ICT705 Applied Data Analytics: midterm study guides

Lectures 1–7 rebuilt as concept-ordered pages with the homework and quiz questions as practice. Midterm: 20 October 2026. Formulas are given in the exam; the skill tested is choosing the right one and showing every step.

Read in this order. Each page builds on the one before it.

  1. Data, vectors and distributions (Lectures 1–2): data matrix, norms, dot product, projection, distance; random variables, pmf, cdf, pdf; mean, median, IQR, variance.
  2. From covariance to PCA (Lectures 3–4): joint distributions, covariance matrix, correlation, normal distribution, eigen-decomposition, projection, PCA.
  3. From kernels to kernel PCA (Lectures 5–6): curse of dimensionality, feature maps, kernel functions and matrices, norms and centring in feature space, kernel PCA.
  4. Linear regression (Lecture 7): least squares with one predictor, residuals and RMSE, the projection view, multiple regression and the normal equations.

Practice covered: HW1 Q1–Q6, Quiz 1, HW2 Q1–Q6, Quiz 2, the practice sheet in quizzes/notes.md, and every worked example on the slides. Errors found in the 2025-era answer keys are flagged in amber boxes on each page.