What 218,812 Airbnb reviews talk about

A Python check and keyword scan of a large review dataset. It turns free text into a service-quality snapshot, with aggregate counts only.

Role
Analyst (own analysis of course data)
Period
Built in 2026 · course data
Stack
Python, pandas, Text analysis
Evidence
Published
Question
How can a large set of review text become a service-quality snapshot that a host team can trust?
Data
218,812 reviews of 4,760 listings by 197,439 reviewers, 15 Feb 2010 – 10 Sep 2022.
Method
A Python script checks row counts, schema, dates, duplicate IDs and missing values. Then it counts reviews by listing, year and comment length, and scans the text for six keyword groups.
ResultPublished
No duplicate review IDs and no missing comments. Location, cleanliness, host contact and comfort are the strongest themes.

The data comes from a COMM 414 assignment. I made this analysis in July 2026; it is not the assignment that the group submitted. The scan is keyword-based and directional. It is not sentiment modelling or proof of customer retention. The page shows no raw comments or reviewer names.

Share of reviews that mention each theme

Share of reviews that mention each theme
ItemValue (%)
Location48.8
Clean33.6
Host31.95
Comfort25.68
Check-in10.1
Value4.31