Understand the main applications of recommenders. Discover similarities.
- [Instructor] Recommendation systems have several different uses. The most common use for a recommendation system is ranking products by how much a user would like them. If a user is browsing or searching for products, we want to show them the products they would like most first in the list. Users expect to find what they want quickly and will move on if they have a hard time finding relevant products. Recommendation systems can also be used to find out how similar different products are to each other. If products are very similar to each other, they might appeal to the same users. When the user clicks on one product, we can use this to give the user links to other products that are very similar.
Or if a user buys a product, we can email the user later with advertisements for similar products. Product similarity is especially useful in cases where we don't know much about a particular user yet. We can recommend similar products, even if the user hasn't entered any of their own product reviews yet. We can also use recommendation systems to figure out if two different users are similar to each other. If two users have similar preferences for products, we can assume they have similar interests. For example, a social network can use this information to suggest the two users should become friends.
Recommendation systems are a key part of almost every modern consumer website. The systems help drive customer interaction and sales by helping customers discover products and services they might not ever find themselves. The course uses the free, open source tools Python 3.5, pandas, and numpy. By the end of the course, you'll be equipped to use machine learning yourself to solve recommendation problems. What you learn can then be directly applied to your own projects.
- Building a machine learning system
- Training a machine learning system
- Refining the accuracy of the machine learning system
- Evaluating the recommendations received