👑 Multivariate exploratory data analysis in Python — PCA, CA, MCA, MFA, FAMD, GPA

IBM Employee Profiling using Clustering

Analysis of Nitrogen and Carbon stable isotopes in wild and farmed scallops from Penobscot Bay, Maine.

This repository contains the codes from Intxaurbe et al., 2024 (Archaeological and Anthropological Sciences).

Python project for the Algorithmic Methods of Data Science class for the MSc. in Data Science at the Sapienza University of Rome. The main purpose of the project is building a Recommendation Engine using Locally Sensitive Hashing and using K-Means to cluster users based on their Netflix activity

Capstone project applying unsupervised learning to segment international tourists in Peru using PROMPERÚ’s 2024 Foreign Tourist Profile Survey. Combines Factor Analysis of Mixed Data (FAMD) with PAM clustering to derive robust segments from mixed categorical and numerical survey data.

This repository contains the implementation of a multivariate control chart with dimension reduction techniques, namely Factor Analysis of Mixed Data (FAMD) and Autoencoder. The control chart is designed for detecting network intrusions using network data traffic.

Use a variety of data science techniques to detect whether a tax return is fraudulent or not.

End-to-end machine learning pipeline designed for high-dimensional, mixed-type survey data. Implemented K-Prototypes Clustering, Factor Analysis of Mixed Data, and t-SNE manifold learning to successfully segment 1,400+ observations. Features rigorous feature engineering (ordinal encoding, MAR analysis) model validation via stability test (ARI=0.9).

Datathon with a retargeting ad company. Churn date prediction (Normalized RSME 38), clustering (98% Silhouette), automated identification of the gaps between best and average client within a cluster.

Implémentation des méthodes statistiques notamment de l'analyse factorielle pour identifier la combinaison des caractéristiques qui sont plus susceptibles d'être associées à un accident vasculaire cérébral (AVC).

We examine the relationship between variables in a fictional Telco customers dataset. The techniques used are Factor Analysis of Mixed Data and logistic regression.