search

soton-data-mining / job-salary-prediction

A regression problem, predicting salaries of jobs in UK based on various criteria
8 stars 3 forks source link

Further cleaned locations based on Charles update #20

Closed utkuozbulak closed 7 years ago

utkuozbulak commented 7 years ago

@charlienewey ' bad practice to do `type(x) == "<class ...>" ' Updated :heart:

utkuozbulak commented 7 years ago

This is a basic cleaning on stuff google couldn't find but was obvious. Some stats are below, uncomment print statements to see what is updated

Some stats:

field 'town' total empty records: 148409 updated: 77348

Examples: Found on raw location, updated on cleaned Found: Bristol Updated: Bristol Found: Grimsby Updated: Grimsby Found: Edinburgh & Lothian Updated: Edinburgh Found: City of London Updated: London

field: 'region' total empty records: 71815 updated: 30280

Examples: Found on raw location, updated on cleaned: Found: Hampshire-wide Updated: Hampshire Found: Staffordshire Staffordshire England Updated: Staffordshire Found: Essex Updated: Essex Found: Berkshire Updated: Berkshire Found: Basingstoke Hampshire England Updated: Hampshire