Skip to main content
B.Sc. DegreeFaculty of Faculty of Computing & Information TechnologyData Science

Bachelor of Science (B.Sc.) in Data Science: Mathematical Foundations, Statistical Learning & Big Data Analytics (NUC CCMAS Standards)

A comprehensive 4-year undergraduate degree curriculum aligned with the National Universities Commission (NUC) Core Curriculum and Minimum Academic Standards (CCMAS) for Data Science in Nigerian universities.

Instructor
Created by PAMCET Learning Team
advanced1 weekEnglish

Course Overview

A comprehensive 4-year undergraduate degree curriculum aligned with the National Universities Commission (NUC) Core Curriculum and Minimum Academic Standards (CCMAS) for Data Science in Nigerian universities. Curriculum Structure & Core Modules: • 1. DTS 101: Foundations of...

Degree Programme Curriculum Context

This course is a curriculum unit of the Bachelor of Science (B.Sc.) in Data Science degree framework in Nigeria.

Earn Your Certificate

Showcase your skills with a certificate! Complete the course, submit your project, and earn your certificate. Here's a sample of what you'll receive to show off your accomplishments.

course-certificate
Lecture Board & Syllabus

Course Curriculum

Browse complete theoretical foundations, worked calculations, laboratory protocols, and board assessment questions under each topic.

Launch Course Reader →
24 Chapters
•
72 Lecture
•
0s
Subtopic 1Document LectureOpen Access~16 min read

The Data Science Lifecycle: Epistemology, Hypothesis Generation & Inductive Modeling

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Theoretical & Epistemological Foundations University Hub > Nigeria > B.Sc. Data Science > Chapter 1 > Lecture 1 The Data Science Lifecycle: Epistemology, Hypothesis Generation & Inductive Modeling Epistemology of empirical inquiry, Breiman's two modeling cultures, mathematical formulation of empirical risk minimization, and ind...

Module 1 • Subtopic 1•13,571 characters of authentic academic content
Subtopic 2Document Lecture

Data Acquisition Paradigms: Structured, Semi-Structured, and Streaming Architectures

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Systems & Data Engineering University Hub > Nigeria > B.Sc. Data Science > Chapter 1 > Lecture 2 Data Acquisition Paradigms: Structured, Semi-Structured, and Streaming Architectures Architectural taxonomy of data representation, schema-on-write vs schema-on-read, row vs columnar storage physics, and stream processing queue dynamics...

Module 1 • Subtopic 2
Subtopic 3Document Lecture

Scientific Computing Environments: Virtualization, Reproducibility & Pipeline Determinism

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Computational Reproducibility & Infrastructure University Hub > Nigeria > B.Sc. Data Science > Chapter 1 > Lecture 3 Scientific Computing Environments: Virtualization, Reproducibility & Pipeline Determinism Epistemic reproducibility in computational science, Linux container isolation mechanisms, IEEE 754 arithmetic determinism,...

Module 1 • Subtopic 3
Subtopic 1Document Lecture

Vector Spaces, Linear Transformations & Orthonormal Bases

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Mathematical Foundations University Hub > Nigeria > B.Sc. Data Science > Chapter 2 > Lecture 1 Vector Spaces, Linear Transformations & Orthonormal Bases Vector space axiomatization, fundamental subspaces, the Rank-Nullity theorem, and Gram-Schmidt orthogonalization with QR decomposition. 1. Theoretical Foundations: Vector Spaces &a...

Module 2 • Subtopic 1
Subtopic 2Document Lecture

Matrix Decompositions: Eigendecomposition, Singular Value Decomposition (SVD) & Principal Components

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Spectral Theory & Dimensionality Reduction University Hub > Nigeria > B.Sc. Data Science > Chapter 2 > Lecture 2 Matrix Decompositions: Eigendecomposition, Singular Value Decomposition (SVD) & Principal Components The Spectral Theorem for symmetric operators, derivation of SVD, the Eckart-Young-Mirsky theorem, and the algebraic...

Module 2 • Subtopic 2
Subtopic 3Document Lecture

Multivariate Gradients, Jacobians & Hessians in High-Dimensional Loss Spaces

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Multivariable Differential Calculus University Hub > Nigeria > B.Sc. Data Science > Chapter 2 > Lecture 3 Multivariate Gradients, Jacobians & Hessians in High-Dimensional Loss Spaces Matrix calculus layout conventions, derivative identities of linear and quadratic forms, Hessian curvature tensors, and gradient descent dynamics. 1....

Module 2 • Subtopic 3
Subtopic 1Document Lecture

Differential Calculus in R^n: Directional Derivatives, Taylor Expansions & Convexity

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Continuous Optimization & Analysis University Hub > Nigeria > B.Sc. Data Science > Chapter 3 > Lecture 1 Differential Calculus in R^n: Directional Derivatives, Taylor Expansions & Convexity Multivariate directional derivatives, multivariate Taylor series approximations, convex analysis, and Jensen's inequality in statistical mo...

Module 3 • Subtopic 1
Subtopic 2Document Lecture

Unconstrained Optimization: Gradient Descent, Newton-Raphson & Quasi-Newton Mechanics

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Algorithmic Optimization University Hub > Nigeria > B.Sc. Data Science > Chapter 3 > Lecture 2 Unconstrained Optimization: Gradient Descent, Newton-Raphson & Quasi-Newton Mechanics Lipschitz continuous gradients, convergence rates of first-order descent, second-order Newton-Raphson dynamics, and the BFGS Quasi-Newton secant update....

Module 3 • Subtopic 2
Subtopic 3Document Lecture

Constrained Optimization: Lagrange Multipliers & Karush-Kuhn-Tucker (KKT) Conditions

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Constrained Mathematical Programming University Hub > Nigeria > B.Sc. Data Science > Chapter 3 > Lecture 3 Constrained Optimization: Lagrange Multipliers & Karush-Kuhn-Tucker (KKT) Conditions Lagrangian duality theory, Karush-Kuhn-Tucker necessary and sufficient conditions, complementary slackness, and support vector machine formul...

Module 3 • Subtopic 3
Subtopic 1Document Lecture

Scales of Measurement, Central Tendency & Robust Measures of Dispersion

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Descriptive Statistics University Hub > Nigeria > B.Sc. Data Science > Chapter 4 > Lecture 1 Measures of Central Tendency & Dispersion: Mathematical Foundations & Robustness Axiomatic characterization of location estimators, L1 and L2 loss duality, Bessel's correction proof, breakdown points, and robust dispersion scaling. 1. T...

Module 4 • Subtopic 1
Subtopic 2Document Lecture

Exploratory Data Analysis: Quantile-Quantile Plots, Heavy Tails & Skewness Metrics

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Distributional Analysis University Hub > Nigeria > B.Sc. Data Science > Chapter 4 > Lecture 2 Moments of a Distribution: Skewness, Kurtosis & Quantile Analysis Raw vs central moments, Fisher-Pearson standardized coefficients, kurtosis as tail weight, and robust quantile-based shape estimators. 1. Theoretical Foundations: Moment-Gen...

Module 4 • Subtopic 2
Subtopic 3Document Lecture

Bivariate Association: Pearson Correlation, Spearman Rank & Kendall Tau Derivations

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Exploratory Data Analysis University Hub > Nigeria > B.Sc. Data Science > Chapter 4 > Lecture 3 Exploratory Data Analysis: Tukey's Fences, Boxplots & Multivariate Cross-Tabulation Order statistics summaries, theoretical derivation of the 1.5 IQR fence multiplier, bivariate contingency analysis, Pearson's Chi-Square, and Simpson's P...

Module 4 • Subtopic 3
Subtopic 1Document Lecture

Data Structures, Memory Allocation & Computational Complexity in Python

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Programming Foundations University Hub > Nigeria > B.Sc. Data Science > Chapter 5 > Lecture 1 Control Structures, First-Class Functions & Functional Programming with Map/Filter/Reduce Lambda calculus foundations, lexical scoping, closure cell mechanics, functional stream combinators, and lazy evaluation pipelines. 1. Theoretical Fo...

Module 5 • Subtopic 1
Subtopic 2Document Lecture

Object-Oriented Programming Paradigms: Encapsulation, Polymorphism & Abstract Classes

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Software Engineering & OOP University Hub > Nigeria > B.Sc. Data Science > Chapter 5 > Lecture 2 Object-Oriented Design: Encapsulation, Polymorphism, Abstract Base Classes & Dunder Methods Type theory, dynamic dispatch, C3 linearization algorithm, abstract base classes, and Python data model dunder protocols. 1. Theoretical Fou...

Module 5 • Subtopic 2
Subtopic 3Document Lecture

Functional Programming, Vectorized Operations & Memory Profiling with NumPy

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Systems & Runtime Engineering University Hub > Nigeria > B.Sc. Data Science > Chapter 5 > Lecture 3 Memory Management, Garbage Collection (C-API/Reference Counting) & High-Performance Generators CPython memory architecture, PyObject reference counting, generational cyclic GC, frame execution suspension, and streaming generators...

Module 5 • Subtopic 3
Subtopic 1Document Lecture

Missing Data Mechanisms: MCAR, MAR, MNAR & Principled Imputation Methods

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Data Cleansing & Preparation University Hub > Nigeria > B.Sc. Data Science > Chapter 6 > Lecture 1 Missing Data Mechanisms (MCAR, MAR, MNAR) & Multiple Imputation by Chained Equations (MICE) Rubin's missing data taxonomy, ignorability conditions, fully conditional specification, and Rubin's pooling rules for total variance. 1....

Module 6 • Subtopic 1
Subtopic 2Document Lecture

Outlier Detection: Mahalanobis Distance, Isolation Forests & Robust Z-Scores

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Statistical Anomaly Detection University Hub > Nigeria > B.Sc. Data Science > Chapter 6 > Lecture 2 Outlier Detection: Mahalanobis Distance, Isolation Forests & Robust Z-Scores Multivariate anomaly geometry, FastMCD covariance estimation, Isolation Forest tree depth derivations, and affine invariance. 1. Theoretical Foundations: Mu...

Module 6 • Subtopic 2
Subtopic 3Document Lecture

ETL Pipeline Construction: Schema Validation, Idempotence & Data Normalization

B.Sc. Data Science NUC CCMAS Benchmark (100 Level) Data Engineering & ETL University Hub > Nigeria > B.Sc. Data Science > Chapter 6 > Lecture 3 ETL Pipeline Construction: Schema Validation, Idempotence & Data Normalization Distributed data ingestion, algebraic idempotence proofs, schema enforcement contracts, and power transformations. 1. Theoretical Foundations: Ingestion Cont...

Module 6 • Subtopic 3
Subtopic 1Document Lecture

Axiomatic Probability, Conditional Probability & Bayes' Theorem Derivation

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) Probability Theory University Hub > Nigeria > B.Sc. Data Science > Chapter 7 > Lecture 1 Axiomatic Probability, Conditional Probability & Bayes' Theorem Derivation Kolmogorov probability spaces, conditional conditioning, partition algebra, and Bayes' theorem with base rate fallacy resolution. 1. Theoretical Foundations: Kolmogorov...

Module 7 • Subtopic 1
Subtopic 2Document Lecture

Univariate & Multivariate Distributions: Normal, Poisson, Exponential & Gamma Families

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) Parametric Families University Hub > Nigeria > B.Sc. Data Science > Chapter 7 > Lecture 2 Univariate & Multivariate Distributions: Normal, Poisson, Exponential & Gamma Families Parametric families, Poisson limit proofs, memoryless Exponential waiting times, Gamma distributions, and multivariate Gaussian geometries. 1. Theoretic...

Module 7 • Subtopic 2
Subtopic 3Document Lecture

Asymptotic Theorems: Law of Large Numbers (LLN) & Central Limit Theorem (CLT) Proofs

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) Asymptotic Statistics University Hub > Nigeria > B.Sc. Data Science > Chapter 7 > Lecture 3 Asymptotic Theorems: Law of Large Numbers (LLN) & Central Limit Theorem (CLT) Proofs Modes of stochastic convergence, Khinchin's WLLN, Kolmogorov's SLLN, and characteristic function proof of the Lindeberg-Lévy CLT. 1. Theoretical Foundations...

Module 7 • Subtopic 3
Subtopic 1Document Lecture

Point Estimation: Maximum Likelihood Estimation (MLE) & Method of Moments

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) STA 202 Mathematical Statistics University Hub > Nigeria > B.Sc. Data Science > Chapter 8 > Lecture 1 Point Estimation: Maximum Likelihood Estimation (MLE) & Method of Moments Statistical likelihood foundations, score vectors, Fisher information matrices, analytical derivations, and numerical optimization via BFGS. 1. Theoretical P...

Module 8 • Subtopic 1
Subtopic 2Document Lecture

Estimator Quality: Unbiasedness, Consistency, Efficiency & Cramér-Rao Lower Bound

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) STA 202 Mathematical Statistics University Hub > Nigeria > B.Sc. Data Science > Chapter 8 > Lecture 2 Estimator Quality: Unbiasedness, Consistency, Efficiency & Cramér-Rao Lower Bound Formal loss metrics, Mean Squared Error decomposition, Cramér-Rao inequality derivation via Cauchy-Schwarz, Rao-Blackwell conditioning,...

Module 8 • Subtopic 2
Subtopic 3Document Lecture

Hypothesis Testing: Neyman-Pearson Lemma, Likelihood Ratio Tests & p-Value Interpretation

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) STA 202 Mathematical Statistics Nigeria > B.Sc. Data Science > Ch 8 > Lec 3 Hypothesis Testing: Neyman-Pearson Lemma, Likelihood Ratio Tests & p-Value Interpretation Statistical decision theory, Type I and II error trade-offs, Most Powerful test optimality proofs, Generalized Likelihood Ratio Tests, Wilks' asymptotic theorem, and p-va...

Module 8 • Subtopic 3
Subtopic 1Document Lecture

Asymptotic Complexity Analysis, Recurrence Relations & Master Theorem

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 201 Data Structures & Algorithms Nigeria > B.Sc. Data Science > Ch 9 > Lec 1 Asymptotic Complexity Analysis, Recurrence Relations & Master Theorem Formal set-theoretic asymptotic bounds, recursion tree geometry, divide-and-conquer recurrence equations, and Master Theorem case analysis. 1. Formal Definitions of Asymptotic Growt...

Module 9 • Subtopic 1
Subtopic 2Document Lecture

Hash Tables, Collision Resolution & Amortized Constant Time Lookups

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 201 Data Structures & Algorithms Nigeria > B.Sc. Data Science > Ch 9 > Lec 2 Hash Tables, Collision Resolution & Amortized Constant Time Lookups Universal hashing families, collision resolution mechanisms, potential method amortized analysis, and high-performance associative mapping. 1. Formal Theory of Universal Hashing &...

Module 9 • Subtopic 2
Subtopic 3Document Lecture

Trees, B-Trees, Graph Traversal Algorithms & Topological Sorting

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 201 Data Structures & Algorithms Nigeria > B.Sc. Data Science > Ch 9 > Lec 3 Trees, B-Trees, Graph Traversal Algorithms & Topological Sorting Self-balancing search trees, B-Tree disk I/O optimization, Breadth-First and Depth-First graph traversals, and Directed Acyclic Graph (DAG) topological ordering. 1. Hierarchical Tree Str...

Module 9 • Subtopic 3
Subtopic 1Document Lecture

Relational Algebra Formalisms: Selection, Projection, Joins & Set Operations

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 202 Relational Database Systems Nigeria > B.Sc. Data Science > Ch 10 > Lec 1 Relational Algebra Formalisms: Selection, Projection, Joins & Set Operations Mathematical relational model foundations, algebraic operators, derived joins, division, and query equivalence transformations. 1. Set-Theoretic Foundations of the Relational Mod...

Module 10 • Subtopic 1
Subtopic 2Document Lecture

Database Normalization: Functional Dependencies, 1NF Through BCNF Decompositions

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 202 Relational Database Systems Nigeria > B.Sc. Data Science > Ch 10 > Lec 2 Database Normalization: Functional Dependencies, 1NF Through BCNF Decompositions Armstrong's inference axioms, attribute closures, candidate key identification, 1NF to BCNF normal forms, and lossless-join decomposition theorems. 1. Formal Theory of Functional...

Module 10 • Subtopic 2
Subtopic 3Document Lecture

Advanced SQL: Window Functions, Common Table Expressions & Query Execution Plans

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 202 Relational Database Systems Nigeria > B.Sc. Data Science > Ch 10 > Lec 3 Advanced SQL: Window Functions, Common Table Expressions & Query Execution Plans Recursive CTE hierarchies, analytic window partition frames, Cost-Based Optimizer mechanics, and physical join algorithm complexity analysis. 1. Analytic Window Functions &am...

Module 10 • Subtopic 3
Subtopic 1Document Lecture

The R S3/S4 Object Model, Vectorization & Memory Management

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 203 Statistical Programming in R Nigeria > B.Sc. Data Science > Ch 11 > Lec 1 The R S3/S4 Object Model, Vectorization & Memory Management Functional programming lineage, Copy-on-Modify semantics, S3 informal vs S4 formal object systems, SIMD vectorized evaluation, and memory profiling. 1. R Architectural Lineage & Memory Seman...

Module 11 • Subtopic 1
Subtopic 2Document Lecture

Tidy Data Philosophy, Grammar of Data Manipulation & Functional Programming with purrr

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 203 Statistical Programming in R Nigeria > B.Sc. Data Science > Ch 11 > Lec 2 Tidy Data Philosophy, Grammar of Data Manipulation & Functional Programming with purrr Relational semantics of tidy formatting, pivoting mathematics, non-standard evaluation in dplyr, and higher-order functional mappings with purrr. 1. Formal Principles...

Module 11 • Subtopic 2
Subtopic 3Document Lecture

Statistical Modeling Syntax, Formula Interfaces & Model Diagnostics in R

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 203 Statistical Programming in R Nigeria > B.Sc. Data Science > Ch 11 > Lec 3 Statistical Modeling Syntax, Formula Interfaces & Model Diagnostics in R Wilkinson-Rogers formula algebra, model matrix contrast parameterizations, GLM link families, and comprehensive regression diagnostics. 1. The R Formula Interface & Design Matri...

Module 11 • Subtopic 3
Subtopic 1Document Lecture

Semiology of Graphics: Bertin's Visual Variables & Cleveland-McGill Perceptual Hierarchy

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 204 Data Visualization Theory Nigeria > B.Sc. Data Science > Ch 12 > Lec 1 Semiology of Graphics: Bertin's Visual Variables & Cleveland-McGill Perceptual Hierarchy Psychophysical foundations of quantitative encoding, retinal channel semiology, Cleveland-McGill perceptual accuracy ranking, and colorimetry uniformity. 1. Jacques Ber...

Module 12 • Subtopic 1
Subtopic 2Document Lecture

Grammar of Graphics Architecture: Coordinate Systems, Faceting & Statistical Transformations

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 204 Data Visualization Theory Nigeria > B.Sc. Data Science > Ch 12 > Lec 2 Grammar of Graphics Architecture: Coordinate Systems, Faceting & Statistical Transformations Wilkinson's layered compositional algebra, aesthetic mapping formalisms, statistical geometric transforms, coordinate projections, and Trellis faceting. 1. Leland W...

Module 12 • Subtopic 2
Subtopic 3Document Lecture

Interactive Dashboards, Visual Integrity Ratios & Executive Decision Communication

B.Sc. Data Science NUC CCMAS Benchmark (200 Level) DTS 204 Data Visualization Theory Nigeria > B.Sc. Data Science > Ch 12 > Lec 3 Interactive Dashboards, Visual Integrity Ratios & Executive Decision Communication Tufte's graphical integrity metrics, Shneiderman's interactive exploration mantra, reactive state management, and executive data presentation. 1. Edward Tufte's Visual Integr...

Module 12 • Subtopic 3
Subtopic 1Document Lecture

Ordinary Least Squares (OLS): Geometric Interpretation, Gauss-Markov Theorem & Proofs

DTS 301 • Supervised Statistical Learning Ordinary Least Squares: Geometry, Gauss-Markov Theorem & Diagnostics Linear model formulation, Euclidean subspace projection, Gauss-Markov BLUE proof, and leverage diagnostics. 1. Classical Linear Regression Model (CLRM) Specification The observational data generating process across $n$ observations and $p$ features is formulated in matrix notatio...

Module 13 • Subtopic 1
Subtopic 2Document Lecture

Ridge, Lasso & Elastic Net Regularization: L1 vs L2 Norms, Sparsity & Shrinkage

DTS 301 • Supervised Statistical Learning Ridge, Lasso & Elastic Net: L1 vs L2 Norms, Sparsity & Shrinkage Ill-conditioned design, Tikhonov regularization, subgradient calculus of L1 sparsity, and coordinate descent. 1. Failure Modes of OLS & The Bias-Variance Tradeoff When explanatory regressors are highly correlated (multicollinearity) or feature dimension $p$ approaches sample...

Module 13 • Subtopic 2
Subtopic 3Document Lecture

Generalized Linear Models (GLMs): Logistic Regression, Poisson Regression & Deviance Analysis

DTS 301 • Supervised Statistical Learning Generalized Linear Models: Logistic, Poisson & Deviance Analysis Exponential dispersion family unification, link functions, IRLS optimization, and likelihood deviance testing. 1. The Exponential Dispersion Family (EDF) Generalized Linear Models (GLMs) extend ordinary linear regression beyond Gaussian continuous outcomes to binary, count, and non-n...

Module 13 • Subtopic 3
Subtopic 1Document Lecture

Decision Trees: Impurity Measures (Gini, Entropy), Recursive Splitting & Cost-Complexity Pruning

DTS 302 • Tree-Based Methods & Ensembles Decision Trees: Impurity Criteria, Recursive Splitting & Cost-Complexity Pruning Recursive space partitioning, Gini and entropy concavity proofs, greedy search, and Breiman weakest-link pruning. 1. Non-Parametric Recursive Partitioning & Geometry Decision tree learning constructs an adaptive non-parametric model by partitioning a $p$-dimens...

Module 14 • Subtopic 1
Subtopic 2Document Lecture

Bootstrap Aggregation (Bagging) & Random Forests: Variance Reduction Proofs & Out-of-Bag Error

DTS 302 • Tree-Based Methods & Ensembles Bootstrap Aggregation & Random Forests: Variance Proofs & OOB Error Bootstrap sampling, mathematical proof of ensemble variance reduction, feature decorrelation, and Out-of-Bag validation. 1. The Bootstrap Principle & Bagging Mechanics Individual decision trees possess low bias when grown deeply, but exhibit high variance: small perturb...

Module 14 • Subtopic 2
Subtopic 3Document Lecture

Gradient Boosting Machines: Functional Gradient Descent, XGBoost Mechanics & Loss Regularization

DTS 302 • Tree-Based Methods & Ensembles Gradient Boosting Machines: Functional Gradient Descent & XGBoost Mechanics Functional optimization in L2 Hilbert space, second-order Taylor approximations, and regularized split gain formulations. 1. Boosting as Functional Gradient Descent Unlike bagging, which builds parallel independent models to reduce variance, boosting sequentially builds...

Module 14 • Subtopic 3
Subtopic 1Document Lecture

K-Means Clustering: Lloyd's Algorithm, Convergence Guarantees & Silhouette Diagnostics

DTS 303 • Unsupervised Learning & Manifold Learning K-Means Clustering: Lloyd's Algorithm, Convergence & Silhouette Diagnostics Within-cluster sum of squares minimization, alternating coordinate descent proof, K-Means++ seeding, and silhouette geometry. 1. Mathematical Formulation & Within-Cluster Sum of Squares (WCSS) In unsupervised learning, clustering partitions an unlabelled...

Module 15 • Subtopic 1
Subtopic 2Document Lecture

Density-Based & Hierarchical Clustering: DBSCAN, HDBSCAN & Agglomerative Dendrograms

DTS 303 • Unsupervised Learning & Manifold Learning Density-Based & Hierarchical Clustering: DBSCAN, HDBSCAN & Linkage Topological density reachability, noise isolation, HDBSCAN mutual reachability, and agglomerative Ward's variance linkage. 1. Limitations of Centroid Models & The Density Paradigm Centroid-based algorithms like K-Means implicitly assume that clusters form conv...

Module 15 • Subtopic 2
Subtopic 3Document Lecture

Nonlinear Dimensionality Reduction: t-SNE, UMAP & Spectral Graph Embeddings

DTS 303 • Unsupervised Learning & Manifold Learning Nonlinear Dimensionality Reduction: t-SNE, UMAP & Spectral Embeddings The manifold hypothesis, crowding problem resolutions, Riemannian fuzzy sets, and normalized Graph Laplacians. 1. The Manifold Hypothesis & Failure Modes of PCA The Manifold Hypothesis posits that high-dimensional observational data $\mathbf{x} \in \mathbb{R}^p...

Module 15 • Subtopic 3
Subtopic 1Document Lecture

Distributed Computing Principles: CAP Theorem, MapReduce Architecture & HDFS Storage

DTS 304 • Distributed Systems & Big Data Distributed Computing Principles: CAP Theorem, MapReduce & HDFS Distributed consensus trade-offs, formal proof of the CAP theorem, HDFS rack-aware replication, and MapReduce shuffling. 1. Distributed Systems Foundations & The CAP Theorem Distributed computing coordinates autonomous compute nodes connected via commodity networks to execute d...

Module 16 • Subtopic 1
Subtopic 2Document Lecture

Apache Spark Architecture: Resilient Distributed Datasets (RDDs), DAG Execution & Catalysts

DTS 304 • Distributed Systems & Big Data Apache Spark Architecture: RDDs, DAG Execution & The Catalyst Optimizer In-memory distributed computing, narrow vs wide lineage graphs, stage boundaries, and query optimization. 1. The In-Memory Computing Paradigm & Topology Hadoop MapReduce required intermediate disk spills and HDFS replication between stages. For iterative algorithms (Pag...

Module 16 • Subtopic 2
Subtopic 3Document Lecture

Modern Data Lakehouses: Parquet Columnar Storage, Delta Lake & Schema Enforcement

DTS 304 • Distributed Systems & Big Data Modern Data Lakehouses: Parquet Columnar Storage & Delta Lake ACID Evolution from data lakes to lakehouses, Apache Parquet columnar encoding, and Delta transaction log concurrency. 1. The Lakehouse Architectural Paradigm Enterprise data management historically oscillated between two distinct architectures: Data Warehouses (EDWs): Highly optimiz...

Module 16 • Subtopic 3
Subtopic 1Document Lecture

Stationarity, Autocorrelation (ACF/PACF) & Spectral Density Analysis

DTS 305 • Time Series Analysis & Forecasting Stationarity, Autocorrelation (ACF/PACF) & Spectral Density Analysis Covariance stationarity foundations, unit root tests, Yule-Walker PACF inversion, and Fourier spectral density. 1. Stochastic Processes & Stationarity Formulations A univariate time series is an empirical realization of a stochastic process $\{X_t\}_{t \in \mathbb{Z}}$...

Module 17 • Subtopic 1
Subtopic 2Document Lecture

ARIMA, SARIMA & Exponential Smoothing Formulations: Parameter Estimation & Forecasting

DTS 305 • Time Series Analysis & Forecasting ARIMA, SARIMA & Exponential Smoothing: Estimation & Forecasting Lag operator polynomial calculus, Box-Jenkins SARIMA identification, Kalman state-space estimation, and Holt-Winters ETS. 1. The Backshift Operator & ARMA Polynomial Calculus The algebraic analysis of linear time series models is formalized through the Backshift (Lag) O...

Module 17 • Subtopic 2
Subtopic 3Document Lecture

Dynamic Regression, ARCH/GARCH Volatility Modeling & Forecasting Evaluation Metrics (RMSE, MAPE)

DTS 305 • Time Series Analysis & Forecasting Dynamic Regression, ARCH/GARCH Modeling & Forecast Evaluation Spurious regression avoidance, ARCH/GARCH volatility clustering mechanics, and scale-independent MASE metrics. 1. Dynamic Regression & The Spurious Regression Hazard Dynamic regression models (ARIMAX) merge causal econometric regressors with autoregressive moving-average erro...

Module 17 • Subtopic 3
Subtopic 1Document Lecture

Enterprise Data Architecture: Data Warehousing, Data Marts & Medallion Architecture

DTS 306 • SIWES / Enterprise Data Pipelines Enterprise Data Architecture: Warehousing, Marts & Medallion Design Kimball dimensional modeling, Slowly Changing Dimensions (SCD Type 2), and the multi-hop Bronze-Silver-Gold lakehouse. 1. Foundational Data Warehousing: Inmon vs Kimball Enterprise data engineering coordinates heterogeneous operational transaction databases (OLTP) into unified a...

Module 18 • Subtopic 1
Subtopic 2Document Lecture

Workflow Orchestration: Directed Acyclic Graphs (DAGs), Airflow & Idempotent Scheduling

DTS 306 • SIWES / Enterprise Data Pipelines Workflow Orchestration: DAGs, Apache Airflow & Idempotent Scheduling Topological sorting in task graphs, Airflow scheduler architecture, execution windows, and idempotent writes. 1. The Orchestration Imperative & Cron Failure Modes Data engineering pipelines coordinate interdependent computational tasks: extracting logs, cleansing data, refr...

Module 18 • Subtopic 2
Subtopic 3Document Lecture

Data Quality Engineering: Great Expectations, Schema Drift & Data Observability Monitoring

DTS 306 • SIWES / Enterprise Data Pipelines Data Quality Engineering: Great Expectations & Data Observability Declarative data contracts, statistical distribution tests, schema drift mitigation, and observability metrics. 1. The High Cost of Bad Data & Data Observability In software engineering, unit and integration tests validate whether application code functions correctly. In data...

Module 18 • Subtopic 3
Subtopic 1Document Lecture

Multilayer Perceptrons (MLPs): Activation Functions, Cross-Entropy Loss & Forward Propagation

DTS 401 • Deep Neural Networks Multilayer Perceptrons: Activation Functions & Forward Propagation Universal approximation theorems, non-linear activation mechanics, matrix tensor propagation, and cross-entropy loss. 1. Perceptron Foundations & The Universal Approximation Theorem The artificial neuron originates from the Frank Rosenblatt (1958) linear threshold perceptron $f(\mathbf{x}...

Module 19 • Subtopic 1
Subtopic 2Document Lecture

Reverse-Mode Automatic Differentiation & Backpropagation: Computational Graphs & Gradient Derivations

DTS 401 • Deep Neural Networks Reverse-Mode Automatic Differentiation & Backpropagation Calculus Computational directed graphs, forward vs reverse accumulation, matrix chain rule derivations, and gradient checking. 1. Computational Graphs & Differentiation Paradigms Training deep neural networks requires computing the exact gradient of a scalar objective loss $\mathcal{L} \in \mathbb{...

Module 19 • Subtopic 2
Subtopic 3Document Lecture

Regularization in Deep Learning: Dropout Mechanics, Weight Decay & Batch Normalization Derivations

DTS 401 • Deep Neural Networks Regularization in Deep Learning: Dropout, Weight Decay & BatchNorm Inverted dropout ensemble mathematics, decoupled weight decay, and the full analytical derivations of Batch Normalization. 1. The Overparameterization Regime & Weight Decay Modern deep learning models operate in the overparameterized regime, where parameter count $P$ exceeds sample size $...

Module 19 • Subtopic 3
Subtopic 1Document Lecture

Text Preprocessing, Tokenization, N-Gram Language Models & TF-IDF Vector Spaces

DTS 402 • Natural Language Processing Text Preprocessing, Subword Tokenization & TF-IDF Vector Spaces Byte Pair Encoding algorithms, statistical N-gram smoothing formulations, and high-dimensional TF-IDF vector spaces. 1. Linguistic Preprocessing & Subword Tokenization Natural Language Processing translates unstructured textual strings into quantitative vector spaces. Classical word-l...

Module 20 • Subtopic 1
Subtopic 2Document Lecture

Word Embeddings: Word2Vec (Skip-Gram, CBOW), GloVe & Vector Space Semantics

DTS 402 • Natural Language Processing Word Embeddings: Word2Vec (Skip-Gram, CBOW), GloVe & Vector Semantics The distributional hypothesis, negative sampling mathematics, global co-occurrence factorizations, and linear analogies. 1. The Distributional Hypothesis & Dense Vector Spaces Classical bag-of-words and one-hot encodings represent words as mutually orthogonal basis vectors in a...

Module 20 • Subtopic 2
Subtopic 3Document Lecture

Sequence Models to Transformers: Recurrent Networks, Attention Mechanisms & Document Classification

DTS 402 • Natural Language Processing Sequence Models to Transformers: RNNs, Self-Attention & Classification Backpropagation through time, LSTM gating, scaled dot-product attention mathematics, and multi-head representations. 1. Recurrent Neural Networks & The Vanishing Gradient Dilemma Standard feedforward neural networks treat inputs as independent vectors, lacking mechanisms to ret...

Module 20 • Subtopic 3
Subtopic 1Document Lecture

Potential Outcomes Framework (Rubin Causal Model) & Causal Directed Acyclic Graphs (DAGs)

DTS 403 • Causal Inference & Experimental Design Potential Outcomes Framework & Causal Directed Acyclic Graphs Rubin Causal Model, SUTVA, Pearl's Structural Causal Models, d-Separation, and Backdoor Identification. 1. The Causal Hierarchy & The Limits of Statistical Association Classical statistical modeling and machine learning excel at estimating conditional expectations: discov...

Module 21 • Subtopic 1
Subtopic 2Document Lecture

Confounding Control: Propensity Score Matching, Inverse Probability Weighting & Doubly Robust Estimation

DTS 403 • Causal Inference & Experimental Design Confounding Control: Matching, IPW & Doubly Robust Estimation Balancing Scores, Propensity Score Matching, Inverse Probability Weighting, and Augmented IPW (Doubly Robust Estimation). 1. The Propensity Score as a Balancing Score In observational studies, treatment assignment is confounded by high-dimensional baseline covariates $\mathbf...

Module 21 • Subtopic 2
Subtopic 3Document Lecture

Quasi-Experimental Methods: Difference-in-Differences (DiD), Instrumental Variables & A/B Testing Design

DTS 403 • Causal Inference & Experimental Design Quasi-Experimental Methods: DiD, Instrumental Variables & A/B Testing Difference-in-Differences, Parallel Trends, 2SLS, Local Average Treatment Effects (LATE), and CUPED Variance Reduction. 1. Difference-in-Differences (DiD) Estimation When unobserved confounders vary across units but remain constant over time, cross-sectional matching...

Module 21 • Subtopic 3
Subtopic 1Document Lecture

Content-Based Filtering & Utility Matrix Representations: Cosine Similarity & TF-IDF Profiles

DTS 404 • Recommender Systems & Ranking Content-Based Filtering & Utility Matrix Representations Utility Matrix Sparsity, Explicit/Implicit Feedback, TF-IDF Item Profiles, User Centroid Embeddings, and Cosine Vector Spaces. 1. The Recommendation Problem & Utility Matrix Topology Information overload across digital platforms necessitates algorithmic filtering systems that curate pe...

Module 22 • Subtopic 1
Subtopic 2Document Lecture

Collaborative Filtering: User-Based, Item-Based & Matrix Factorization with SVD/ALS

DTS 404 • Recommender Systems & Ranking Collaborative Filtering: Neighborhoods & Matrix Factorization User-Based and Item-Based CF, Regularized Latent Factor Models, SVD Biases, Stochastic Gradient Descent, and Alternating Least Squares. 1. The Collaborative Filtering Paradigm Unlike content-based filtering, which requires domain-specific metadata features, Collaborative Filtering (CF...

Module 22 • Subtopic 2
Subtopic 3Document Lecture

Ranking Metrics & Cold-Start Solutions: Precision@K, Recall@K, MAP, NDCG & Hybrid Recommenders

DTS 404 • Recommender Systems & Ranking Ranking Metrics, Cold-Start Solutions & Hybrid Recommenders Top-K Evaluation, Precision@K, Recall@K, MAP, NDCG Formulations, Cold-Start Remediation, and Two-Stage Funnel Architectures. 1. The Fallacy of Rating Prediction: Why Ranking Dominates Early recommender research focused primarily on minimizing regression error metrics, such as Mean Squar...

Module 22 • Subtopic 3
Subtopic 1Document Lecture

Ethical Foundations in Data Science: Algorithmic Bias, Fairness Metrics & Disparate Impact Analysis

DTS 405 • Data Governance & Ethics Ethical Foundations: Algorithmic Bias & Fairness Metrics Statistical Parity, Equalized Odds, Predictive Parity, The Impossibility Theorem of Fairness, and Bias Mitigation. 1. The Mechanics of Algorithmic Bias in Machine Learning Machine learning models do not operate in a moral or sociological vacuum. When trained on historical observational records,...

Module 23 • Subtopic 1
Subtopic 2Document Lecture

Nigeria Data Protection Act (NDPA 2023): Principles, Data Controller Obligations & NDPC Enforcement

DTS 405 • Data Governance & Ethics Nigeria Data Protection Act (NDPA 2023): Principles & Governance Statutory Architecture, Data Controller Obligations, Data Subject Rights, NDPC Enforcement, and Cross-Border Transfers. 1. Legislative Evolution: From NDPR 2019 to NDPA 2023 Data privacy in Nigeria underwent a historic structural transformation with the signing into law of the Nigeria D...

Module 23 • Subtopic 2
Subtopic 3Document Lecture

Privacy-Preserving Data Science: k-Anonymity, l-Diversity, t-Closeness & Differential Privacy Mathematics

DTS 405 • Data Governance & Ethics Privacy-Preserving Data Science: Anonymization & Differential Privacy Quasi-Identifiers, k-Anonymity, l-Diversity, t-Closeness, and Differential Privacy Laplace/Gaussian Mechanisms. 1. The Re-Identification Hazard & Linkage Attacks Naively removing Direct Identifiers (such as Names, National Identity Numbers (NIN), Bank Verification Numbers (BVN)...

Module 23 • Subtopic 3
Subtopic 1Document Lecture

Capstone Research Formulation: Problem Definition, Literature Gap & Methodology Protocol

DTS 406 • Undergraduate Capstone Project Capstone Research Formulation & Methodology Protocol Problem Framing, FINER Criteria, PRISMA Systematic Reviews, Scientific Gaps, and Pre-Registered Protocols. 1. Scientific Problem Formulation in Data Science An undergraduate capstone in Data Science represents the intellectual culmination of the Bachelor of Science degree. Unlike introductory cou...

Module 24 • Subtopic 1
Subtopic 2Document Lecture

System Implementation & Model Validation: Cross-Validation Schemes, Ablation Studies & Reproducibility

DTS 406 • Undergraduate Capstone Project System Implementation, Validation Schemes & Ablation Studies Modular Scientific Engineering, Group/Spatial K-Fold, Purged Walk-Forward Splits, Component Ablation, and Dietterich's 5x2cv Test. 1. Scientific Software Engineering & Reproducibility The reproducibility crisis in modern computational science stems primarily from ad-hoc, non-determini...

Module 24 • Subtopic 2
Subtopic 3Document Lecture

Scientific Thesis Writing, Oral Defense Preparation & Production Deployment Architecture

DTS 406 • Undergraduate Capstone Project Thesis Writing, Oral Defense & Production Deployment Dissertation Architecture, Viva Voce Defense Protocol, ONNX Serialization, FastAPI Microservices, and PSI Drift Monitoring. 1. Academic Thesis Architecture & Scientific Discourse The final graduation requirement for the B.Sc. Data Science degree is the submission and defense of a formal Bache...

Module 24 • Subtopic 3

Instructor

PAMCET Learning Team

PAMCET Learning Team

Institutional Course Curation · PAMCET

The PAMCET Learning Team is the institutional curriculum and content group behind course publishing on this platform. PAMCET (Professional and Academic Management Centre for Education and Training) is operated by Digitpen Hub Ltd, a company registered in Lagos, Nigeria. Courses credited to the PAMCET Learning Team are structured, organized, and published centrally by the platform rather than authored by a single named individual instructor — each course's syllabus, learning outcomes, and content are reviewed before publication under this institutional account. For questions about a specific course or to report a content issue, please use the platform's Help & Support or Contact Us pages.

Qualifications:

PAMCET Learning Team — Institutional Course Curation, Digitpen Hub Ltd

My Skills:

  • Curriculum Curation
  • Course Content Review
  • Learning Outcome Design
  • Course Publishing & Quality Assurance
  • Platform Content Operations

Reviews

No reviews available
Bachelor of Science (B.Sc.) in Data Science: Mathematical Foundations, Statistical Learning & Big Data Analytics (NUC CCMAS Standards)
Free100% Off
Certificate available separately for $7 after completion

Course Includes

24 Chapters
72 Lectures
1 week
Skill Level advanced
Taught in English
Lifetime Access
Certification of Completion
Category: University - Data Science
Share On

Download Syllabus & Study Guide

Receive the full lecture syllabus, reading list, and milestone schedule for Bachelor of Science (B.Sc.) in Data Science: Mathematical Foundations, Statistical Learning & Big Data Analytics (NUC CCMAS Standards).

By submitting, you agree to receive educational updates. You can unsubscribe anytime. Read our Privacy Policy.