About Stabilarity Research Hub
Stabilarity Research Hub is an open-access scientific platform for research at the intersection of artificial intelligence, economics, and data science. We publish academic and community-reviewed work, build open-source research tools, and provide free API access to AI systems developed by our team.
The hub originated from the MedAI Hackathon 2025 in Odessa, Ukraine and operates under the institutional affiliation of Odesa National Polytechnic University (ONPU), Department of Economic Cybernetics and Information Technologies, with legal representation by Stabilarity OU (Estonian Registry: 17150040).
Research Team
| Researcher | Role & Affiliation | ORCID |
|---|---|---|
| Oleh Ivchenko | Lead Researcher · PhD Candidate, Economic Cybernetics, ONPU · Innovation Tech Lead, Capgemini Engineering | 0000-0002-9540-1637 |
| Dmytro Grybeniuk | Co-Researcher · Anticipatory Intelligence, predictive system architecture | 0009-0005-3571-6716 |
| Iryna Ivchenko | Co-Researcher · Intellectual Data Analysis, knowledge extraction, data mining taxonomy | 0000-0002-1977-0342 |
Research Areas
The hub maintains active research series across multiple domains. Each series is archived on Zenodo with permanent DOIs.
| Series | Focus |
|---|---|
| Medical ML Diagnosis | Machine learning in healthcare imaging, diagnostic accuracy, and clinical decision support |
| AI Economics | Economic frameworks for AI investment, ROI modelling, and enterprise decision-making |
| Cost-Effective AI | LLM deployment strategies, model selection, and efficient AI patterns for production |
| Anticipatory Intelligence | Predictive systems, forecasting methodologies, and proactive decision frameworks |
| Intellectual Data Analysis | Data mining taxonomy, pattern discovery, and knowledge extraction |
| Geopolitical Risk Intelligence | 87-country geopolitical risk quantification and cross-domain scientific synthesis |
| Future of AI | Emerging trends, long-range implications, and the evolving AI landscape |
| Spec-Driven Development | Formal specification methods and structured approaches to reliable systems |
| HPF-P | Heuristic Prediction Framework for pharmaceutical decision readiness and portfolio scoring |
| AI Observability | Monitoring, tracing, and explainability for production AI systems |
Publishing Standards
Zenodo DOI Registration
Every paper published on Stabilarity Hub is permanently archived on CERN Zenodo with an immutable DOI. This ensures long-term citability, version tracking, and integration into global citation networks.
Citation Standards (2026)
Publications in primary research series must meet the 80% 2026 citation rule — at least 80% of cited sources must originate from 2026 or later, ensuring engagement with the current state of the field.
Peer Review Threshold
Research series designated as academic require editorial review and community critique before addition to the canonical Zenodo archive. The review process is documented and transparent.
Open Research Tools
Beyond publications, the hub develops and maintains working AI systems freely accessible to researchers.
Six diagnostic models for medical imaging, including chest X-ray and retinal scan analysis. Open inference API for researchers.
Pharmaceutical decision readiness scoring based on the HPF = DRI + DRL framework. Supports portfolio prioritisation and pipeline decisions.
87-country geopolitical risk index updated quarterly. Quantified scoring across political, economic, and conflict dimensions.
Free API access to all Stabilarity AI systems for researchers. Enables reproducible, API-driven experiments. No registration fee.
Scientific Philosophy
- Open by default. All publications are free, publicly accessible, and permanently registered with DOIs.
- Post-facto moderation. Research is published directly and reviewed after publication, enabling rapid scientific discourse.
- Author accountability. Authors are responsible for accuracy and integrity. Fabrication or plagiarism results in removal.
- Community critique. Every publication is subject to public critique. This is a feature, not a limitation.
- Reader responsibility. Readers are responsible for independent verification before acting on any published information.
Collaboration & Contact
We welcome collaboration with researchers, institutions, and practitioners working in AI, economics, or adjacent fields. If you are interested in contributing to an active research series, co-authoring a publication, or integrating our tools into your research workflow, please reach out.
- Join the community: Register for free
- API access: Get your free API key
- ORCID profiles listed above for direct academic contact
Platform Integrity & Research Infrastructure
Stabilarity Research Hub was built by a research community that values transparency, reproducibility, and long-term accessibility. Every system described below is operational and verifiable.
| Capability | Implementation |
|---|---|
| Regular Backups | Full-system backups (WordPress, database, research files, configuration) stored independently from the production server. Restorable via single-command script. |
| Blue/Green Mirror | Mirror server provisioned from the latest backup for disaster recovery. Enables near-zero-downtime restoration if the primary instance fails. |
| Full Version Control | All research tools and platform code are maintained under Git version control with public repositories on GitHub (github.com/stabilarity). Article content is tracked through WordPress revision history — every edit is logged with timestamp, author, and diff. Currently tracking 1,300+ revisions across 250+ publications. |
| Article Version History | Every published article displays a transparent version history widget showing all revisions, editorial changes, and content evolution from first draft to current version. Publicly visible to all readers. |
| Responsive Design | All pages, tools, and interactive applications are fully adaptive for mobile, tablet, and desktop. Tested across viewport sizes with consistent reading experience. |
| Uptime & Accessibility | Platform maintains 98%+ uptime with monitoring. HTTPS enforced via Let’s Encrypt TLS. All content accessible without registration or paywall. |
| Reference Tracking System | Database-backed reference manager tracking 3,000+ external citations across all publications. Link validation, content caching, DOI verification via CrossRef, and duplicate detection via content similarity analysis. |
| Zenodo DOI Registration | Every publication is registered on Zenodo with a permanent DOI (10.5281/zenodo.*), ensuring long-term citability independent of this platform. Currently 210+ articles with DOIs assigned. |
Editorial Review Process
Every publication undergoes a structured post-publication moderation period:
- 30-day review window. Each paper is actively moderated for one month from publication date.
- Editorial team: Iryna Ivchenko and Oleh Ivchenko — representing direct involvement of the ONPU Department of Economic Cybernetics and Information Technologies in the review process.
- Quality standards. Citation freshness (80% from current year), compliance checks, scientific value scoring, and format consistency are enforced as community rules on every article.
- Transparent revisions. All editorial changes are logged in the article’s version history, visible to readers.
This process is designed to support the transition from preprint to external journal submission (Scopus, Web of Science, Springer, IEEE). The review period strengthens each paper before formal peer review.
Compliance Readiness
The platform is prepared for compliance with the following standards and frameworks:
| Standard | Status | Scope |
|---|---|---|
| OWASP Top 10 | Prepared for compliance | HTTPS enforcement, input sanitization, authenticated admin access, dependency updates, no known injection vectors |
| FAIR Data Principles | Substantially compliant | Findable (DOIs + search), Accessible (open access, no paywall), Interoperable (standard formats, API), Reusable (Zenodo archive, open licensing) |
| Plan S / Open Access | Compliant | All publications immediately open access with no embargo period. Permanent DOIs via Zenodo. |
| WCAG 2.1 (Level AA) | Prepared for compliance | Semantic HTML, sufficient contrast ratios, responsive layout, keyboard navigation support |
| GDPR | Prepared for compliance | No user tracking cookies without consent. Matomo analytics with IP anonymization. No personal data collection beyond voluntary API registration. |
| CII Best Practices | Prepared for compliance | Open source repositories with version control, issue tracking, and documented contribution guidelines (Linux Foundation Core Infrastructure Initiative) |
Hub Release Notes
+Individual publications represent the views of their respective authors and do not constitute the official position of Stabilarity OU or ONPU. The hub does not guarantee the accuracy, completeness, or reliability of any individual publication. Readers should verify sources independently and consult qualified professionals before acting on any information published here.