Summary

This chapter presented an opportunity to learn about a social network which has so far passed the test of time. Flickr has evolved with the Internet and has been, for a long time now, a popular site for professional and amateur photographers alike. Since its very inception, Flickr has been part of numerous research studies and their APIs have provided analysts/researchers access to their data. With about 10 billion photos, Flickr is a goldmine of sorts. In this chapter, we covered ways of interacting with Flickr using its APIs, and prepared some useful utilities to extract and preprocess the deeply nested response objects. We learnt a few basics related to photography, in particular EXIF, and used this domain knowledge to cluster our set of extracted photos using the K-means clustering algorithm. We also utilized photo attributes such as iso, focal_length, views, and so on, to build a random forest based classifier to help us identify whether a photo will land on the Explore page...