journal-matcher

Journal Matcher

When to Use

• User pastes title + abstract (or keywords) and asks for journal suggestions
• "Where should I submit this paper?", "Find journals matching my manuscript", "Compare journal options"
• After a rejection when the user wants alternatives

Prerequisites / Inputs

Collect (or ask for):

• Title and abstract (preferred)
• Field / keywords
• Preferences: OA required? Target quartile/IF range? Max APC? Preferred speed? Geographic focus? Article type?
• Any prior rejections or must-avoid journals

Privacy note: Prefer not storing full manuscripts long-term; process ephemerally where possible.

Procedure

1. Normalize input (clean abstract, extract key terms).
2. Query multiple independent sources in parallel where possible:
   • JANE (jane.biosemantics.org) for biomedical
   • Publisher finders: Elsevier Journal Finder, Springer Nature Journal Suggester, Wiley, Taylor & Francis, MDPI, etc.
   • Broader tools / indexes: Paperpile-style, Web of Science / Clarivate if accessible, SCImago, Scopus sources, DOAJ
   • Semantic search / recent similar papers via web or available APIs
3. Aggregate candidates and enrich each with:
   • Scope fit / similarity rationale + example similar papers
   • Indexing (Scopus, WoS, PubMed, DOAJ…)
   • Metrics (SJR, CiteScore, JIF if relevant — present DORA-aware)
   • Reported or estimated review/acceptance timelines, acceptance rates if available
   • OA model + APC / waivers
   • Predatory-risk signals (Think.Check.Submit checklist items, known lists, red flags)
4. Filter and rank according to user preferences + quality signals.
5. Produce a clear shortlist (3–8 journals) with:
   • Ranked table or cards
   • "Why this one" explanations
   • Risks / caveats
   • Direct links to guidelines / submission
   • Suggested next actions (e.g., check special issues, prepare cover letter points)
6. Optionally offer deeper checks or a human-style "expert shortlist" rationale.

Pitfalls & Guardrails

• Publisher tools are biased — always cross-check with independent sources.
• Acceptance rates and real timelines are often incomplete or outdated.
• Interdisciplinary work needs broader search + manual scope reading.
• Never recommend journals that fail basic legitimacy checks.
• Be transparent about data gaps.

Verification / Acceptance

• Shortlist contains at least 3 plausible options with explicit fit reasoning.
• User can act immediately (links + clear next steps).
• Quality flags are visible.

References / Further

• Link to key sites and Think.Check.Submit.
• (Add any local templates or example reports in references/ or templates/)