despite me trying to understand I can’t still understand why the collaborative filtering is unsupervised since we have a target which in this case the rating, which we use to derive our cost function.
The objective of collaborative filtering is to cluster the consumers based on their past behavior using similarities to make recommendations. The rating is not a target variable in this case, but rather an input variable. So this is a clustering / unsupervised learning technique. Hope this clarifies the confusion!
I am curious. Would you mind sharing where it says Collaborative Filtering is unsupervised learning? Is it in one of this course’s materials?
This is explained throughout the course. Collaborative Filtering is part of recommender systems. You can even find this in the introduction about the course: “It provides a broad introduction to modern machine learning, including supervised learning (multiple linear regression, logistic regression, neural networks, and decision trees), unsupervised learning (clustering, dimensionality reduction, recommender systems), and some of the best practices used in Silicon Valley for artificial intelligence and machine learning innovation (evaluating and tuning models, taking a data-centric approach to improving performance, and more.)”. Hope that helps!
Thanks @Samuel_Chazy. I found that line in the MLS’s page under “About this specialization”, but is that all? I am wondering whether learners get that impression in any course materials too, and if so, from which material. It is better to know it from the learners because each of them may have a different experience.
Hi Raymond thank you for your reply. Maybe I got that impression as soon as I saw unsupervised now, I see the comma. so, I guess it’s a separate type of ML algorithm (Recommender system). if I understood clearly now, right?
thank you for the insight. it does clarify my confusion plus I found that the recommender system is a separate type of ML .If I am correct?
I think you are talking about the course 3’s title: “Unsupervised Learning, Recommenders, Reinforcement”. Yes, I think you are right that they are just the topics of week 1, week 2, and week 3 respectively.