Contact data for identity verification

1 . A computer-implemented method for identity verification during a payment transaction comprising:

causing presentation, by one or more servers of a service provider and on a user interface of a user device of a user during pendency of a payment transaction, at least one prompt for a name and a phone number of the user;

receiving, by the one or more servers from the user device during the pendency of the payment transaction, an indication of input comprising the name and the phone number of the user,

collecting, by the one or more servers and from each mobile device of a plurality of mobile devices, a respective set of contact records, wherein the respective set of contact records is stored on each respective mobile device, includes metadata, and corresponds to user accounts of a plurality of users associated with the service provider and corresponding contact record signals;

analyzing, by the one or more servers, the metadata collected from each mobile device of the plurality of mobile devices, wherein the metadata includes, for an individual contact record of the set of contact records, data indicating a length of time that the individual contact record has been stored by the respective mobile device, and wherein the individual contact record of the set of contact records is associated with a phone number of a contact and wherein the corresponding contact record signals include at least a portion of the metadata;

responsive to receiving the name and the phone number of the user, identifying, during the pendency of the payment transaction by the one or more servers via a query of the set of contact records, particular contact records, of the set of contact records, that contain the phone number of the user;

receiving, during the pendency of the payment transaction by the one or more servers and as a result of the query, a plurality of name options associated with the phone number of the user, wherein individuals of the plurality of name options are stored in at least one of the particular contact records;

responsive to receiving the plurality of name options associated with the phone number of the user, determining, during the pendency of the payment transaction by the one or more servers and utilizing a first machine-trained model trained based at least in part on first training data representing a corpus of at least one of similar name options, common names, or rules of grammar, scores for individual ones of the plurality of name options, wherein the first machine-trained model outputs the scores based at least in part on a similarity of the individual ones of the plurality of name options to other name options;

responsive to determining the scores for the individual ones of the plurality of name options, determining, during the pendency of the payment transaction by the one or more servers and utilizing a second machine-trained model trained at least in part on second training data representing a history of true name options and the other name options used to derive the true name options, a predicted name option of the plurality of name options to associated with the phone number of the user, wherein the second machine-trained model outputs the predicted name option based at least in part on interpreting the scores for the individual ones of the plurality of name options and particular contact record signals associated with the particular contact records;

verifying, by the one or more servers during the pendency of the payment transaction, an existing user account of the user based at least in part on the analyzing of the metadata, and a comparison of the predicted name option and the name provided via the user device; and

causing presentation, by the one or more servers and via the user interface during the pendency of the payment transaction, an indication that an identity of the user has been verified.

2 . The computer-implemented method of claim 1 , wherein the similarity comprises one or more of a spelling similarity or a phonetic similarity.

3 . The computer-implemented method of claim 1 , wherein the scores based at least in part on the similarity of the individual ones of the name options to the other name options are further based on one or more parameters to account for noise associated with the plurality of name options.

4 . The computer-implemented method of claim 1 , wherein the corresponding contact record signals comprise at least one contact record signal comprising one or more of a quantity of contact records associated with the predicted name option, trustworthiness of the individual contact record, or trustworthiness of a particular user account associated with the individual contact record, wherein verifying the existing user account is further based on the at least one contact record signal.

5 . A computer-implemented method for identity verification during a payment transaction comprising:

causing presentation, by one or more servers of a service provider and on a user interface of a user device of a user during pendency of a payment transaction, at least one prompt for a name and an identifier of the user;

receiving, by the one or more servers from the user device during the pendency of the payment transaction, an indication of input comprising the name and the identifier of the user;

collecting, by the one or more servers and from each mobile device of a plurality of mobile devices, a set of contact records, wherein the set of contact records is stored on each respective mobile device, includes metadata, and corresponds to user accounts of a plurality of users associated with the service provider and corresponding contact record signals;

analyzing, by the one or more servers, the metadata collected from each mobile device of the plurality of mobile devices, wherein the metadata includes, for an individual contact record of the set of contact records, data indicating a length of time that the individual contact record has been stored by a mobile device, and wherein the individual contact record of the set of contact records is associated with an identifier of a contact and the contact record signals include at least a portion of the metadata;

responsive to receiving the name and the identifier of the user, identifying, during the pendency of the payment transaction by the one or more servers via a query of the set of contact records, particular contact records, of the set of contact records, that contain the identifier of the user;

receiving, during the pendency of the payment transaction by the one or more servers and as a result of the query, a plurality of name options associated with the identifier of the user, wherein individuals of the plurality of name options are stored in at least one of the particular contact records;

responsive to receiving the plurality of name options associated with the identifier of the user, determining, during the pendency of the payment transaction by the one or more servers and utilizing a first machine-trained model, a score for an individual name option of the plurality of name options based at least in part on a similarity of the individual name option to other name options in the plurality of name options;

verifying, by the one or more servers during the pendency of the payment transaction and utilizing a second machine-trained model, an existing user account of the user based at least in part on the analyzing of the metadata, and a comparison of a determined name option and the name provided via the user device, wherein the second machine-trained model outputs the determined name option based at least in part on interpreting the score for the individual name option of the plurality of name options and particular contact record signals associated with the particular contact records; and

causing presentation, by the one or more servers and via the user interface during the pendency of the payment transaction, an indication that an identity of the user has been verified.

6 . The computer-implemented method of claim 5 , further comprising generating, by the one or more servers, a similarity matrix comprising a measure of similarity of the individual name option to the other name options of the plurality of name options.

7 . The computer-implemented method of claim 5 , wherein the identifier of the user comprises a phone number of the user or a device identifier.

8 . The computer-implemented method of claim 5 , wherein the similarity comprises a spelling similarity of at least components of the individual name option to the other name options of the plurality of name options.

9 . The computer-implemented method of claim 5 , wherein the similarity comprises a phonetic similarity between the individual name option and the other name options in the plurality of name options.

10 . The computer-implemented method of claim 5 , wherein the similarity of individual candidate name options comprises a weighted combination of at least a first similarity score and a second similarity score, wherein the first similarity score and the second similarity score are associated with different measures of similarity.

11 . The computer-implemented method of claim 5 , further comprising calculating, by the one or more servers, a probability of the determined name option being a true name associated with the identifier of the user.

12 . The computer-implemented method of claim 5 , further comprising determining, by the one or more servers, that the similarity of the determined name option to the other name options of the plurality of name options satisfies a threshold similarity.

13 . The computer-implemented method of claim 5 , further comprising determining, by the one or more servers, that the similarity of the determined name option to the other name options of the plurality of name options satisfies a threshold similarity for a threshold number of the other name options.

14 . The computer-implemented method of claim 5 , wherein:

each of the set of contact records are retrieved from a database accessible by the service provider; and

each of the set of contact records are provided to the database from a contact record stored by user devices of respective users of the service provider after the respective user has granted permission to the service provider to collect the contact record.

15 . A payment service system comprising:

one or more processors; and

one or more computer-readable non-transitory storage media coupled to one or more of the processors and comprising instructions operable when executed by one or more of the processors to cause the payment service system to perform operations comprising:

causing presentation, on a user interface of a user device of a user during pendency of a payment transaction, at least one prompt for a name and an identifier of the user;

receiving, from the user device during the pendency of the payment transaction, an indication of input comprising the identifier of the user;

collecting, from each mobile device of a plurality of mobile devices, a set of contact records, wherein the set of contact records is stored on each respective mobile device, includes metadata, and corresponds to user accounts of a plurality of users associated with the payment service system and corresponding contact record signals;

analyzing the metadata collected from each mobile device of the plurality of mobile devices, wherein the metadata includes, for an individual contact record of the set of contact records, data indicating a length of time that the individual contact record has been stored by the mobile device, and wherein the individual contact record of the set of contact records is associated with an identifier of a contact and the contact record signals include at least a portion of the metadata;

responsive to receiving the name and the identifier of the user, identifying during the pendency of the payment transaction via a query of the set of contact records, particular contact records, of the set of contact records, that contain the identifier of the user;

receiving, during the pendency of the payment transaction and as a result of the query, a plurality of name options associated with the identifier of the user, wherein individuals of the plurality of name options are stored in at least one of the particular contact records;

responsive to receiving the plurality of name options associated with the identifier of the user, determining, during the pendency of the payment transaction and utilizing a first machine-trained model, scores for individual ones of the plurality of name options based at least in part on a similarity of the individual ones of the plurality of name options to other name options-;

during the pendency of the payment transaction, verifying an existing user account of the user associated with a service provider based at least in part on the analyzing of the metadata, and a comparison of a predicted name option and the name provided via the user device, wherein a second machine-trained model outputs the predicted name option based at least in part on interpreting the scores for the individual ones of the plurality of name options and particular contact record signals associated with the particular contact records; and

causing presentation, via the user interface during the pendency of the payment transaction, an indication that an identity of the user has been verified.

16 . The payment service system of claim 15 , wherein the instructions are further operable when executed by at least one of the one or more processors to cause the payment service system to perform further operations comprising generating a similarity matrix comprising a measure of the similarity of the individual ones of the plurality of name options to the other name options.

17 . The payment service system of claim 15 , wherein the set of contact records are stored at a datastore associated with the service provider.

18 . The payment service system of claim 15 , wherein the input comprises first input, wherein the at least one prompt comprises a first prompt, wherein the instructions are further operable when executed by at least one of the one or more processors to cause the payment service system to perform further operations comprising:

during the pendency of the payment transaction:

causing presentation, on the user interface of the user device of the user, of a second prompt, wherein the second prompt comprises a request for credentials of the user for a social media account of the user;

responsive to presentation of the second prompt, receiving an indication of a second input to the user interface;

receiving, from a computing device associated with a social media platform via an application programming interface (API) or software development kit (SDK) integration of the payment service system with the social media platform, data associated with the social media account of the user; and

determining that the second input comprises valid credentials based at least in part on the data associated with the social media account of the user,

wherein verifying the existing user account of the user is further based at least in part on the valid credentials.

19 . The payment service system of claim 15 , wherein the input comprises first input, wherein the at least one prompt comprises a first prompt, wherein the instructions are further operable when executed by at least one of the one or more processors to cause the payment service system to perform further operations comprising:

during the pendency of the payment transaction:

causing presentation, on the user interface of the user device of the user, of a second prompt, wherein the second prompt comprises a request for credentials of the user for an email account of the user;

responsive to presentation of the second prompt, receiving an indication of a second input to the user interface;

receiving, from a computing device associated with an email platform via an application programming interface (API) or software development kit (SDK) integration of the payment service system with the email platform, data associated with the email account of the user; and

determining that the second input comprises valid credentials based at least in part on the data associated with the email account of the user,

wherein verifying the existing user account of the user is further based at least in part on the valid credentials.

20 . The payment service system of claim 19 , wherein the instructions are further operable when executed by at least one of the one or more processors to cause the payment service system to perform further operations comprising:

receiving an indication of a third input to the user interface, wherein the third input comprises permission for the payment service system to access the data associated with the email account, wherein receiving the data associated with the email account is based at least in part on the permission.