This content originally appeared on DEV Community and was authored by DEV Community
Today I am going to break some myths like cognitive learning, prescriptive learning, etc.
A picture paints a thousand word. I have taken this from the web.
The whole life cycle is as follows:
Data Collection/Foundation -> Descriptive -> Predictive -> Prescriptive -> Cognitive
Most people have no problem with Data Collection/Foundation and Descriptive. In fact most organization have tonnes of dormant data lying around in digital or non digital forms. Digital data can be stored in databases, excels, logs, etc.
Descriptive is simply having dashboards like kibana, Tableu or even Frontend UI dashboard (mchart, d3js)
Predictive is like machine learning from data collected. For example in programming learn it is a ability to program a model using Python-based Scikit Learn. A simple linear progression program, u can give it 3 points it will project/predict the next few points. So Predictive is a program taking in parameters and output a result according to the ML equation. So people's ML could be using rule-based if-else as discussed in my previous post.
Prescriptive is deploy the predictive model to the right environment. An analogy is doctor giving the correct treatment to patients after all the blood test.
Cognitive is the ultimate. The model is apply with feedback loops back to the model. It will react and learn. This is what Tesla is doing in the humanoid development. Another example is the Chappie movies by Hugh Jackman. Scary right?
This content originally appeared on DEV Community and was authored by DEV Community
DEV Community | Sciencx (2022-03-08T01:42:32+00:00) Data Analytics Progression. Retrieved from https://www.scien.cx/2022/03/08/data-analytics-progression/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.