Usage guide

How to use every tool on the platform — bibliography check, citation verification, citation repair, manuscript formatting, journal finder & check, thesis assistant, author search, and the alert suite.

1. Check bibliography

Paste a reference list (or upload a .txt/.docx/.pdf) and we verify each entry against scholarly databases — catching fabricated, mismatched, retracted or wrongly-formatted citations.

Paste your references, or upload a file — from a full paper we auto-detect the bibliography section. .bib/.ris imports are also supported.
Pick a citation style (APA 7, IEEE, Vancouver, MLA or a CSL style).
Click “Check”. Each reference is checked for reality (DOI / year / authors / journal), for formatting, and against the retraction database.
Read the result: a colored badge (verified / suspicious / not found) and a 0–100 confidence score per entry.
Click any citation to see the matched source record and where it was found.
Copy the “Repaired bibliography” (complete, in-order list — verified entries rewritten, the rest kept as-is and numbered for manual checking) or download it as Word.
About to submit? Use the “Pre-submission check” card: enter the target journal and get a single Word report combining bibliography health, retractions and the journal's reliability.

Sources covered: journal articles, conference papers, Turkish national journals (TR Dizin, İSAM, SOBIAD), education research (ERIC), books (TO-KAT union catalog / Google Books / OpenLibrary), web sources (page-title confirmation), arXiv preprints, DOI / PubMed / Europe PMC / OpenAlex / Crossref / DBLP / DOAJ records. Thesis citations are checked in SOBIAD (which indexes the YÖK National Thesis Center) and, if not found, routed to the YÖK National Thesis Center.
Author ORDER is checked too: you are warned when the authors in a reference are swapped. Indexes sometimes carry the order reversed; when records disagree, the publisher's own page wins.
Limits — Guest: 5 searches/day (5 references/search). Member: first day unlimited, then 20 references/day, up to 75/month (20 per search). Pro: unlimited.
A low score does not always mean fabricated — very new preprints can be under-indexed; click through and verify manually.
Tick “save to my account” to keep the report on My activity — saved reports are then watched daily for later retractions, you can share them via link, and re-runs show what changed.

2. Verify citations

One step beyond the reference check: even when a source is real, does the sentence citing it actually match what the source says? Each citation is marked Full / Partial / Unsupported / Could-not-verify.

Open “Verify citations” and paste your text (or upload the file) together with its bibliography.
If you have the full text of closed-access sources, upload it (PDF, Word .docx, .rtf or .txt) — the text is extracted IN YOUR BROWSER and auto-matched to the right reference, so subscription-only sources get verified with full text too.
Glance at the matching table: it shows which file was linked to which reference and why (DOI, or title/author). Fix any wrong link with one click.
Start the run. The bibliography is verified first, then your text is split into claims and each claim is compared, by meaning, to the cited source's abstract or full text.
Watch progress — a time estimate is shown; expect 1–5 minutes depending on the number of sources.
Read the report: per-citation verdicts with evidence, plus a corrected bibliography. When the evidence comes from a file you uploaded it is labelled as such, and for PDFs the page is shown too (“your file · p. 7”), so you can confirm it in seconds. Download as Word or CSV, or share it via link.

A Turkish sentence can be matched to an English source — comparison is at the meaning level.
“Could not verify” means the source's text could not be reached (paywall); it does not mean the citation is wrong.
If a file cannot be linked to any reference with confidence it is never assigned at random: it is set aside and you are asked to pick among the three most likely candidates — a wrongly matched full text is more harmful than a missing one.
If a claim is not found in the full text you uploaded, the result is not lumped in with “could not be reached”; it is reported separately as “Not found in the file you uploaded”.
Uploaded file contents are never dumped on screen and never stored anywhere. To re-run a saved report later, you will need to upload the files again.
Your first day as a member is unlimited; afterwards a shared daily pool of 20 citations applies. Pro members get 250 citations a month (Max: 1,000); academic e-mail addresses (.edu / .edu.xx / .ac.xx) are exempt from the shared-network limit.

3. Find scientific evidence

You have the sentence but not the source. This tool finds a publication that ACTUALLY supports the claim: the citation, verbatim evidence in the source's own words, and wording the source can safely support.

Paste the sentence you need a source for — with its population, the relationship, and any number (one sentence is enough).
Pick a citation style (APA 7, IEEE, Vancouver, TÜBİTAK, TÜSEB/TBYS) and, if it matters, whether the source should be international or Turkish-language (“No preference” is the default).
Run it: scholarly indexes are searched, reachable full texts are opened and the relevant passages read. The stage you are in is shown live; a run typically takes 20–45 seconds.
Read the result: the citation, the VERBATIM evidence quote, the evidence level (full text / abstract) and where the evidence came from.
Use the “wording the source supports”: if your sentence claims more than the source does, you get it rewritten down to what the source actually establishes.
Copy the citation and the in-text label, download the result as Word, and — if you are signed in — tick “save to my account” to keep it on My activity.
Want to see the output first? “See a sample result” makes no request and costs no quota (direct link: /en/find-scientific-evidence?sample=1).

When no supporting quote exists, nothing is invented: you are told “no source supports this sentence as written” and shown the closest sources with what each one actually establishes — so you can rewrite the sentence or narrow the claim.
Turkish claims also search TR Dizin and the Turkish national literature; a Turkish sentence can match an English source — comparison is at the meaning level.
Retracted publications are NEVER offered as support (checked separately against our Retraction Watch extract).
The tool does not generate citations; it finds a sentence that genuinely appears in an existing publication. Still open the source yourself before submitting.
Limits — Guest: 3 runs/month. Member: 5/month. Campus-licensed institutions: 30/month. Pro: unlimited.

4. Convert manuscript format

Convert an entire Word manuscript to a major publisher's general submission template — document layout, typography, headings, in-text citations and references together.

Open “Manuscript format” and upload the editable .docx manuscript.
Choose a general Springer Nature, Elsevier, IEEE, Wiley, Taylor & Francis or SAGE submission template and the article type.
Convert: page layout, font, spacing, columns and heading styles are applied; in-text citations and references are verified and rewritten in the profile's citation style.
Download the converted .docx and review the target journal's current Guide for Authors before submission.
For an exact journal-specific format, request the separately priced custom service from the Contact page.

Publisher profiles are general templates; individual journals and article types may require different title pages, section order, declarations and figure placement.
Full conversion requires .docx. PDF can be inspected, but cannot reliably preserve editable document structure.
Free members have a monthly conversion quota; Pro is unlimited.

5. Find a journal

Evidence-based journal suggestions for your manuscript: we look at where similar publications actually appeared, not at keyword lists.

Enter your paper's title (an abstract improves the match). For Turkish-language titles, journals published in Türkiye are pulled into the pool as well.
Use the DECISION BAR at the top of the results in the order academics actually decide: indexing first (SCI-E/SSCI/AHCI · all WoS incl. ESCI · Scopus · TR Dizin · UBYT), then “minimum quartile” and the quartile source (Scopus CiteScore / WoS JCR), then access and APC.
Read each suggestion with its justification: “published N papers like yours in the last 3 years” is evidence; a “Scope match” badge means NO similar publication was found there and the journal was suggested only because its subject profile is close — a weaker signal.
If the filters empty the list, look at “Journals closest to your filters”: each one is listed with the condition that excluded it (quartile / indexing / access / APC / risk).
Check the money: journals badged “may be covered by a TÜBİTAK agreement” can let you publish open access without paying an APC, and Diamond journals charge nothing at all.
Compare the process: the “~N weeks” badge and the “shortest process” sort show submission-to-publication time.
Run “Check a journal” on your shortlist before submitting; you can export the list as CSV to share with your supervisor.

Suggestions come from OpenAlex publication data joined with Scopus CiteScore and Web of Science JCR quartiles, WoS index membership and the TÜBİTAK ULAKBİM UBYT incentive list.
ESCI is a separate option: it is a large part of Web of Science but is not counted the same as SCI-E, SSCI and AHCI in Turkish promotion criteria. Results are grouped accordingly.
By default the quartile filter uses the BETTER of the Scopus and WoS quartiles; pin it to whichever your institution recognises via the “quartile source” box. The same journal can sit in different quartiles in the two sources.
A TÜBİTAK waiver is not automatic: the annual quota can run out, your institution must be an EKUAL member, you must be the corresponding author and the article type must qualify (cabim.ulakbim.gov.tr).
Turnaround is the journal's OWN declaration to DOAJ, not a measurement — publishers largely do not share acceptance dates. Use it to compare, not as a commitment.
The match score is absolute (the top journal does not automatically get 100) and is not a probability of acceptance: it combines how much of the journal's output is in your topic, how many papers in your topic were found, and how recent they are.
“Hide risky journals” removes journals carrying a hijacked-clone record, heavy retractions or a DOAJ removal signal.
Sign-in is required; free members get 10 searches per month, Pro is unlimited.

6. Check a journal

Vet a journal before you submit: is it actually indexed in Scopus and Web of Science, in which quartile, has it been discontinued, and what explainable publishing-risk signals does it carry?

Enter the journal's name or ISSN. Use the ISSN for the most precise result — different journals can share a title.
Look at COVERAGE first: is it indexed in Scopus, and in which Web of Science index (SCIE/SSCI/AHCI/ESCI)? That question comes before “which quartile?”.
Then the QUARTILES: the Scopus quartile (CiteScore) and the JCR quartile are shown separately and can differ for the same journal. The per-category breakdown is included.
Read the TÜBİTAK UBYT incentive amount, the DOAJ / ISSN-registry / TR Dizin checks and the red-flag list.
For a Turkish associate-professorship application, open the “Doçentlik değerlendirmesi” section: the YÖK predatory-journal decision is applied to this journal with the quartile pre-filled.

MOST IMPORTANT WARNING: if you see “this quartile may no longer hold”, be careful — the journal shows a strong quartile but the source database no longer lists it. Quartile lists keep carrying the old value for a while after a journal is dropped.
A journal “discontinued in Scopus” is not necessarily predatory (it may have ceased publishing), but NEW articles published there will not be indexed in Scopus — decisive for incentives and promotion files.
A journal can sit in different quartiles across subject categories; check which category applies to your submission.
Red flags include: unregistered ISSN, high APC with no DOAJ listing, low impact with high APC, sudden volume spikes, very young journals with high volume, a documented hijacked-clone site, a high retraction rate.
A flag is a signal, not a verdict — combine it with your field's own lists and the library's guidance.
If you are an editor and something here is wrong, write to us from the contact link at the bottom of the page — we review and correct it.

7. Check a conference

The call for papers says “Scopus indexed” — is it true? Check whether the proceedings are really covered, year by year, against Elsevier's official Scopus source list.

Search the ACRONYM (e.g. IMCET, ISCTURKEY) — long names change every year, acronyms do not.
Add a year if you care about one edition (“ISCTURKEY 2022”); the verdict is then given for that year.
If you have the ISBN of an earlier proceedings volume, search that — it is the most precise match.
Read the verdict: “Yes — indexed in Scopus (2022 proceedings)”, “indexed, but not the year you asked”, or “No — no Scopus-indexed record”.
Compare the listed volumes (year, title, publisher, ISBN) with what the organiser claims.

SCOPUS INDEXES THE SOURCE THAT PUBLISHES THE PROCEEDINGS, NOT THE EVENT. There is no permanent “this conference is Scopus indexed” label: one edition may be covered and the next not. Always decide per year.
Proceedings of an upcoming edition are not published yet, so they cannot be in the list — “not found” alone does not mean “will not be indexed”. Ask the organiser for previous editions and verify them here.
If your query is only generic words (“international engineering congress”), no verdict is given — thousands of conferences share them. Similar conferences are listed separately, but they are not the answer: search the acronym or ISBN.
Some large conferences are covered as a SERIES rather than volume by volume; then we say the series is covered, which does not verify this conference's own volume.
Data source: Elsevier's official Scopus source list (updated a few times a year), over 177,000 proceedings volumes. The list date is shown on the page.

8. Thesis / Paper Assistant

Give a topic and get a cited Introduction, Literature Review, or research-Gap analysis — grounded in the abstracts of ~50 scanned publications.

Type your topic (sloppy keywords are fine — we turn them into academic search queries).
Choose the mode: Introduction, Literature review, or Gap analysis.
Optional: add a key paper’s DOI/title to also scan its citation neighborhood (seed + snowball).
Click Generate. A fluent, cited draft is written from the publications’ abstracts.
Click any [citation] to see the exact abstract it relies on — every sentence is grounded.
Pick a citation style; the reference list is built automatically from only the cited works.
Run “Verify & fix references” to check each source and correct its metadata.

Sign-in is required for this tool.
This is a rapid draft from abstracts, not a systematic (PRISMA) review — verify sources before use.
The in-text citation label and the reference entry always come from the same (verified) record.

9. Find by author

Search an author’s name and list their publications, ready to copy as formatted references.

Type the author’s name (and, if helpful, a topic to narrow it).
Search; matching publications are listed — pick the right person by institution, ORCID and publication count.
Pick a citation style and copy the formatted references you need.
If it is a researcher you want to keep up with, hit “Follow this author” — you get a weekly email when new work appears.

Useful for building a reference list from a known author quickly.
Combine with “Check bibliography” to verify the entries you collect.
Following is limited to 3 authors on the free plan and 10 on Pro; manage the list under My account → Alerts.
The second tab of the same screen is for EDITORS: “Find reviewers” takes the manuscript topic and lists candidates who published on it recently, with ORCID, sample publications and conflict-of-interest warnings (1 trial for guests, 5/month for members).

10. Alerts: citations, topics, authors and retractions

Free weekly email alerts for academic updates — set them up under My account → Alerts.

Citation alert: add your ORCID; when your publications receive new citations you get a weekly email. Every citation is vetted — the citing journal's quartile (Scopus/WoS), predatory-journal signals and self-citations are flagged.
Topic alert: enter up to 3 topics; when new publications appear in that area you get a weekly email.
Author alert: follow a researcher from the “Find by author” screen; you get a weekly email when they publish (3 authors free, 10 on Pro).
Retraction tracking: automatic — if a source in a saved report, or an article citing your work, is later retracted, you get a separate notice.

The citation alert is ORCID-based, so there are no name mix-ups; only indexed, verifiable citations are tracked, which is why counts may be lower than Google Scholar.
The first citation-alert scan sets a baseline; emails start the following weeks with NEW citations only.