This is the homework 3 for Data Science ADM
In this homework we analyze the text of Airbnb property listings and building a search engine. In particular we build three increasingly performing search engines:
-
The first give back all the announcements that contain the key words queried
-
The second give back the top k announcements more similar with the query. k is decided by the user
-
The third do the same thing of the previous but use another idea of "similar", decided by us
For this purpose we use the data in the Kaggle site. Dataset contains more than 18,000 property listings from Texas, United Staes. Given below are the data fields: Rate per night, Number of bedrooms, City, Joining month and year, Longitude, Latitude, Property description, Property title, Property URL.
Homework 3 - Group 13.ipynb
This script is divided in this way:
- First we download the data, then we clean it by stopword, punctuation and those that seemed us wrong. At the end we store each review in a single tsv file.
- Then we do all the search engines requested.
- At the end there's a bonus part with the map.
A detailed description is provided in the file.
If you can't see the map or the tables woth the results look strange, please try to see at this link: NB Viewer