Abstract: According to an example, a processor may receive a request via voice command and obtain a response to the request. The processor may also obtain a confidence level of the obtained response, in which the confidence level corresponds to an accuracy of the identified response to the received request, identify an indication aspect corresponding to the identified confidence level, wherein different indication aspects correspond to different confidence levels, and output the obtained response with the identified indication aspect. The processor may also receive user feedback on the outputted response, in which the received user feedback is used to improve an accuracy of responses provided by the processor to requests received via voice command.

Abstract: A system and a method to match advertisement requests with campaigns using targeting attributes, and campaigns are selected for fulfillment of the advertisement request according to a priority algorithm. The targeting uses end user information that is verifiable, and which the user has granted permission to use, improving the granularity and accuracy of the targeting data. The algorithm includes load balancing and campaign state evaluation on a per campaign, per user basis. The algorithm enables control over the frequency and number of exposures for a campaign, optimizing the advertising both from the perspective of the user and the advertiser.