Machine Learning and Generative AI for Marketing by Yoon Hyup Hwang;Nicholas C. Burtch;

Author:Yoon Hyup Hwang;Nicholas C. Burtch;
Language: eng
Format: epub
Publisher: Packt Publishing
Published: 2024-08-23T14:24:27+00:00

In this chapter, we are going to focus on the collaborative filtering algorithm as it is a frequently used backbone of lots of recommender systems, and more specifically, how to build recommendation systems using user-based collaborative filtering and item-based collaborative filtering algorithms. This technique is frequently used in recommendation systems as it captures the interactions among the users and items very well in a very intuitive way. Lots of organizations where recommendations are critical in their businesses, such as Netflix, Amazon, and Spotify, use collaborative filtering algorithms as part of their recommendation systems.

The key to the collaborative filtering algorithms is to find similar users or items. There can be various metrics that can be used to measure the similarities between users or items, such as Euclidean distance, Manhattan distance, or Jaccard distance. However, cosine similarity is one of the most frequently used similarity metrics in collaborative filtering, as it focuses more on the directional similarity over the magnitude of the distance.

Cosine similarity is simply the cosine of the angle between two vectors. Here, vectors can be user vectors or item vectors in our case of collaborative filtering algorithms. As cosine similarity has strength in high-dimensional space, it is often chosen as the distance metric for collaborative filtering algorithms. The equation for computing the cosine similarity between two vectors looks as follows:

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.