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.
Dive deeper into content-based recommendations - Python Tutorial
From the course: Building Recommender Systems with Machine Learning and AI
Dive deeper into content-based recommendations
- [Instructor] As an exercise, try modifying contentKandNrecommender.py to use genre, release year, and mise en scene data independently. Which produces the best accuracy, and which produces the most satisfying results subjectively to you? If you have lots of time, you can pass true into the evaluate function in contentrecs.py to compute more evaluation metrics to work with. I'll go over my results in the next slides so hit pause if you don't want any spoilers. So here are the results I got by trying each content attribute out independently. As far as accuracy goes, genre is the winner, and qualitatively, I think genre is the winner as well. You can see that release year just ended up finding the the year this user liked films from the best and recommended whatever it could find from that year. Since we only have release dates down to the year level, what really happened here is every movie from 1994 was tied for first place. So it's kind of arbitrary which ones made it into the top…
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)
Content-based recommendations and the cosine similarity metric8m 58s
-
K-nearest neighbors (KNN) and content recs3m 59s
-
(Locked)
Producing and evaluating content-based movie recommendations5m 23s
-
(Locked)
Bleeding edge alert: Mise-en-scene recommendations4m 31s
-
(Locked)
Dive deeper into content-based recommendations4m 26s
-
(Locked)
-
-
-
-
-
-
-
-
-