Systems and methods for a unified matching engine
A computer-implemented method for implementing a uniform matching engine. In aspects, the method includes receiving consumer data from a plurality of data sources, the consumer data include a plurality of attributes that include personally identified information, non-personally identifiable information, or a combination thereof. In some aspects, the method associates a first attribute of the plurality of attributes with a second attribute of the plurality of attributes and creates combinations of attributes based upon the associations. In some aspects, the method includes matching first and second combinations of the multiple combinations and associating one or more attributes included in the first combination with one or more attributes included in the second combination based upon the matching. In addition, the method includes outputting the association of the one or more attributes included in the first combination with the one or more attributes included in the second combination.
1 . A computer-implemented method for implementing a uniform matching engine, comprising:
receiving consumer data from a plurality of data sources, including digital device interactions from a plurality of digital devices, the consumer data comprising a plurality of attributes that include personally identified information, non-personally identifiable information, or a combination thereof;
associating a first attribute of the plurality of attributes with a second attribute of the plurality of attributes;
assigning a confidence score to the association of the first attribute with the second attribute, using a same digital device interaction, wherein the same digital device interaction comprises respective digital device interactions by a same digital device from the plurality of digital devices and same user of the same digital device, based on one or more of: (i) how frequently the first attribute and the second attribute have been included in the same digital device interaction, where the more frequent the inclusion the greater the confidence score, (ii) how recently the first attribute and the second attribute have been included in the same digital device interaction, where the more recent the inclusion the greater the confidence score, (iii) whether the first attribute and the second attribute are of a particular type, and (iv) the association of the first attribute and the second attribute each with other attributes;
creating a plurality of combinations of attributes of the plurality of attributes based upon respective associations;
assigning a combination score to each combination of the plurality of combinations of attributes based on one or more of: (i) how frequently the combination of attributes have been included in the same digital device interaction, where the more frequent the inclusion the greater the combination score, (ii) how recently the combination of attributes have been included in the same digital device interaction, where the more recent the inclusion the greater the combination score, (iii) whether the combination of attributes are of a particular type, and (iv) a set of rules selected based at least one of the attributes;
adjusting, for respective combinations of attributes of the plurality of combinations of attributes, an initial combination score for a first combination of attributes by (i) removing an attribute from the first combination of attributes to form a subcombination, (ii) calculating a combination score for the subcombination, and (iii) adjusting the combination score for the subcombination based on the initial combination score;
matching a set of combinations from the plurality of combinations of attributes based on the combination score and the confidence score of the plurality of combinations of attributes to associate different attributes of the set of combinations, wherein the set of combinations includes a first combination of the plurality of combinations of attributes to a second combination of the plurality of combinations of attributes;
associating one or more attributes included in the first combination with one or more attributes included in the second combination based upon the matching;
using an identity graph component to generate an identity graph of association of the one or more attributes included in the first combination with the one or more attributes included in the second combination; and
providing an interface to permit a user to select sets of output data using the identity graph.
2 . The method of claim 1 , further comprising assigning a matching score to the set of combinations based on the combination score of the first combination and the second combination.
3 . The method of claim 1 , wherein the plurality of data sources comprise at least one data source selected from the following data sources: public records, public censuses, property data, realty data, telecom data, purchase propensity data, content publisher data, e-commerce data, supply-side publisher data, mobile application data, customers data sets, clickstream data, and/or digital metadata.
4 . The method of claim 1 , wherein the identity graph is based at least in part upon (i) the association of the first attribute with the second attribute, (ii) the plurality of combinations, (iii) the matching of the first combination to the second combination, (iv) the association of the one or more attributes included in the first combination with the one or more attributes included in the second combination, or a combination thereof.
5 . The method of claim 1 , wherein the first attribute and the second attribute are selected for use in the identity graph are based on at least one of (i) a hierarchical structure, (ii) a relationship between the first attribute and the second attribute, (iii) a physical geographical address, and (iv) a common digital device.
6 . The method of claim 1 , wherein the plurality of data sources comprise data from digital device interactions, including data regarding digital identity data, device identifiers, attribute data, browser and app-level data, location data, and/or online behavioral data.
7 . The method of claim 6 , wherein digital device interaction data is obtained from one or more of the following: content & e-commerce publishers; supply-side platform aggregators; mobile application SDKs or aggregators; customer provided data; clickstream data aggregators; device metadata.
8 . The method of claim 1 , wherein the plurality of data sources include data from one or more digital device interactions, and wherein the step of associating a first attribute of the plurality of attributes with a second attribute of the plurality of attributes involves the step of determining if the first attribute and the second attribute were obtained from data from the same digital device interactions and, if so, associating the first attribute and the second attribute.
9 . The method of claim 1 , creating a plurality of combinations of attributes of the plurality of attributes based upon the respective associations.
10 . The method of claim 1 , further comprising a step of adjusting a first combination score for the first combination of attributes using a machine learning algorithm, the machine learning algorithm using feature engineering and model generation to determine combination scores associated with different combinations of attributes.
11 . The method of claim 1 , wherein the step of the matching the first combination of the plurality of combinations of attributes to the second combination of the plurality of combinations of attributes is based at least in part on the combination score.
12 . A computer-implemented method for implementing a uniform matching engine, comprising:
receiving an association of a first attribute of a plurality of attributes with a second attribute of the plurality of attributes, from digital device interactions from a plurality of digital devices, wherein the plurality of attributes comprises personally identified information, non-personally identifiable information, or a combination thereof;
assigning a confidence score to the association of the first attribute with the second attribute, using a same digital device interaction, wherein the same digital device interaction comprises respective digital device interactions by a same digital device from the plurality of digital devices and same user of the same digital device, based on one or more of: (i) how frequently the first attribute and the second attribute have been included in the same digital device interaction, where the more frequent the inclusion the greater the confidence score, (ii) how recently the first attribute and the second attribute have been included in the same digital device interaction, where the more recent the inclusion the greater the confidence score, (iii) whether the first attribute and the second attribute are of a particular type, and (iv) the association of the first attribute and the second attribute each with other attributes;
receiving an association of one or more attributes included in a first combination of a plurality of combinations with one or more attributes included in a second combination of a plurality of combinations of attributes, wherein the plurality of combinations combines at least a portion of the plurality of attributes into different combinations, each combination of the plurality of combinations of attributes is assigned a combination score based on one or more of: (i) how frequently the combination of attributes have been included in the same digital device interaction, where the more frequent the inclusion the greater the combination score, (ii) how recently the combination of attributes have been included in the same digital device interaction, where the more recent the inclusion the greater the combination score, (iii) whether the combination of attributes are of a particular type, and (iv) a set of rules selected based at least one of the attributes;
adjusting, for respective combinations of attributes of the plurality of combinations of attributes, an initial combination score for a first combination of attributes by (i) removing an attribute from the first combination of attributes to form a subcombination, (ii) calculating a combination score for the subcombination, and (iii) adjusting the combination score for the subcombination based on the initial combination score;
matching the first combination and the second combination among the plurality of combinations of attributes based on the combination score and the confidence score of the plurality of combinations;
generating an identity graph based at least in part upon (i) the association of the first attribute with the second attribute, (ii) the plurality of combinations, or a combination thereof; and
providing an interface to permit a user to select sets of output data using the identity graph.
13 . The method of claim 1 , wherein the plurality of attributes excludes a predetermined type of a cookie.
14 . The method of claim 13 , wherein the predetermined type of the cookie comprises a third-party cookie.
15 . The method of claim 1 , wherein the identity graph indicates a first user and a second user sharing a physical address, and the first user at the physical address is as indicated in the associated one or more attributes of the first combination.
16 . The method of claim 1 , wherein the identity graph indicates a first user and a second user shaing a digital device to enable cross-walking from the first user to the second user, and the first user using the digital device is indicated as at least a part of the associated one or more attributes of the first combination.