Preparation

%%{init: {'gantt': {'axisFormat': '%Y-%m-%d'}}}%%
gantt
    title Big Data & Analytics — Prep plan (everything done before teaching)
    dateFormat YYYY-MM-DD
    axisFormat %Y-%m-%d
    excludes weekends

    section Milestones
    First teaching day (Sessions 1–2) :milestone, teach, 2026-03-23, 0d

    section Week 1
    Foundations (Session 1) :found, 2026-03-02, 2d
    Structured Data (Sessions 2–7) :struct, after found, 3d

    section Week 2
    Big Data (Sessions 8–9) :big, 2026-03-09, 3d
    Organizations (Sessions 10–11) :org, after big, 2d

ImportantMaterials

See /home/gerit/workspace/projects/bda/nextcloud (lecture_x)

Preparation overview

Generally: generation of PDFs: local only?

  • Include “must-learn” markers: <!-- _class: must_learn -->

Plan to record timing (data in the repository) and data from google surveys - TODO: session/exercise structure should be reflected in each survey -> run separate surveys for the lecture/exercise to use the timestamps for the duration

TODO: html notebook creation (via make exercises) appears to be overriden by the quarto render (e.g., html assignemnts are slides not jupyter notebooks on the overview page)

Consistently include “vocabulary” sections at the end (like in regression 1)

Session 0 — Course logistics

Session 1 — The rise of analytics

TODO: Exercise session: divide class (one part does the setup/exercise, the other reads the paper, then switch)

Session 2 — Exploratory data analysis

Session 3 — Analytical data architecture

Session 4 — Regression I

Session 5 — Regression II

Session 6 — Machine learning I

Session 7 — Machine learning II

Session 8 — Big Data I

Session 9 — Big Data II

Session 10 — Governance of data analytics

Session 11 — Recap, exam preparation, Q&A

Cross-cutting tasks

TBD:

    • show and read code for the analysis
    • include code reading and interpretation, maybe even pseudo-code writing in the exam (and exercises)
    • surveys during the lecture: vote for “additional exercise/more difficult exercise”

Work through the different parts (easier to identify materials for a part instead of comparing multiple lectures for multiple topics)

Slides:

Exercise sheets:

https://www.coursera.org/learn/data-analysis-with-python - python, exploratory data analysis -> accessible (good examples?!)

Switch to font Inter (Google: https://fonts.google.com/specimen/Inter) - similar to official Sergoe UI

inspiration for model choice/test/exam questions:

Use Cases Quiz:

Notes (Quarto/revealjs)

Hide slide in PDF export:


::: {.content-hidden when-profile="pdf"}

## temp

:::