StopFire is a campaign started by Monash University to predict and stop the fire in Victorian…

StopFire is a campaign started by Monash University to predict and stop the fire in
Victorian cities. They have employed sensors in different cities of Victoria and have
collected a large amount of data. The data is so big that their techniques have failed to
provide the results on time to predict fire. They have hired us as the data analyst to
migrate their data to the NoSQL database (MongoDB). They want us to analyse their data
and provide them with results. In addition, they want us to build an application, a
complete setup from streaming to storing and analyzing the data for them using Apache
Kafka, Apache Spark Streaming and MongoDB.
What you are provided with
? Five datasets:
? hotspot_historic.csv
? climate_historic.csv
? hotspot_AQUA_streaming.csv
? hotspot_TERRA_streaming.csv
? climate_streaming.csv
? These files are available in Moodle in the Assessment section in Assignment Data
folder.
Information on Dataset
Climate data is recorded on a daily basis whereas Fire data is recorded based on the
occurrence of a fire on a particular day. Therefore, for one climate data, there can be zero
or many fire data.
The data is NOT row per weather station basis. You can simply think of it as, Station 1 was
reporting data for X number of days and then Station 2 started reporting data because
Station 1 was shut down for instance.
Attachments:
3089470_1_FIT5148—Assignment-Specification-3.pdf
3089470_2_Assignment-Data-20190521.zip