TECHNOLOGIES FOR COLLECTING ADVERTISING STATISTICS IN A PRIVACY SENSITIVE MANNER

Generally, this disclosure provides technologies for collecting ad statistics in a privacy sensitive manner. In some embodiments the technology includes a system which includes a plurality of client devices, each hosting a context information management (CIMM) module in a secure processing environment. Each CIMM may be operable to select ads for display and calculate statistics for each of the selected ads. The CIMMs may generate a vector representative of those statistics, and may encrypt that vector using additive homomorphic encryption. The encrypted vector may be associated with a statistics collection counter, which may be incremented each time an encrypted vector is calculated. Each CIMM may compare the incremented counter value to a threshold, and may distribute the encrypted vector to another CIMM for further statistical tabulation if the incremented value is less than the threshold. In this way, the technologies described may ensure that a minimum statistical sample size is collected prior to the transmission of ad statistics to an ad network or advertiser, potentially maintaining or protecting user privacy.