Webscale Recommender Systems

Webscale Recommender Systems

Problem being addressed

Deep learning systems have made significant progress over the last couple of years. But can they handle data on the order of magnitude of the web?

Solution

In this research, a deep learning recommendation system for graph structured data is developed. This involves the creation of an efficient algorithm called PinSage which uses a combination of random walks and graph convolutions. PinSage is then deployed on Pinterest web scale data that involves a network data set of 7,5 billion examples and 3 billion nodes. It is found that PinSage generates recommendation benchmarks that are much better than any other deep learning or graph based architecture that was considered state-of-the-art at the time of publication.

Advantages of this solution

This is a highly scalable algorithm that outperforms the best benchmarks at the time for making recommendations to users on Pinterest.

Solution originally applied in these industries

internet

Internet Industry

Possible New Application of the Work

/static/media/chemicalAndMaterials-1.0f0d457a.png

Chemical and Materials Industry

Chemical and material manufacture can sometimes require large scale simulations of molecular structure. This deep learning approach and its ability to scale up could be of use for the materials industry.

data:image/png;base64,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

Real Estate Industry

Real estate data is quite extensive, and real estate recommendation engines often serve up information to users. Perhaps this algorithm could be adapted to the real-estate industry as a way of making recommendations to prospective customers. One possibility is, for example, to embed it into AirBNB type businesses.

Source URL: #############check-icon


search-iconBrowse all blitzcards