jev-reviewer
Browser-based systematic-review extraction tool where Jev points at the lines that answer each question and code copies verbatim quotes.
Overview
Jev Reviewer helps with data extraction for systematic reviews. You open a trial report with its supplements, protocol, analysis plan or data tables, then ask what your extraction form needs, such as inclusion criteria for age, how many were randomised or who funded the trial, by voice, typing, a questions file or a template (trial characteristics, RoB 2, ROBINS-I, QUADAS-2, TIDieR, outcome data, transparency). Every answer is a verbatim quote with its file and page, paragraph, row or slide, highlighted in place; you check it, write your value and tick it, and an extraction table exports per study or per quote.
Jev never writes text here: it only points at line ids, and code copies the quote, so nothing is paraphrased. Files are segmented in the browser into sentences and table rows with ids; a screening pass asks per page chunk which line answers each question (with a none option) and whether a passage answers it, then a verify pass asks for the best lines and neighbours whether each line itself answers the question. Many file types are supported, projects can import studies from reference managers, and everything stays in the browser. The list summary notes findings require human review and are not clinical decisions.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which line answers this question (or none)? | — |
| Yes / no | Does this passage answer the question? | yes · no |
| Yes / no | Does this line itself answer the question? | yes · no |
How it works
- Load study files in the browser and segment them into lines with ids
- Screen each chunk with a Choice and Noul per question
- Verify the best lines and neighbours with a Noul each
- Show verbatim quotes with locations for human checking
- Export the extraction table
Strengths
- Quotes are copied, never generated
- Files stay in the browser
- Supports standard review templates and many file types
Caveats
- Every answer still needs human checking
- Not for clinical decisions
- Extraction quality depends on document parsing
Getting started
Use it at https://jevreviewer.xera.ac; the guide walks through a full review.At a glance
- Type
- Project
- Status
- beta
- Language
- JavaScript
- Runtime
- Browser app
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
A systematic-review tool that selects verbatim evidence from papers and supplements for human checking and export.
logicrw/awesome-jev-projectsResearch-document extraction aid where Jev selects and verifies source lines for verbatim quotes; findings require human review and are not clinical decisions.
cobanov/awesome-jevData extraction for systematic reviews, quoted from the papers. Ask a trial report and its supplements your extraction form or a RoB 2, ROBINS-I, QUADAS-2 or TIDieR template; Jev points at the lines, every answer is a verbatim quote with its page, you check it and export the table. Files stay in your browser.
hellogumbo/awesome-jevExtracts data for systematic reviews from a trial report and its supplements, quoted from the paper, against your own form or a RoB 2 template.
MrJev/awesome-jev
Frequently asked questions
What does Jev decide?
Which line in a document answers each extraction question, and verifies that the line itself answers it.
Can it invent answers?
No. Jev only points at line ids; code copies the exact quote.
Are my files uploaded?
No. Files and projects are stored in your browser.
Related projects
Interactive explorer that uses Jev to tag TC39 proposals by adoption path, cognitive overhead, web-compat risk and intent archetype.
ProjectprototypeTypeScriptDaily pipeline that reads Spain's official gazette with Jev, scoring relevance, tagging categories and picking a summary paragraph.
AppprototypeTypeScriptTitle and abstract screening for systematic reviews with Jev: include, exclude or maybe per record, with reasons, PRISMA counts and recall.
ProjectbetaPythonResearch pipeline that scores NYC 311 complaint descriptions with Jev and maps severity and safety signals on a city grid.
ServiceresearchPythonChecks whether each cited paper supports its citing sentence: code fetches papers, Claude proves a quote, Jev scores it, a human decides.
ServiceproductionPublished Jev client and claim check from the Fake/Real evidence-review site, where Jev filters sources and judges each claim.
ProjectproductionTypeScript