Data Mining and Process mining
In this post we’ll directly start with what process science is and how this is connected with data science. You can read my last post on brief about motivation behind exploring new field.
Before we go into Process mining let’s try to understand in brief what Process Science and how is this related to Data Science.
The diagram above shows different sub-areas of data sceince and process science and intersection. As we see that data science is a well established field with state-of art tools and techniques avaliable but process science this is not the case, tools are still getting developed, so more research required
So, lets go back to the basic question - what is the process that people really follow?
This picture above shows a entry gate of a park with a gate to stop cyclists to enter the park so that is process but if you see the actual path its different story. This means that people actually follow something other than the defined path. So question comes that how do we know what actual path user has followed? With advancement in ICT one thing the best way to get which path was followed is to study the event data from IT system.
So looking into above picture, more questions comes:
- What is the process that people really follow?
- What are the bottlenecks in my process?
- What do people (or machines) deviate from the expected or idealized process?
Just to summarise, in this post we understood in brief about data sceince and process science, junction of these two. We also tried to see how people deviate from defined processes.
In my next post we’ll dig deep into data mining and process mining and will look into various process mining techniques.
Thanks for reading!
IMG Source: “Process mining” by TU Eindhoven, © by Springer