Skip to content
View SamInMotion's full-sized avatar

Block or report SamInMotion

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.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. 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
SamInMotion/README.md

Hi, I'm Sam

NLP engineer with a linguistics background (MPhil, University of Bergen). Most of my work sits at the intersection of medical text and structured knowledge including ontologies, clinical terminologies, that sort of thing.

At Nemoest AS ,I built a RAG-based document retrieval system for Norwegian legal texts. My thesis was on automating article screening for systematic reviews using ontology-enriched classification. More recently I built MedTermCheck, a tool that verifies LLM-extracted medical codes against SNOMED-CT and ICD-10.


What I work on

Clinical NLP: classifying medical texts using ontology-based feature enrichment, verifying LLM outputs against structured terminologies

Legal document retrieval: RAG pipelines for Norwegian regulatory text

Explainable ML : SHAP-based interpretation for classification models in healthcare and business contexts


Stack

Python, scikit-learn, TensorFlow, SHAP, Hugging Face, Gradio, Azure, REST APIs, SNOMED-CT, ICD-10, SQL, MLflow

Certs: Azure AI Engineer Associate (AI-102), Azure Fundamentals, ML Engineer (DataCamp)


Current focus

Extending my thesis classification pipeline with benchmark datasets (Cohen et al. 2006, 2,744 abstracts) and 5-fold cross-validation. Recent finding: mechanical ontology enrichment doesn't reliably improve classification with broader implications for ontology-grounded NLP.

Building out local RAG infrastructure for document retrieval workflows.


sammy.okmens@gmail.com | LinkedIn

Pinned Loading

  1. Medical-intervention-text-classification Medical-intervention-text-classification Public

    Master's thesis (UiB, Computational Linguistics, 2023) and extension work. Text classification for systematic review screening. Original 11-workflow analysis with NEO ontology integration plus Cohe…

    Jupyter Notebook 2

  2. llm-medical-verification-system llm-medical-verification-system Public

    MedTermCheck: verifies LLM-extracted medical codes against SNOMED-CT and ICD-10. Four-signal confidence scoring. Deployed on HuggingFace Spaces.

    Python 1

  3. gdpr-healthcare-ai-compliance-scorer gdpr-healthcare-ai-compliance-scorer Public

    Ontology-first, rules-first framework for assessing GDPR compliance in healthcare AI systems.

    Python 1