What Data Science Actually Is & The Data Science Workflow

Data Science Fundamentals

Chapter 1 · What Data Science Actually Is & The Data Science Workflow

This course assumes real, working Python fluency — py1–py4 already cover variables, control flow, functions, files, JSON, and packaging in depth, and none of that gets re-taught here. This chapter picks up almost exactly where py1-9's own brief pip/virtual-environment coverage left off, moving from "how to manage Python packages in general" into the specific stack this course is actually about.

Data Science Is Not a Synonym for Machine Learning

Four terms get used almost interchangeably in casual conversation, and they shouldn't be: Business Intelligence (BI) is retrospective reporting on known questions — dashboards showing what already happened. Data Analytics is answering a specific business question with data — "did the redesign increase signups?" Machine Learning is building models that make predictions or find patterns automatically from data. Data Science is the broader discipline that combines statistics, programming, and domain knowledge to extract insight from data — and machine learning is one tool inside that discipline, not a synonym for the whole thing.

Why this course draws the line where it does
A huge amount of real, valuable data science work — cleaning a messy dataset, understanding what it actually contains, finding the pattern that answers the question someone actually asked — happens before any model gets built, and plenty of real data science projects never need a model at all. This course covers that entire "before modeling" layer in depth. Building predictive models is a genuinely separate skill, covered start to finish in ml1.

The Data Science Workflow

Five stages, in order, form the backbone this whole subject is organized around:

  • Collect — gather data from files, APIs, databases, or the web. Covered only briefly here; a full hands-on project (a real web-scraping tool) is deferred to dsproj1, deliberately chosen there as a good first project because it needs no ML at all.
  • Clean — handle missing values, duplicates, inconsistent types, and messy strings so the data can actually be trusted. Covered in full in ds1-4.
  • Explore — understand what the data actually contains: summary statistics, distributions, relationships, visualizations. Covered in ds1-5 through ds1-9 — the largest share of this course by chapter count, because it's the largest share of real data science work.
  • Model — build a predictive or classification model. Deliberately out of scope for this course — this is ml1's entire job, start to finish.
  • Communicate — turn findings into something another person can actually use: a chart, a report, a clear conclusion. Threaded through ds1-7–ds1-9 and the capstone rather than given its own dedicated chapter.

    • Jupyter Notebooks — The Standard Data Science Environment

    py1-1 introduced running Python as a plain script, top to bottom, once. Data science work is exploratory and iterative by nature — you look at a dataset, try something, look at the result, adjust, try again — and a plain script re-run from the top every time that happens is genuinely slow to work with. A Jupyter notebook is a document made of individual, independently runnable cells: code cells that keep their own output (including inline charts) visible directly beneath them, and text cells for notes, mixed freely in one file. Running one cell at a time, keeping earlier results in memory, and only re-running what actually changed is the standard way real data science work gets done — not a replacement for py1's own script model in general, just the better-fitting tool for this specific, exploratory kind of work.

    The Python Data Science Stack — A First Look

    LibraryJobCovered in
    NumPyFast, array-based numerical computingds1-2
    pandasLabeled, tabular data — the DataFrameds1-3 through ds1-5
    MatplotlibThe foundational plotting libraryds1-7
    SeabornStatistical plots, built on Matplotlibds1-8
    scikit-learnClassical machine learning modelsNot this course — ml1
    Why this course's own chapter order matches the table above
    Every chapter in this course exists to deliver one row of that table in depth. By the time ds1-9 formally names the EDA methodology, every tool it draws on will already be familiar — nothing in this course's own capstone (ds1-10) introduces a new library for the first time.

    A Short, Grounding Example

    A retail company wants to know why weekend sales dipped last month. A data scientist starts by collecting the relevant sales records (a database export, in this case — ds1-3's own CSV-reading skills apply directly). The raw export has missing store IDs and a few duplicated rows from a failed sync — cleaning (ds1-4) fixes that before anything else happens. Exploring the cleaned data (ds1-5–ds1-9) reveals the dip is concentrated at one specific store, not company-wide — a pattern no amount of staring at raw numbers would have surfaced as quickly as a grouped chart does. No model was ever built; the answer came entirely from disciplined cleaning and exploration. Communicating that finding — one clear chart, one clear sentence — closes the loop.

    Hands-On Exercises

    Exercise 1

    Explain, using this chapter's own four-term distinction, why "I want to learn data science" and "I want to learn machine learning" are not the same request, and identify which of the four terms this course itself is scoped around.

    📄 View solution
    Exercise 2

    Using this chapter's own five-stage workflow, explain which stage is deliberately marked out of scope for this course, which stage gets the most chapters, and why that allocation matches this chapter's own stated reasoning.

    📄 View solution
    Exercise 3

    Explain why a Jupyter notebook's cell-based execution model is a better fit for data science work than py1-1's own plain top-to-bottom script model, without concluding that notebooks make script-based Python obsolete in general.

    📄 View solution

    Chapter 1 Quick Reference

    • Data science ≠ machine learning — ML is one tool inside the broader discipline of data science, not a synonym for it
    • The workflow: Collect → Clean → Explore → Model → Communicate — this course covers everything except Model (that's ml1's job)
    • Jupyter notebooks — cell-based, iterative, inline output — the standard exploratory environment, distinct from py1-1's own script model
    • Stack roadmap: NumPy (ds1-2) → pandas (ds1-3–ds1-5) → Matplotlib (ds1-7) → Seaborn (ds1-8)
    • Next chapter: NumPy Fundamentals