datasets

Contents

• data.html: pure raw data from the data source. Retrieved using inspect element and save the HTML.
• raw_data.csv: table extracted from data.html. Columns are fixed (see schema)
• clean_data.csv: cleaned with these steps in order
  • Set lower case of feature with string data type
  • Normalize salary to monthly IDR
  • Fix company name
  • Remove duplicates

Note that in clean_data.csv, no outlier were removed. Please see notebook for more details on data cleaning.

Schema

1. role: str. Role/title of the job
2. company: str. Not all of the companies are disclosed, only Gojek, Shopee, Tiket.com, Traveloka, Tokopedia, Bukalapak companies are disclosed by default. Undisclosed company is set as Purchase to unlock 👆
3. years_of_exp: int. Years of experience, ranging from 0 to 20 or more
4. city: str. City where the salary owner is living
5. country: str. Country code such as "ID", "SG", "MY", "DE"
6. gender: str. Male, female, or prefer not to tell
7. currency: str. Salary currency code such as "IDR" or "USD"
8. salary: int. The salary amount based on the currency and the period
9. mode: str. Gross or net salary
10. period: str. Monthly or annually paid salary
11. compensation: str. Non-cash compensation
12. verified: str. Whether the salary owner also attach salary slip during the data submission