Applied ML

A methods portfolio spanning 17 analyses across 9 domains — applying classical ML, deep learning, ensemble modeling, and statistical regression to real-world problems in healthcare, agriculture, retail, finance, and engineering. Each project demonstrates deliberate method selection, rigorous comparative evaluation, and interpretable results — the analytical foundations underlying applied research and production systems.

18

Applied ML
Analyses

9

Domains

12

Datasets &
Real-World Problems

Domain

Healthcare

Clinical risk prediction · Diagnostic classification · Patient data modeling

Applied ML Healthcare
👁️

Diabetic Retinopathy Prediction

Binary Classification

A binary classification model predicting diabetic retinopathy probability from retinal image features. Uses backward elimination and AIC for variable selection on 1,151 patient records — providing a clinically interpretable probability score for early DR detection.

⭐ Clinically interpretable probability score for early diabetic retinopathy detection

R Binary Logistic Regression AIC
View Project →
Applied ML Healthcare
🫀

Coronary Heart Disease Risk Prediction

Ensemble Classification

A binary classification system predicting 10-year coronary heart disease risk from patient vitals and lab results. Benchmarks Random Forests against Mixture Discriminant Analysis across 10 random seeds on 4,080 Framingham Heart Study patients for consistent, reliable predictions.

⭐ 84.3% classification accuracy; RF outperformed MDA with lower variance

R MDA Random Forest
View Project →

Domain

Computer Vision & NLP

Spatial & temporal learning · Sequence modeling · Architecture implementation

Applied ML Computer Vision
🎥

DeepVision: Video Analysis

ConvLSTM Architecture

A ConvLSTM deep learning model that combines spatial feature extraction with temporal sequence modeling for automated video classification. Processes raw video frames through an OpenCV pipeline and outputs class predictions for video-heavy operational workflows.

⭐ 84% accuracy and 97% AUC; Combines spatial and temporal learning in one model

Python TensorFlow OpenCV
View Project →
Applied ML NLP
🤖

Transformer Architecture

Encoder-Decoder · Multi-Head Attention

A full Transformer architecture built from scratch in PyTorch — no Hugging Face, no abstraction shortcuts. Covers positional encoding, multi-head self-attention, feedforward layers, encoder and decoder blocks, and the complete encoder–decoder stack with configurable depth and heads.

⭐ Full architecture from scratch — every component implemented in detail

Python PyTorch Lightning
View Project →

Domain

Retail & Commerce

Demand forecasting · Customer behavior · Market basket analysis

Applied ML Retail & Commerce
🛍️

Product Purchase Prediction

CNN+LSTM · Multi-Output Forecasting

A CNN+LSTM hybrid pipeline forecasting per-product daily purchases using lagged and calendar features. Predicts next-day demand for multiple products simultaneously in a single forward pass — enabling smarter inventory planning and procurement decisions.

⭐ Multi-output sequence model predicting demand for N products in a single forward pass

Python PyTorch pandas scikit-learn
View Project →
Applied ML Retail & Commerce
🛒

Market Basket Analysis

Apriori · Unsupervised Association Mining

An unsupervised market basket analysis system applying the Apriori algorithm to uncover hidden product co-purchasing patterns from transaction data. Generates rules ranked by support, confidence, and lift — surfacing actionable bundling and cross-sell opportunities.

⭐ Actionable co-purchase patterns with high-lift rules for the retail domain

Python pandas mlxtend matplotlib
View Project →
Applied ML Retail & Commerce
🏥

PolicyPredict

Deep Learning · Binary Classification

A deep neural network pipeline predicting insurance policy purchase likelihood from customer data. Combines scikit-learn preprocessing with TensorFlow model training to output purchase probability scores for targeted marketing and personalized policy recommendations.

⭐ 70% accuracy and 85% ROC-AUC - Deep learning outperforms classical baselines

Python TensorFlow scikit-learn pandas
View Project →
Applied ML Retail & Commerce
🏷️

PolicyPredict Lite

Classical ML · Multi-Model Comparison

A lightweight scikit-learn pipeline predicting insurance purchase behavior using Logistic Regression, Random Forests, and SVMs. Built as the classical ML counterpart to PolicyPredict — directly comparing interpretability and performance tradeoffs against the deep learning version.

⭐ Side-by-side deep vs. classical ML comparison on the same insurance dataset

Python scikit-learn pandas NumPy
View Project →

Domain

Finance & Economics

Customer targeting · Economic forecasting · Behavioral pattern mining

Applied ML Finance & Economics
🏦

Bank Campaign Subscription Analysis

Apriori · Association Rule Mining

Association rule mining on 41,188 Portuguese bank clients to identify which attributes drive term deposit subscriptions. Rules ranked by standardized lift surface the most actionable client segments for targeted campaign outreach and rejection avoidance.

⭐ 13 rules generated; Marital status and prior contact identified as top subscription drivers

R arules Apriori Algorithm
View Project →
Applied ML Finance & Economics
📈

China's GDP Growth Modeling

Standard Cubic Splines · Smoothing Splines

A non-parametric spline regression system modeling China's GDP growth from 1960 to 2014. Compares standard cubic splines across varying degrees of freedom against smoothing splines across bandwidth parameters — analyzing the bias-variance tradeoff for complex economic trend forecasting.

⭐ Smoothing splines outperformed cubic splines for exponential growth modeling

R Cubic Splines Smoothing Splines
View Project →

Domain

Agriculture

Crop classification · Quality control automation · Agricultural ML

Applied ML Agriculture
🌾

AgriGrain: Grain Type Classification

Ensemble Learning · Neural Networks · Discriminant Analysis

A multiclass grain classification system built entirely in R, comparing five algorithms — Bagging, Random Forests, Boosting, MDA, and ANNs — on geometric and shape features. Automates grain type identification to reduce manual inspection effort in agricultural quality control.

⭐ Random Forest and Boosting consistently outperformed baseline classifiers across all grain types

R scikit-learn Boosting
View Project →

Domain

Real Estate

Property value prediction · Regularization benchmarking · Non-linear modeling

Applied ML Real Estate
🏘️

Boston House Value Prediction

Ridge · LASSO · LSLR

A regression benchmark comparing Least-Squares, Lasso, and Ridge on the Boston Housing dataset across 506 properties and 13 predictors. Demonstrates regularization benefits in the presence of multicollinearity — with Lasso producing the most parsimonious model by excluding INDUS and AGE.

⭐ Lasso is the best performer. Excluded INDUS and AGE for most parsimonious model

R glmnet Cross-Validation
View Project →
Applied ML Real Estate
🏙️

Taipei House Value Prediction

GAMs · MARS · Ridge

A non-linear regression benchmark comparing GAMs, MARS, and Ridge on 414 Taipei City properties. Exploits known non-linearities — house age takes a parabolic shape, MRT distance is inversely proportional to price — to demonstrate where non-parametric methods outperform linear regularization.

⭐ GAMs is the best performer; MRT distance identified as dominant price predictor

R earth gam ridge
View Project →

Domain

Materials Science & Forensics

Compositional analysis · Structural property prediction · Forensic classification

Applied ML Materials Science
🏗️

Material Property Prediction

Tree Ensemble Regression Benchmark

A tree ensemble regression benchmark predicting compressive strength from compositional features — comparing Bagging, Random Forests, and Gradient Boosting on 1,030 specimens. Demonstrates ensemble method tradeoffs in low-sample, high-feature-interaction regression settings.

⭐ Boosting significantly outperformed Random Forest and Bagging algorithms

R randomForest gbm
View Project →
Applied ML Forensics
🔬

Glass Type Forensic Clustering

Unsupervised Clustering

An unsupervised clustering system classifying forensic glass samples from crime scenes by chemical composition. Compares Hierarchical Clustering, K-Means, and Gaussian Mixture Models on 214 samples across 9 chemical variables to support forensic evidence identification.

⭐ K-Means best performer. Separated headlamp glass from all other types

R mclust K-Means
View Project →

Domain

Public Health & Sustainability

Population health modeling · Environmental prediction · Non-parametric methods

Applied ML Public Health & Sustainability
🌍

Life Expectancy Trend Modeler

k-NN Regression · Gaussian Kernel Smoothing

A non-parametric regression system modeling Russian life expectancy trends from 1800 to 2016. Compares k-NN regression and Gaussian kernel smoothing across a range of hyperparameters — analyzing the bias-variance tradeoff for complex non-monotonic time-series data.

⭐ Kernel smoothing preferred for smoother boundary behavior

R k-NN Regression Kernel Smoothing
View Project →
Applied ML Public Health & Sustainability
⛽

Fuel Economy Prediction

LASSO Regression · MARS

A comparative regression system predicting vehicle fuel economy from 7 characteristics on 392 vehicles. Benchmarks LASSO against MARS — demonstrating MARS's advantage in handling non-linearity and variable interactions while both address multicollinearity in the Auto MPG dataset.

⭐ MARS significantly outperformed LASSO

R glmnet earth
View Project →

Domain

ML Tooling

Model benchmarking · Hyperparameter optimization · Automated model selection

Applied ML ML Tooling
🌲

OptiTree: Tree Model Benchmarking

PSO · Multi-Model Benchmarking

A model benchmarking framework comparing seven tree-based classifiers — Decision Tree, Random Forest, AdaBoost, Gradient Boosting, LightGBM, XGBoost, and CatBoost — under automated PSO-based hyperparameter search. All experiments tracked via MLflow for reproducible comparison and downstream deployment.

⭐ PSO-tuned models outperformed default settings; All experiments logged in MLflow for reproducibility

Python scikit-learn MLflow hyperparameter-optimizer
View Project →