Here's a next step to understanding Recommendation Engine in AI.
Collaborative filtering
Want to know how it works?
How can it help you skyrocket your company sales?
checkout our website to know more! http://bit.ly/2nYcPB1
2. What is it?
Collaborative filtering is used to tailor
recommendations based on the behavior of
persons with similar interests. Sometimes it
can be based on an item bought by the user.
Since this method does not require a person
himself to always contribute to a data store,
voids can be filled by the actions of other
persons, or the same person on
other items.
Similar product
Purchased by both users
Purchased by himRecommended to him!
3. Types of Collaborative filtering
Neighborhood-based approach
Item-based approach
Classification approach
Neural Collaborative Filtering
In this approach, users are selected based
on their similarity to the active user.
Instead of using ratings given by the users
to calculate the neighborhood, the ratings
are used to find similarity between items.
In the classification approach, items are represented as vectors and
they are classified and suggested to the user based on the ratings
provided by the active user to each class of items.
Collaborative filtering is a method of making predictions about the
interests of a user by collecting preferences from many users.
Collaborative filtering is perhaps the most common machine
learning technique used by recommender systems.
4. Upside
No domain knowledge necessary
Serendipity
Great starting point
Source: developers.google.com
Collaborative filtering doesn’t need domain knowledge
because the embeddings are automatically learned.
The model can help users discover new interests. In isolation, the ML
system may not know the user is interested in a given item, but the
model might still recommend it because similar users are interested in
that item.
To some extent, the system needs only the feedback matrix to train a
matrix factorization model. In particular, the system doesn’t need
contextual features. In practice, this can be used as one of the multiple
candidate generators.
5. Downside
Cannot handle fresh items
If an item is not seen during training, the system can’t create an
embedding for it and can’t query the model with this item.
This issue is often called the cold-start problem.
Hard to include side features for query/item
Side features are any features beyond the query or item ID. For movie
recommendations, the side features might include country or age.
6. | 866-628-2302 | ctoboost.com
User Experience is Key!
Recommendation engine improves with more information. Recommen-
dation engines that display smart, intuitive, visualization techniques for
their results are much likelier to ensure repeat visits
See into the future with CTO
Boost’s recommendation engine;
predicting what your customer
wants before they even know
they want it.