From the course: Building Recommender Systems with Machine Learning and AI
Unlock this course with a free trial
Join today to access over 22,600 courses taught by industry experts.
Measure the performance of SVD recommendations - Python Tutorial
From the course: Building Recommender Systems with Machine Learning and AI
Measure the performance of SVD recommendations
- [Instructor] So it took a few minutes, but we got some valid results here for our SVD recommender. Our RMSE score is about 0.9, and MAE is about 0.7. That means that on average, our guess of a rating for a given movie for a given user was off by about 0.7 stars. RMSE is higher, meaning that we got penalized for being way off more often than we'd like. Again, remember, higher error metrics are bad. You want these values to be as low as possible if accuracy is your goal. Our hit rate was just about 3%, which actually isn't that bad, considering that only one movie was left out from each user's ratings to test with. But this number by itself is going to be hard to interpret as good or bad until you have other recommenders to compare it against. If we break it down by a rating value, you can see that our hit rates did better at higher rating predictions, which makes sense and is what we want to see. Cumulative hit rate with a 4.0 threshold isn't much lower than the raw hit rate, meaning…
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)
Train/test and cross-validation3m 49s
-
(Locked)
Accuracy metrics (RMSE and MAE)4m 6s
-
Top-N hit rate: Many ways4m 35s
-
(Locked)
Coverage, diversity, and novelty4m 55s
-
(Locked)
Churn, responsiveness, and A/B tests5m 6s
-
(Locked)
Review ways to measure your recommender2m 55s
-
(Locked)
Walkthrough of RecommenderMetrics.py6m 53s
-
(Locked)
Walkthrough of TestMetrics.py5m 8s
-
(Locked)
Measure the performance of SVD recommendations2m 24s
-
(Locked)
-
-
-
-
-
-
-
-
-
-
-