Data Warehousing and Data Mining
Complete course on Data Warehousing and Data Mining. Learn Data Warehousing (Kimball, Data Lakehouses, ETL with PySpark) and the CRISP-DM cycle (EDA, Preprocessing, Predictive Modeling).
Computer Science
Course Notes · Semester 2026–II
Data Warehousing &
Data Mining
From data warehouses and ETL pipelines with PySpark to supervised classification, clustering, and dimensionality reduction — complete with mathematical rigor, Python implementations, Databricks labs, and over 80 interactive visualizations.
5
Parts
47+
Chapters
80+
Visualizations
100%
Reproducible
Preface
These lecture notes cover the full end-to-end data value cycle: analytical warehouse design, pattern mining, preprocessing, supervised classification, and probabilistic modeling. Each topic combines rigorous theoretical foundations with executable Python code.
Part I
Data Warehouses & ETL
DW architecture, multidimensional modeling, ETL/ELT pipelines with PySpark on Databricks, heterogeneous sources, and OLAP cubes.
Part II
Data Mining Process
CRISP-DM methodology, univariate and multivariate EDA, preprocessing with sklearn Pipelines, missing data imputation, PCA, and feature selection.
Part III
Pattern Mining
Association rules: Apriori and FP-Growth. Support, confidence, lift, conviction, leverage, and Zhang metrics.
Part IV
Supervised Classification
CART decision trees, Random Forests, XGBoost, SVMs, logistic regression, Naive Bayes, GMMs, neural networks, and ROC-AUC evaluation.
Part V
Unsupervised Learning
K-Means, hierarchical clustering, DBSCAN, t-SNE, UMAP, and autoencoders for uncovering latent structure without labels.
Cross-Cutting
Python & Modern Tooling
PySpark, Databricks, Delta Lake, dbt, scikit-learn, PyTorch, and JAX with interactive visualizations in every chapter.
Syllabus
I
Data Warehouses
8 chapters
DW Architecture
Multidimensional Model
ETL vs ELT
PySpark & Databricks
Heterogeneous JSON/CSV
Delta Lake
dbt & Airflow
OLAP Cubes
II
Data Mining Process
17 chapters
CRISP-DM
Univariate EDA
Central Tendency · Dispersion · Shape
Heterogeneity · Concentration
Correlation · Multivariate
sklearn Pipeline
Imputation · Noise
PCA · χ² Selection
III
Pattern Mining
3 chapters
Apriori
FP-Growth
Support · Confidence · Lift
Conviction · Leverage · Zhang
IV
Supervised Classification
14 chapters
CART Trees
Bagging · Random Forests
XGBoost · SHAP
Logistic Regression
SVM · Kernels
Naive Bayes · GMM
Neural Networks · CNN
PyTorch · JAX
ROC · AUC · k-fold
V
Unsupervised Learning
5 chapters
Clustering Metrics
K-Means · K-Medoids
Hierarchical · Ward
DBSCAN · HDBSCAN
t-SNE · UMAP
Autoencoders
Prerequisites
Targeted at students in Computer Science, Mathematics, Physics, and related quantitative fields.
Calculus & Linear Algebra
Partial derivatives, gradients, matrix operations, and eigenvalues.
Probability & Statistics
Random variables, distributions, expectation, variance, and Bayes' theorem.
Python Programming
Basic–intermediate level: numpy, pandas, and tabular data manipulation.
Databases
Basic SQL and relational modeling (recommended, not mandatory).
Instructor
DV
Diego Villalba
Faculty of Sciences ·
National Autonomous University of Mexico (UNAM)
These notes bridge mathematical theory with reproducible Python implementations,
prioritizing geometric intuition and formal rigor in equal measure.
Contact: diego.villalba@ciencias.unam.mx