Submit documents, run similarity analysis, review AI advisory signals and export signed reports.
Research Integrity & Knowledge Platform
EVIDENCE by WACREN for West and Central Africa
A regional platform for similarity detection, responsible AI review, integrity policies, knowledge resources, training and evidence-based governance.
Multi-institution by design
Human review before decisions
Built for trusted institutional review
14Member countries
1.9B+Documents compared
100+Languages
10Fraud types detected
Why EVIDENCE exists
Research integrity is not a score. It is a shared institutional practice.
Universities need more than a similarity percentage. They need a way to guide students, support supervisors, protect authors, publish clear policies and understand integrity trends across institutions.
EVIDENCE gives WACREN a regional platform where checks, recommendations, resources, training and insights work together, without exposing the technical engine as the main brand.
Shared institutional services
A strategic ecosystem for research integrity, not a simple anti-plagiarism interface.
EVIDENCE gives WACREN and member institutions a complete governance, knowledge, training and analysis platform under one trusted regional brand.
Platform model
Three layers built for regional scale.
The platform combines operational checks, institutional knowledge and regional intelligence so each stakeholder can act from the same trusted environment.
Publish policies, recommendations, resources and training paths aligned with WACREN guidance.
Monitor indicators by institution and country while preserving tenant isolation and privacy.
Detection workflow
From submission to an auditable institutional decision.
1Submit document and metadata
2Extract text and queue analysis
3Compare against trusted source collections
4Review similarity and AI indicators
5Export report and record actions
Inside the platform
The engines behind every signal
EVIDENCE brings together similarity, source-level evidence, AI-writing indicators, cross-language checks, stylometry and evasion detection — each one visible, filterable and exportable in a single report.
Unified report
One report, every signal
A single report unifies the similarity score, source evidence, AI indicators, obfuscation and language breakdown — with an explicit call for human review before any decision.
- Global and per-source scores
- AI-writing and obfuscation flags
- Certificate and alert exports
Source evidence
Every match traced to its source
Each similarity is traced to a web, archive or primary-collection source, with repetition counts and a per-section heatmap. Reviewers activate, mute or split any source.
- Primary vs inspiration sources
- Per-section match heatmap
- Activate or mute any source
AI review
Calibrated, advisory AI-writing signals
A detector calibrated to a 1% false-positive rate flags likely AI passages, internal style breaks and the human/AI split — always framed as an aid to human judgement, never proof.
- Calibrated, not punitive
- Pure / reformulated / human split
- AI intensity by position
Cross-language
Plagiarism that crossed a language
Every document is compared across eight languages at each scan, exposing text translated from one language to another — the plagiarism single-language checks never see.
- 8 languages on every scan
- Per-language coverage
- Automatic, no setup
Stylometry
A writing fingerprint per author
EVIDENCE profiles eight stylistic features and compares them to the institution's baseline, surfacing internal style breaks that suggest a passage written by another hand.
- Eight-feature style radar
- Institution baseline
- Internal style-break detection
Evasion
Evasion tricks, made visible
Look-alike letters, invisible characters, private-font mappings and bidirectional tricks used to fool detection are found, counted and located — down to the exact character.
- Cyrillic and Greek look-alikes
- Invisible and bidi characters
- Density map by position
Report anatomy
A report designed for judgement, not automatic punishment.
Each report combines a global score, source-level evidence, highlighted passages, applied filters, AI advisory signals, recommendations and a clear disclaimer.
EVIDENCE Report
Signed by WACREN
Similarity18%
AI indicator20%
Requires academic review and institutional context.
WACREN mission
A regional platform for trusted research and open knowledge.
EVIDENCE supports national research and education networks, universities, libraries, researchers and policy teams with shared digital services and regional integrity indicators.
--Published articles
--Resources
--Events
--Recommendations
Role-based experience
Each user sees the right responsibilities, data and tools.
Regional oversight, institutions, recommendations, statistics and content governance.
Users, departments, reports, internal trends and institutional settings.
Submissions, report review, comments, student follow-up and PDF exports.
Manuscript checks, authorized reports, training courses and certificates.
Guides, policies, templates, repositories and responsible research resources.
News, events, public resources and general information about EVIDENCE.
Knowledge hub
Guidance, resources and recommendations for member institutions.
The CMS supports editorial pages, policy notes, downloadable resources, recommendations, events and multilingual publishing.
Regional network
Multi-institution by design.
Each institution has isolated data, dashboards and users, while WACREN keeps the regional aggregate view.
Events and academy
Training, workshops and certification pathways.
EVIDENCE connects operational integrity checks with capacity building for students, supervisors, administrators and open science teams.
Multilingual by design
Detection in 100+ languages, tuned for the region
EVIDENCE analyses and cross-checks documents in more than a hundred languages, with particular care for the languages of higher education across West and Central Africa.
Regional languages of instruction
FrançaisEnglishPortuguêsالعربية
African languages
WolofBamanankanHausaYorùbáIgboFulfulde / PulaarAkan / TwiEweDioulaMooréLingalaKiswahili
World languages
EspañolItalianoDeutschNederlands中文日本語Русский+90
Cross-language detection exposes plagiarism translated between any pair of supported languages — a text copied in French and submitted in English is still caught.
Designed for sensitive academic records.
Role-based access, institution scoping, audit logs, file validation, secure authentication and explicit AI-score disclaimers are part of the product baseline.
Tenant isolationInstitutions only access their own users, documents and reports.
Human reviewSimilarity and AI indicators support decisions; they do not replace academic judgement.
Multilingual growthEnglish and French are active, with the structure prepared for Arabic and RTL layouts.
Access
Open secure workspace
Deploy EVIDENCE for a WACREN member institution.
Institution administrators can manage users, documents, reports, resources, recommendations and training from the secure workspace.