From Data Experiments to Business Impact

Explore real-world data science & machine learning work. Each post breaks down the approach and takeaways — and how the same methods can uncover opportunities, reduce risk, and guide better decisions.

Recent writing & project breakdowns

Insider Threats: Anomaly Detection on User Logs

Unsupervised ML flags suspicious behavior (USB events, file access) to surface potential insider risks early.

Facial Expression Classification with CNNs

Training a convolutional model to classify emotions from 2D images — data prep, architecture, and results.

Detecting Metastatic Cancer (ResNet18, PCam)

Binary classification on histopathologic scans; pipeline, metrics, and a clean Kaggle-ready submission.

Zeek Logs: Finding C2 / Beaconing with ML

Supervised learning over MITRE-labeled network data to highlight command-and-control patterns.

Life Stressors & Depression — Regression in R

Stepwise regression linking environmental and genetic factors (5-HTT) with depression outcomes.

Pitch Mining: A Case Study on Gerrit Cole

Feature engineering from pitch-level data to predict run-risk innings with ~83% accuracy.

CycleGAN: Turning Photos into Monet-Style Art

Training a CycleGAN to translate real-world images to Monet-style outputs; data, training, and MiFID.

Disaster Tweets with DistilBERT

Transformer-based classification for disaster vs. non-disaster tweets with minimal preprocessing.

Classifying News Articles with NLP & ML

Dimensionality reduction + supervised/unsupervised models to categorize news at scale.

Spotting Spam Job Posts via Logistic Regression

From feature selection to evaluation — a practical model to filter low-quality job listings.

Online Retail: Price Optimization

Finding revenue-maximizing price points by product with historical demand and price variation.

Single-Predictor Linear Regression in R

Transformations, lack-of-fit testing, and why the power model won — a compact walkthrough.