Skip to content
View seun-john's full-sized avatar

Block or report seun-john

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
seun-john/README.md

John Oluwaseun Oluwawalemi

Researcher and data analyst based in Ibadan, Nigeria. I lead Novapoint Consulting, where I have supported research, data analysis and training work for more than ten years. I also teach computer applications to health students.

I build open-source tools that check AI-assisted research for fake citations, broken instructions and unverified numbers. An open preprint evaluates fifteen of them with and without each tool: https://doi.org/10.5281/zenodo.23404947

Research verification tools

  • verify-references catches fabricated and mis-cited references.
  • citeproof audits references and the claims they support.
  • specguard tests whether AI-generated work followed its instructions.
  • clean-output keeps prompt scaffolding out of deliverables and flags prompt injection.
  • research-document-auditor audits theses, reports and manuscripts.
  • reviewtrace traces reviewer feedback to the evidence that it was addressed.

Open data and analytics

Elsewhere

Novapoint Consulting | ORCID | Google Scholar | LinkedIn

Pinned Loading

  1. citeproof citeproof Public

    Audit references and the claims they support: DOIs, metadata, duplicates, retractions and evidence passages. Offline.

    Python

  2. customer-churn-intelligence customer-churn-intelligence Public

    Explainable telecom churn intelligence: validated ML pipelines, SHAP explanations, customer analytics and retention scenarios in Streamlit.

    Python

  3. marketing-experiment-intelligence marketing-experiment-intelligence Public

    Hillstrom marketing experimentation: randomised treatment effects, DuckDB SQL, statistical inference and an interactive Streamlit decision workspace.

    Python

  4. nigeria-health-dashboard nigeria-health-dashboard Public

    Six public health indicators for Nigeria from World Bank and WHO open data, with gaps and source disagreements left visible.

    Python

  5. specguard specguard Public

    Verify that AI-generated work actually follows the instructions.

    Python

  6. verify-references verify-references Public

    Detect AI-hallucinated and mis-cited references. A Claude skill that verifies bibliographies against Crossref, OpenAlex, PubMed, CORE and DataCite.

    Python