Data and Processing
Data source: Fire Department Calls for Service, San Francisco (Our dataset was downloaded on 02/18/2019)
| Attribution | Value | Sreenshot |
|---|---|---|
| Size | 1.7 GB |  |
| Rows | 4,886,220 |  |
| Columns | 34 |  |
| Data Types | String, Number, Date, Datetime, Boolean, Location (Latitude and Longitude) | |
Data Processing: (See script source code)
1. Filtered "City" column with key word San Francisco.
2. Filtered "Call Date" column to get the year from date string and removed data not in 2016 and 2018.
3. Filtered "Call Type Group" column with key word Fire.
4. Removed unused columns such as "Call Number", "Unit ID", "Incident Number", "Location" and "RowID".
5. Removed "City" and "Call Type Group" columns since we did filtering on those columns and they only remained the records of key words we've used.
6. Aggregated a new column "Time Diff" using "Received DtTm" and "On Scene DtTm", grouped them by "Call Type", "Year", and "Quarter".