Machine learning: a quick review (part 1)
1 – Introduction
1-1- What is machine learning
Machine learning is using computers to automatically detect patterns in data and use these to make predictions or decisions. It is usually useful when: we want to automate something a human can do or want to do things a human can not do.
Deep learning is a subset of machine learning. Big data is more towards software engineering while machine learning is more towards artificial intelligence. Both methods include statistics and math as backbone, but they place more emphasis on prediction instead of description and try to provide flexible models.
1-2- Applications
Recently, the application of machine learning is immersing and the intention to it is bursting as the 2018 Neurips conference tickets sold out under 12 minutes. Some applications includes: spam filtering, credit card fraud detection, product recommendation, motion capture, speech recognition, face/object detection, sports analytics, personal Assistants, medical imaging, self-driving cars, scene completion, image annotation, discovering new cancer subtypes, automated Statistician, fast physics-based animation, beating humans in Go and Starcraft
1-3- Visualization methods
There are many visualization methods, including: time based plot, histogram, boxplot, bar plot, map color, contour plot, stream graph, and etc.
Time based plot:
Histogram plot:
Boxplot [1]:
Bar plot:
map color:
Contour graph [2]:
Stream graph:
