Abstract: Systems, methods, and non-transitory computer-readable media can determine a first dataset provided by a first party, wherein the first dataset includes a set of vectors that are each associated with a user identifier. A second dataset provided by a second party can be determined, wherein the second dataset includes a set of vectors that are each associated with a user identifier. One or more vectors in the first dataset can be matched to vectors in the second dataset based on a secure multi-party computation without revealing respective graph information of the first party or the second party. Respective mappings between vectors in the first dataset to a set of shared universal identifiers can be provided to the first party. Respective mappings between vectors in the second dataset to the set of shared universal identifiers can be provided to the second party.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a set of mappings between vectors in a first dataset associated with a first party to a set of shared universal identifiers based on a secure multi-party computation. A set of mappings can be determined between vectors in a second dataset associated with a second party to the set of shared universal identifiers based on the secure multi-party computation. Membership information for each vector in the first dataset can be obtained. The membership information indicating whether an individual associated with the vector is assigned to a test group, a control group, or neither. Conversion information for each vector in the second dataset can be obtained. The conversion information indicating whether an individual converted. Conversion counts for the test group and the control group can be determined based at least in part on the membership information and the conversion information.

Abstract: An online system receives impression data from one or more content publishers. The impression data describes impressions provided to users of the online system on behalf of an agent. The online system selects a randomly selected number of impressions in the received impression data. The online system generates an impressions block by encrypting impression data that describes the selected set of impressions using a unique cypher, and adds the impressions block to a blockchain. The online system further generates a cypher block by encrypting the cypher and an identifier of the impressions block to which the cypher applies using a public key provided by the agent to the online system. The online system adds the cypher block to the blockchain. The agent can recover the cypher from the cypher block based on a private key, and the agent can then recover the impression data using the recovered cypher.

Abstract: Systems, methods, and non-transitory computer readable media can determine a representation of an advertisement based on a first machine learning model. The representation can be provided to a second machine learning model. One or more qualitative ratings associated with the advertisement can be determined based on the second machine learning model.

Abstract: The disclosed computer-implemented method may include (1) accessing, by a computing device and from a record stored in an immutable distributed ledger, information describing characteristics of a target audience, (2) ascertaining a target audience member identifier of a target audience member by matching information describing characteristics of the target audience member with the information describing characteristics of the target audience, (3) generating a unique and encrypted secure identifier linking an advertiser identifier to the target audience member identifier, and (4) committing the secure identifier to the immutable distributed ledger as an update transaction to the record. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: The disclosed computer-implemented method for securing data on blockchains may include receiving a smart contract from a third-party for a designated party and a data set, transmitting, to a network of nodes, a request to add the smart contract and the data set to an immutable distributed ledger, receiving a digital signature from the third-party, receiving a digital signature from the designated party, validating the smart contract, the digital signature from the third-party, and the digital signature from the designated party, and in response to validating the smart contract, the digital signature from the third-party and the digital signature from the designated party, adding the smart contract and the data set to the immutable distributed ledger. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Systems, methods, and non-transitory computer readable media can obtain an advertisement via a user interface associated with an application. One or more qualitative ratings associated with the advertisement can be predicted based on a machine learning model. One or more recommendations for improving the qualitative ratings associated with the advertisement can be provided, via the user interface, based at least in part on one or more advertisements that are visually similar to the advertisement.

Abstract: Systems, methods, and non-transitory computer readable media can predict one or more qualitative ratings associated with an advertisement based on a machine learning model. One or more advertisements that are visually similar to the advertisement can be identified. At least one difference between the advertisement and the one or more advertisements can be determined. A recommendation for improving the one or more qualitative ratings associated with the advertisement can be provided based on the at least one difference.

Abstract: Systems, methods, and non-transitory computer readable media can determine qualitative ratings associated with a plurality of advertisements based on a machine learning model. One or more clusters of the plurality of advertisements can be generated based on representations of the plurality of advertisements. One or more advertisements visually similar to an advertisement can be identified based at least in part on a cluster of the one or more clusters and qualitative ratings of advertisements in the cluster.

Abstract: Systems, methods, and non-transitory computer readable media can determine a representation of an advertisement based on a first machine learning model. The representation can be provided to a second machine learning model. One or more qualitative ratings associated with the advertisement can be determined based on the second machine learning model.