Sachin Kumar Gupta

WORK HISTORY

EDUCATION

ABOUT SACHIN KUMAR GUPTA

Machine Learning and Artificial Intelligence in GISYou’ve probably heard about machine learning (ML). But you’re not exactly sure how to use it in the context of GIS.Simply, machine learning makes sense out of noisy data finding patterns that you’d never think existed. In other words, it’s software that writes software.Instead of applying a pre-built function, ML gains experience through repeated seen conditions and builds a model to apply in new situations.For example, Google might use Bayesian classification to filter spam emails. Alternatively, Facebook might use it for facial recognition and automatically identify faces in images. And ML can even render Nicholas Cage in every movie ever made.Types of Machine Learning (ML)The two broad categories of machine learning are supervised and unsupervised. And they both can apply to GIS applications in various ways. First, what’s the difference between the two?SUPERVISED LEARNING is just fitting data to a function for prediction. For example, if you plot millions of sample points in a graph, you can fit a line to approximate a function.UNSUPERVISED LEARNING recognizes what the data is using patterns from unlabelled data. For example, it takes millions of images and runs them through a training algorithm. After trillions of linear algebra operations, it can take a new picture and segment it into clusters.Most importantly, machine learning is about optimally solving a problem. So it automatically learns on its own and improves from experience.Lately, GIS is applying artificial intelligence in areas such as classification, prediction, and segmentation.But how can we use it in the context of GIS?