Four things, in order of how much they are worth:
Exploration and data together carry 26 of 40 points. Prior work carries 4.
A literature review is one of the things generative AI writes most easily, so it demonstrates little. Digging into your actual data does.
Prior work is about one slide: what is already known about your topic and which questions it leaves open. Cite what you drew on, and put the references on a closing slide rather than inline.
There is no separate code submission. Whatever is in your group GitHub repository at the deadline (Oct 25, 2026, 11:59 PM US Eastern Time) is what gets graded, and it should be the analysis behind the presentation you just handed in.
Do not paste code into the deck.
Roughly 10–15 slides, most of them exploratory analysis.
A specific, answerable question — with who cares about the answer and what changes if you find one. "We will analyze housing data" is not a question.
What is already known on the topic, and what remains unanswered. Make the gap you are addressing explicit. No citation minimum; any consistent style.
Sources with provenance and access method, key variables with types and units, and how anything is joined if you are joining. Must meet the size and shape requirements.
Notes should carry the detail a slide cannot hold — row counts, coverage, licensing.
Show what the data actually looks like and what surprised you. Distributions, relationships, outliers, missingness, and anything that will shape how you model.
Deliberate chart choices suited to the variables, captions that carry the takeaway, and interpretation in the notes. Plotting everything once and moving on scores poorly; finding something and explaining why it matters scores well.
Given what the exploration showed, what do you expect to model and why. This is a direction, not a commitment — Deliverable 2 is where methods get settled.
Submit a single PDF. A normal "export to PDF" throws your presenter notes away — export using the notes layout instead, so each page shows a slide with its notes underneath.
Open the PDF before you submit and confirm your notes are actually in it. Every semester someone submits slides only.
How Deliverable 1 is graded, out of 40.