Task driven context-aware search

摘要

A processing device of an information processing system is configured to detect bursts of activity associated with a particular user, to map the bursts to tasks performed by the user, to generate a user model based on the mapping of bursts to tasks, and to provide personalized query processing to the user based on the user model. The user model may comprise a Hidden Markov Model. In one embodiment, a query is received from the user in a current time window, and a model parameter is obtained. Based on the model parameter, a most likely task for the user in the current time window is identified, and based on the query, the most likely task, and user classification, search results are ranked for presentation to the user. The processing device may be configured to communicate the user model to a cloud service provider, and to receive from the cloud service provider the ranked search results determined based on the user model.

主权项

1. A method comprising:
detecting bursts of activity associated with a particular user, wherein a given one of the bursts of activity is indicated by an increased level of activity of the particular user above a threshold level of activity of that particular user and wherein the given one of the bursts of activity comprises one or more of opening, modifying and closing a local resource of a computing device associated with the particular user; mapping the bursts to tasks performed by the particular user; generating a user model specific to the particular user based on the mapping of bursts to tasks, wherein the user model comprises a Hidden Markov Model and generating the user model comprises generating a first version of the user model and generating a second version of the user model, the first version of the user model being associated with a first set of services and the second version of the user model being associated with a second set of services, the first set of services being different than the second set of services; and providing personalized query processing to the particular user based on the user model; wherein providing personalized query processing comprises:
receiving one or more queries from the particular user;sending the one or more queries and at least one of the first and second versions of the user model to one or more cloud service providers selected from a plurality of cloud service providers, wherein the one or more cloud service providers are selected based at least in part on one or more user classifications associated with the particular user, the one or more user classifications comprising at least one of a classification based on a role of the particular user in a given entity and a classification based on a location of the particular user; andproviding ranked search results from the one or more cloud service providers to the particular user responsive to the one or more queries and the at least one of the first and second versions of the user model.