You need to deploy a KNN classifier for a real-time recommendation system serving millions of queries per second. The training set has 10 million records with 100 features. Which design choice best addresses the scalability challenge?

Question

Answer

Normalise all features to speed up Euclidean distance calculations

Answer

Switch to Manhattan distance because it's faster than Euclidean

Answer

Use an approximate nearest neighbour (ANN) structure like KD-trees or locality-sensitive hashing to avoid brute-force distance computation

Answer

Set K=1 to minimise computation during prediction

Crowdly

You need to deploy a KNN classifier for a real-time recommendation system servin...

Want instant access to all verified answers on moodle.ucu.ac.ug?