Naive Bayes in practice

The first step is to choose some data that will be used for classification. We have chosen some data from the UK Government data website at http://data.gov.uk/dataset/road-accidents-safety-data.

The dataset is called Road Safety - Digital Breath Test Data 2013, which downloads a zipped text file called DigitalBreathTestData2013.txt. This file contains around half a million rows. The data looks as follows:

Reason,Month,Year,WeekType,TimeBand,BreathAlcohol,AgeBand,GenderSuspicion of Alcohol,Jan,2013,Weekday,12am-4am,75,30-39,MaleMoving Traffic Violation,Jan,2013,Weekday,12am-4am,0,20-24,MaleRoad Traffic Collision,Jan,2013,Weekend,12pm-4pm,0,20-24,Female

In order to classify the data, we have modified both the column ...

Get Apache Spark 2: Data Processing and Real-Time Analytics now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.