Machine Learning Part 5

In this blog, we will describe another example that utilize KMeans & Spark to determine locations.

Before that, we would suggest you to got to previous blogs -

• https://querydb.blogspot.com/2019/12/machine-learning-part-4.html
• https://querydb.blogspot.com/2019/12/machine-learning-part-3.html
• https://querydb.blogspot.com/2019/12/machine-learning-part-2.html
• https://querydb.blogspot.com/2019/12/machine-learning-part-1.html

In this blog, we will analyze and try to make predictions on Fire detection GIS Data- https://fsapps.nwcg.gov/gisdata.php

1. We will have historical data of Wild fires.
2. And, we will try to analyze. That is eventually helpful to reduce response time in case of fire, reduce cost, reduce damages due to fire, etc.

Fire can grow exponentially based on various factors like - Wild life, Wind Velocity, terrain surface, etc. Incident tackle time is limited by various factors one of which is moving firefighting equipment. If we are plan in advance where to place equipment's such that it is close to a fire prone area. It is possible to reduce response time.

So, we can analyze the historical data and try to group them in a cluster, which has a centroid. This centroid can be helpful to determine location to make water tanks, keep fire equipment's, etc. As, this location will be close to the locations where fire has affected in past. 

K-Means is a clustering algorithm that fits suitable in this scenario and we will try to use the same.

Mostly, what we have seen a simplistic singular pipeline. But, we can have a pipeline implemented as per Rendezvous Architecture, where-in, multiple models can be applied to an request, and a user defined logic can be applied to return the best fit results to requester.  For example, We might want to locate a correct GPS address for some grocery store, or in case of Fire: we may require to alert a Fire site which can reach the location quickly, rather than the location which is close to Fire. So, an incoming request may be required to optimize with different models for different objectives. Like - Speed vs Accuracy.

We will discuss more about Rendezvous Architecture in future blogs.

This example is pretty much same as what we have seen in previous blog. But, Refer - @ https://github.com/dinesh028/SparkDS/tree/master/src/indore/dinesh/sachdev/forestfire/

Next Blog - https://querydb.blogspot.com/2019/12/machine-learning-part-6.html