Is machine learning hard? Not always - The Intercom Blog

Summary

The post argues that machine learning does not always require advanced mathematics and, as described by Intercom’s Fergal Reid, is about systems that improve with more training data. It contrasts a naive keyword-matching approach in Ruby for a ‘You might also like’ feature, which produced irrelevant results, with a basic machine learning method using K-Means clustering. The machine learning approach first represents each blog post as a binary vector indicating the presence of each word from the entire corpus. The K-Means algorithm then iteratively assigns each post to the nearest cluster center and recalculates centers as the average of assigned posts until assignments stabilize. The post demonstrates that roughly 40 lines of Ruby code can produce relevant recommendations and encourages readers to try libraries like Scikit Learn.

Takeaways

1. 1

   Machine learning can be applied to well-defined problems where performance improves with more training data, without requiring advanced mathematics.

2. 2

   A simple K-Means clustering algorithm, implemented in about 40 lines of Ruby, can produce relevant content recommendations by clustering binary word-presence vectors.

3. 3

   The K-Means algorithm iteratively assigns items to the nearest cluster center and recalculates centers as the average of assigned items until assignments stabilize.

FAQ