From the course: Building Recommender Systems with Machine Learning and AI
Unlock this course with a free trial
Join today to access over 23,200 courses taught by industry experts.
The cold start problem (and solutions) - Python Tutorial
From the course: Building Recommender Systems with Machine Learning and AI
The cold start problem (and solutions)
- [Instructor] One of the better known issues with recommender systems is what is known as the cold-start problem. If a brand new user arrives at your site, what do you recommend to them when you know nothing about them yet? And new users aren't the only problem. What about new items in your catalog? How do they get recommended when there is no data on them yet to pair them with other items? When faced with a new user, your options are limited. The one saving grace is that a new user isn't new for long. Assuming they stay in your site at all, you'll soon have some information to work with that indicates their interests. As soon as this new user looks at a new item, you'll have at least some implicit information about this user's interests, even if it's just that this user looked at this product. If you're lucky, they landed on your site on an actual product page, and you can just recommend other items similar to the item they are looking at. As it turns out, the thing a person is…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
-
-
-
-
-
(Locked)
The cold start problem (and solutions)6m 12s
-
(Locked)
Implement random exploration54s
-
(Locked)
Exercise solution: Random exploration2m 18s
-
(Locked)
Stoplists4m 48s
-
(Locked)
Implement a stoplist32s
-
(Locked)
Exercise solution: Implement a stoplist2m 22s
-
(Locked)
Filter bubbles, trust, and outliers5m 39s
-
(Locked)
Identify and eliminate outlier users44s
-
(Locked)
Exercise solution: Outlier removal4m
-
Fraud, the perils of clickstream, and international concerns4m 33s
-
(Locked)
Temporal effects and value-aware recommendations3m 30s
-
(Locked)
-
-
-