I am interested in a framework for learning the similarity of different input representations based on some common context. I have looked into word2vec, SVD and siamese networks, all of which are similar to what I want.

For example, suppose we have some customers we are sending different advertisements to, and I would like to create a system to map offers to customers. I am thinking in the lines of creating a customer representation, and a representation of the offers, and feeding them in parallel to a neural network that has a label of whether they acted on the advertisement or not. The idea is that I should be able to locate the best offer for any customer given these representations.

I have looked into siamese networks and word2vec, both are close to what I want. The problem differs slightly in that for the siamese networks, it tends to be identical parallel networks, which I don't want because my inputs are not equivalent. And for word2vec the vectors tend to be in the same domain, while I want to apply this in a more general setting.

If anyone has any resources on a similar problem statement, I would be very interested in it.


The idea is that I should be able to locate the best offer for any customer given these representations.

I think you need a Recommender System. As you want to map the offers to customers based one their representation you can check Content-Based Recommender System.

The method is by taking pattern from a customer history and try to find similarities with the new offers. You can use many techniques for a recommender system, from a simple TF-IDF or Deep Learning for more complex problems.

  • $\begingroup$ Thanks for the article on Deep Lerning methods, it looks very relevant. $\endgroup$ – user10283726 Mar 6 at 7:59
  • $\begingroup$ @user10283726 Is the idea to use recommender system suitable for your problem? $\endgroup$ – malioboro Mar 6 at 9:06
  • $\begingroup$ Yes, and I have been reading about it but haven't found exactly what I am looking for yet. But I will read the article you provided, hopefully it will be what I am looking for. $\endgroup$ – user10283726 Mar 6 at 9:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.