Re-ranking search results using blended learning models

A method to re-rank a plurality of search results, the method comprising:

• receiving, by a client device, a search query from a user;
• generating, on the client device, the plurality of search results over a plurality of search domains, wherein the plurality of search results is ranked according to a first ranking;
• generating a re-ranking model, wherein the re-ranking model includes a plurality of intra-domain models that are generated based on at least based on-device interactions of a plurality of users interacting with a plurality of other devices and each of the plurality of search domains corresponds to one of the plurality of intra-domain models;
• re-ranking the plurality of search results, by the client device, using the re-ranking model;
• and
• presenting the plurality of search results using the second ranking.

A device to re-rank a plurality of search results, the comprising:

• a processor;
• a memory coupled to the processor though a bus;
• and
• a process executed from the memory by the processor causes the processor to receive a search query from a user, generate the plurality of search results over a plurality of search domains, wherein the plurality of search results is ranked according to a first ranking, generate a re-ranking model, wherein the re-ranking model includes a plurality of intra-domain models that are generated based on at least based on-device interactions of a plurality of users interacting with a plurality of other devices and each of the plurality of search domains corresponds to one of the plurality of intra-domain models, re-rank the plurality of search results, by the client device, using the re-ranking model, and present the plurality of search results using the second ranking.