engleski

Exploratory and Data Mining Analysis in Big Data Environment

s global datasphere is growing faster every year, data analysis is becoming an increasingly challenging task. This paper goes through a big data exploratory analysis and data mining workflow using a distributed data processing environment. Specifically, it presents an analysis of tips in New York City's taxi rides from January 2009 to June 2016 using the R programming language and the Apache Spark platform. Although twenty-seven features have been explored and three different models have been trained, the research shows that predicting tips in taxi rides is a difficult task and that tipping is mostly motivated by social pressure.

Apache Spark ; R ; big data ; sparklyr ; exploratory data analysis ; data mining ; tipping system ; taxi rides

nije evidentirano