CPC G06F 16/3322 (2019.01) | 21 Claims |
1. A computer-implemented method performed by data processing apparatus comprising one or more computers in data communication, the method comprising:
analyzing, for one or more native applications on a user device, an aggregate feedback of the one or more native applications on a user device;
identifying one or more native applications that have a reduced level of usage compared to other applications based on the aggregate feedback based on a click-through rate of one or more sources on each of the one or more native applications;
demoting data associated with one or more native applications that have a reduced level of usage compared to other applications within a ranking of native application history data or removing the data associated with one or more native applications that have poor utility within the native application history data;
receiving, from a first user device, a search query input for a search application that provides a search environment in which a user can input a search query and receive search results from a search service, the search query input comprising one or more characters;
accessing the native application history data for the first user device, the native application history data describing content presented by a plurality of native applications at the first user device, each of the plurality of native applications being different from the search application;
generating, based on the search query input and the native application history data for the first user device, one or more query suggestions; and
providing, to the first user device, the one or more query suggestion for presentation in the search environment of the search application.
|