her bibliometric standing includes an h-index of 16 and an i10-index of 33. Reproducibility scores 70 (mean across 3 assessed papers), matching Evolve verdicts that praise careful, honest scoping while flagging limited vocabularies and missing significance tests. Focus scores 0, signaling deliberately broad interests, and collaboration scores 22 with only 2 distinct recent coauthors, consistent with a compact working group.

Collaboration pattern

Recent collaboration in this snapshot is deliberately compact: 2 distinct coauthors are recorded, with Md Saikat Islam Khan Bappy and Jui-Chien Lin the recurring partners, yielding a collaboration score of 22. Across all three cataloged 2026 papers Seneviratne sits in the last author position without a corresponding-author designation, a pattern consistent with supervising a small working group on federated knowledge graphs and streaming entity integration. Larger collaborative networks may exist in earlier work not covered by this extract.

Research keywords

federated knowledge graphsvertical partitioningfederated question answeringknowledge graph question answeringgraph partitioningfederated knowledge graph embeddingmulti-hop question answeringheterophilymulti-view graph clusteringgraph neural networkssemantic webblockchain technology applications

Highlights

Cost map for vertically partitioned federated knowledge graphs
As last author, Seneviratne's team compares four vertical KG-partitioning strategies across five cost dimensions and shows three of the costs are invariant to the partitioning choice; Evolve credits a careful, honest study while noting tiny relation vocabularies limit reach. Reproducibility readiness: high.
FedV-KGQA: federated KG question answering nears centralized accuracy
An ISWC poster distilling design lessons from FedV-KGQA alongside an interactive prototype: federated fusion approaches centralized accuracy on three KGQA benchmarks; reviewers note honest scope but the absence of significance tests or a formal privacy proof. Reproducibility readiness: medium.
Charting the homophily boundary of SNAP-KG
A companion scope study shows SNAP-KG and homophily-assuming baselines collapse on heterophilous graphs during streaming entity integration; Evolve credits honest analysis with no new method proposed, and flags the roadmap as untested. Reproducibility readiness: medium.

Timeline

2026Releases a companion scope study charting the homophily boundary of SNAP-KG for streaming entity integration on arXiv.
2026Presents FedV-KGQA in Practice, an ISWC poster sharing design lessons and an interactive prototype for federated knowledge graph question answering.
2026Publishes a cost characterization of vertically partitioned federated knowledge graphs on arXiv, in the last author position.

Papers in the Evolve catalog

Frequent co-authors

Md Saikat Islam Khan Bappy×2 Jui-Chien Lin×1

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.

All bibliometric indicators are computed from OpenAlex only; coverage may differ from other scholarly databases.
Output and impact axis scores are unavailable in this snapshot and are left unscored rather than estimated.
Evolve's review coverage is limited to three 2026 arXiv preprints, all with last-author position; earlier and peer-reviewed work was not independently reviewed here.
Identity resolved via ORCID 0000-0001-8518-917X and OpenAlex; 2026 counts are partial as indexing continues.