scimapR

The reproducible, equity-aware, question-driven, AI-assisted bibliometric toolkit for working biomedical researchers.

scimapR is a comprehensive R package for bibliometric and scientometric analysis. It provides unified ingestion from 12+ bibliographic sources, classical and modern science-mapping analytics, embedding-based research cluster discovery, and a publication-ready Shiny application – all in one CRAN-compatible package with tibble-based outputs and viridis-themed visualisations.

What makes scimapR distinctive

• Live corpus refresh. Your corpus knows when it was last refreshed, what is stale, and how to update itself (sm_refresh(), sm_staleness(), sm_lock()).
• Research questions as first-class objects. Build structured PICO/PECO questions, auto-generate search queries, and screen with optional LLM grounding (sm_question(), sm_screen_against_question()).
• Reproducible-by-construction corpus certificates. A YAML document that another researcher can use to re-derive your exact corpus (sm_certificate(), sm_rebuild_from_cert()).
• Author trajectory analysis. Track topic pivots, collaborator turnover, and productivity curves across a career (sm_author_trajectory()).
• Equity and representation auditing. Geographic, gender, funding, and OA audits with built-in confidence reporting and epistemic humility (sm_audit_geographic(), sm_audit_gender()).
• LLM-grounded corpus chat. Ask questions about your corpus with every claim anchored to actual works – no hallucinated references (sm_chat()).

Relationship to bibliometrix

scimapR is inspired by and designed as a complement to the excellent bibliometrix package by Massimo Aria and Corrado Cuccurullo (2017, Journal of Informetrics, doi:10.1016/j.joi.2017.08.007).

bibliometrix is the foundational R package for science mapping. It pioneered many of the analyses that scimapR also provides. scimapR is not a fork, not a derivative, and contains no code copied or adapted from bibliometrix. For shared bibliographic formats, scimapR ships clean-room parsers written from public format specifications.

First-class round-trip interop is provided:

M <- sm_to_bibliometrix(corpus)            # use with bibliometrix
corpus <- as_sm_corpus(M)                  # come back to scimapR

See vignette("relationship-to-bibliometrix") for details.

Installation

# Install from GitHub (development version)
# install.packages("pak")
pak::pak("CTTIR/scimapR")

Quick example

library(scimapR)

# Generate a synthetic corpus
corpus <- sm_example_corpus(n_works = 100, seed = 42)
print(corpus)
#> 
#> ── <sm_corpus> ─────────────────────────────────────────────────────────────────
#> Works: 100 | Authors: 80 | Institutions: 0
#> Years: 2015 - 2024
#> Sources (journals): 10
#> Embeddings: 100 x 64
#> Provenance: synthetic (100)
#> Status: Unlocked (last refreshed: 2026-05-09 19:13:44)

# Visualise production
sm_plot_production(corpus)

Feature overview

Module	Functions
File ingestion	sm_read_bib(), sm_read_ris(), sm_read_wos(), sm_read_scopus(), sm_read_pubmed_xml(), …
API ingestion	sm_fetch_openalex(), sm_fetch_crossref(), sm_fetch_pubmed(), sm_fetch_semantic_scholar(), …
Enrichment	sm_enrich_unpaywall(), sm_enrich_altmetric(), sm_enrich_concepts(), …
Networks	sm_network_citation(), sm_network_cocitation(), sm_network_coupling(), sm_network_collab(), sm_network_coword()
Embeddings	sm_embed_works(), sm_cluster_hdbscan(), sm_cluster_leiden(), sm_cluster_label()
Indicators	sm_metric_h_index(), sm_metric_disruption(), sm_metric_rcr(), sm_metric_fnci(), sm_metric_novelty()
Visualisation	sm_plot_landscape(), sm_plot_thematic_map(), sm_plot_production(), sm_plot_equity_dashboard(), …
Export	sm_export_figure(), sm_export_table(), sm_export_zip(), sm_export_gephi()
Shiny app	sm_run_app()

Documentation

• vignette("scimapR") – Getting started
• vignette("ingestion") – Building a corpus
• vignette("embeddings-and-clusters") – Semantic landscape
• vignette("modern-indicators") – CD/RCR/FNCI/novelty
• vignette("question-driven-reviews") – Research questions + screening
• vignette("reproducibility-and-certificates") – Corpus certificates
• vignette("equity-trajectory-and-chat") – Equity audit + trajectories
• vignette("relationship-to-bibliometrix") – Interop and credit

Acknowledgements

scimapR stands on the shoulders of the bibliometrix project. We are deeply grateful to Massimo Aria and Corrado Cuccurullo for creating the foundational R package for science mapping, and for their landmark 2017 paper which defined the field of R-based bibliometrics.

We also acknowledge the many data sources that make scimapR possible: OpenAlex, Crossref, PubMed, Semantic Scholar, Unpaywall, and others.

Citation

If you use scimapR in your research, please cite both scimapR and the foundational bibliometrix package:

citation("scimapR")

BibTeX entries:

@Manual{scimapR,
  title = {scimapR: Reproducible, Question-Driven, Embedding-Aware Science Mapping},
  author = {Raban Heller},
  year = {2026},
  note = {R package version 0.1.0},
  url = {https://github.com/CTTIR/scimapR},
}

@Article{bibliometrix,
  title = {bibliometrix: An R-tool for comprehensive science mapping analysis},
  author = {Massimo Aria and Corrado Cuccurullo},
  journal = {Journal of Informetrics},
  year = {2017},
  volume = {11},
  number = {4},
  pages = {959--975},
  doi = {10.1016/j.joi.2017.08.007},
}

For a complete citation block including each data source used in your corpus, run sm_cite_corpus(your_corpus).

License

MIT

Contributing

Issues and pull requests are welcome at github.com/CTTIR/scimapR.