openvenues / libpostal

A C library for parsing/normalizing street addresses around the world. Powered by statistical NLP and open geo data.
MIT License
4.09k stars 421 forks source link

Floor parsing issue (US address) #580

Open codekrolik2 opened 2 years ago

codekrolik2 commented 2 years ago

Hi!

I was checking out libpostal, and saw something that could be improved.


My country is

USA

Here's how I'm using libpostal

Parsing address components


Here's what I did

Tried to parse "777 S. Figueroa St, 41st Floor Los Angeles CA 90017"


Here's what I got

HOUSE_NUMBER: 777 ROAD: s. figueroa st 41st floor CITY: los angeles STATE: ca POSTCODE: 90017


Here's what I was expecting

HOUSE_NUMBER: 777 ROAD: s. figueroa st LEVEL: 41st floor CITY: los angeles STATE: ca POSTCODE: 90017


For parsing issues, please answer "yes" or "no" to all that apply.

LEVEL: 41st floor HOUSE_NUMBER: 777 ROAD: s. figueroa st CITY: los angeles STATE: ca POSTCODE: 90017

HOUSE_NUMBER: 777 ROAD: s. figueroa st CITY: los angeles STATE: ca POSTCODE: 90017


Here's what I think could be improved

The addresses in which Floor information is specified in the same way can be found everywhere. It would be great to add parsing support for this format as well.

ktmohammedsulaim commented 1 year ago

Same here. Any update on this issue? It parses wrongly in some cases

brianmacy commented 1 year ago

I believe the latest Senzing alternative model handles that.

{"CITY":["LOS ANGELES"],"HOUSE_NUMBER":["777"],"LEVEL":["41ST FLOOR"],"POSTCODE":["90017"],"ROAD":["S FIGUEROA ST"],"STATE":["CA"]}