Skip to content
Hassan Farid
Portfolio & Development JournalAbuja, Nigeria · Available to relocate

Hassan Farid

Data ScientistMachine Learning EngineerData Analyst

Portrait of Hassan Farid

Bio

I am a results-driven Data Scientist and Machine Learning Engineer with a mathematics background, transforming complex datasets into actionable insights and intelligent solutions. My work spans predictive modelling, data analytics and independent machine learning research — with a particular interest in healthcare AI and topological data analysis.

Read full bio

“Building Intelligent Solutions Through Data, Research & Machine Learning.”

I turn complex datasets into actionable insight and intelligent systems — from predictive modelling and analytics to independent machine learning research.

View ProjectsExplore ResearchGet in Touch
Also on LinkedIn(opens in a new tab)
Vietoris–Rips complex of a sample point cloud at scale ε = 34, with 5 connected components.
β0 = 5
Fig. 01A Vietoris–Rips filtration on an illustrative point cloud. Drag the scale to grow the complex; the H₀ barcode below is computed live by single-linkage, and the vertical line marks the current scale.

Two directions

I.Direction

Research

Scientific investigation, mathematical reasoning, experimentation and machine learning research.

Read the research
II.Direction

Applied AI/ML

Practical machine learning, predictive analytics, intelligent systems and real-world problem-solving.

See the engineering work
§ 01 Featured researchCompleted

An Integrated Topological and Machine Learning Approach for Breast Cancer Classification

Investigating whether topological features from persistent homology complement classical morphological features for breast cancer classification on the WDBC dataset.

Observations
569
Classical / Topological / Combined
30 / 50 / 80
Models × CV folds
3 × 5
Best mean AUC · Classical, LR
0.9953
Read the case studyKaggle notebook(opens in a new tab)

ROC AUC by model

Mean, 5-fold
  • Classical · 30
  • Topological · 50
  • Combined · 80

Logistic Regression

Classical
0.9953
Topological
0.9482
Combined
0.9951

RBF SVM

Classical
0.9945
Topological
0.9280
Combined
0.9921

Random Forest

Classical
0.9885
Topological
0.9402
Combined
0.9915
AUC
0.900.930.950.971.00

Axis begins at 0.90. Whiskers show ± variation across folds. Classical features performed strongest overall; combined features did not consistently outperform them.

§ 02Applied AI/ML · In development

Four systems on the bench, four more on the drawing board.

All projects
  1. 01Ongoing
    Applied AI/MLHealthcare AI · Applied Machine Learning · Explainable AI

    Explainable Diabetes Risk Prediction

    An explainable machine learning pipeline for diabetes risk prediction, comparing interpretable and ensemble models with threshold-based evaluation and SHAP explanations.

  2. 02Ongoing
    Applied AI/MLApplied AI/ML · Signal Processing · TDA

    Topological Anomaly Detection in ECG Signals

    Exploring topological data analysis for detecting anomalous ECG signals, and comparing it against conventional signal features.

  3. 03Ongoing
    Applied AI/MLApplied AI/ML · Time-Series Forecasting

    Forecasting Nigerian Food Prices

    Exploring forecasting methods for staple food prices in Nigeria, from seasonal baselines to gradient boosting and neural forecasters.

  4. 04Ongoing
    Applied AI/MLApplied AI/ML · Natural Language Processing

    Student Wellbeing Survey Analysis Using NLP

    Exploring NLP methods for analysing anonymised student wellbeing survey responses, including English and Nigerian Pidgin text.

Planned: Customer Churn Prediction and Retention Analytics · Financial Fraud Detection Using Machine Learning · Agricultural Yield Prediction and Crop Analytics · Nigerian Energy Consumption Forecasting

§ 03Development journal

Notes from the work, as it happens.

Open the journal

Research notes

Writing up my TDA and machine learning breast cancer study.

A structured write-up of my independent research comparing classical, topological and combined feature representations on the WDBC dataset.

Career milestone

Relaunching this portfolio as a development journal.

This site is a long-term record of my projects, research and technical growth, not just a static portfolio.

Skills and experience

§ 04Toolkit

What I work with.

All skills
Programming & Data Manipulation
Python, Pandas, NumPy
Data Analysis & Visualisation
Matplotlib, Data analysis, Data pipelines
Machine Learning
Scikit-learn, Logistic Regression, Support Vector Machines, Random Forests, SHAP, Topic modelling, Sentiment analysis
Statistical Analysis
statsmodels, Principal Component Analysis, Cross-validation, Time-series forecasting
Research & Computational Methods
Topological Data Analysis, Persistent homology, Vietoris–Rips complexes, GUDHI
Database Management
Database management

§ 05Experience

Highlights.

Full record

October 2024 – January 2025

Data Science Intern

Data Science Nigeria (DSN)

  • Built end-to-end data pipelines covering collection, preprocessing and structuring.
  • Collaborated on developing, testing and deploying machine learning models on live problems.
  • Co-facilitated AI+ meetups, hackathons and workshops.
  • Campus Lead · ATC Africa, UDUS ChapterFebruary – August 2026
  • Student Ambassador & Intern · YhillsMay – July 2026
  • Pioneer Health Technologist · Inner Compass ProjectDecember 2025 – June 2026

§ 06 Contact

Have a dataset, a role or a question?

Based in Abuja, Nigeria. Available to relocate. I respond to email first.

Email meAll contact details
0903 808 1183