SYSTEMS AND METHODS FOR ESTIMATING PAST AND PROSPECTIVE ATTRIBUTE VALUES ASSOCIATED WITH A USER ACCOUNT
Systems and techniques are disclosed for accessing accounts associated with a user and estimating a value of an attribute associated with the user based upon the retrieved account information. Transaction data associated with an account at an external user account system is received. The transactions are categorized into transaction groups. For each transaction group, a confidence value that the group is associated with the attribute is estimated, based at least in part upon a distribution of transaction amounts for the transactions of the group over a time period associated with the group. An attribute value is estimated for each group, based at least in part upon the transaction amounts of the transaction of the group. In addition a value of the attribute for a future time period may be predicted based upon the transaction groups.
1 . A first device, comprising:
one or more memories; and
one or more processors, coupled to the one or more memories, configured to:
instantiate a simulated instance of a first-party application associated with an external account system,
wherein the first-party application and the simulated instance are configured to interface with the external account system via a first application programming interface (API) of the external account system,
wherein the simulated instance is configured to appear to the external account system as the first-party application executing on a second device associated with an account, and
wherein the simulated instance is configured to query the external account system for transaction data associated with the account;
categorize transactions associated with the transaction data into one or more transaction groups based on sources of the transactions;
estimate, for a group of the one or more transaction groups and based on transaction amounts associated with transactions related to the group, an estimated attribute value associated with an attribute for the group;
determine, based on the estimated attribute value, an expected future value associated with the attribute; and
provide, via a second API that is different from the first API, the expected future value.
2 . The first device of claim 1 , wherein the second API comprises a normalized API.
3 . The first device of claim 1 , wherein the one or more processors are further configured to:
receive a request in a normalized API format; and
convert the request to a format compatible with the first API.
4 . The first device of claim 3 , wherein the one or more processors are further configured to:
query the external account system using the converted request.
5 . The first device of claim 2 , wherein an underlying non-public API is abstracted away for the normalized API.
6 . The first device of claim 1 , wherein estimating the estimated attribute value is based on a confidence value associated with the group satisfying a threshold value, and
wherein the confidence value is determined based on a distribution of the transaction amounts over a time period associated with the transaction group.
7 . The first device of claim 1 , wherein categorizing the transactions is based on at least one of:
dates associated with the transactions, or
sources associated with the transactions.
8 . A method, comprising:
instantiating, by a first device, a simulated instance of a first-party application associated with an external account system,
wherein the first-party application and the simulated instance are configured to interface with the external account system via a first application programming interface (API) of the external account system,
wherein the simulated instance is configured to appear to the external account system as the first-party application executing on a second device associated with an account, and
wherein the simulated instance is configured to query the external account system for transaction data associated with the account;
categorizing, by the first device, transactions associated with the transaction data into one or more transaction groups based on sources of the transactions;
estimating, by the first device, for a group of the one or more transaction groups and based on transaction amounts associated with transactions related to the group, an estimated attribute value associated with an attribute for the group;
determining, by the first device and based on the estimated attribute value, an expected future value associated with the attribute; and
providing, by the first device and via a second API that is different from the first API, the expected future value.
9 . The method of claim 8 , wherein the second API comprises a normalized API.
10 . The method of claim 8 , further comprising:
receiving a request in a normalized API format; and
converting the request to a format compatible with the first API.
11 . The method of claim 10 , further comprising:
querying the external account system using the converted request.
12 . The method of claim 9 , wherein an underlying non-public API is abstracted away for the normalized API.
13 . The method of claim 8 , wherein estimating the estimated attribute value is based on a confidence value associated with the group satisfying a threshold value, and
wherein the confidence value is determined based on a distribution of the transaction amounts over a time period associated with the transaction group.
14 . The first device of claim 8 , wherein categorizing the transactions is based on at least one of:
dates associated with the transactions, or
sources associated with the transactions.
15 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a first device, cause the first device to:
instantiate a simulated instance of a first-party application associated with an external account system,
wherein the first-party application and the simulated instance are configured to interface with the external account system via a first application programming interface (API) of the external account system,
wherein the simulated instance is configured to appear to the external account system as the first-party application executing on a second device associated with an account, and
wherein the simulated instance is configured to query the external account system for transaction data associated with the account;
categorize transactions associated with the transaction data into one or more transaction groups based on sources of the transactions;
estimate, for a group of the one or more transaction groups and based on transaction amounts associated with transactions related to the group, an estimated attribute value associated with an attribute for the group;
determine, based on the estimated attribute value, an expected future value associated with the attribute; and
provide, via a second API that is different from the first API, the expected future value.
16 . The non-transitory computer-readable medium of claim 15 , wherein the second API comprises a normalized API.
17 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions further cause the first device to:
receive a request in a normalized API format; and
convert the request to a format compatible with the first API.
18 . The non-transitory computer-readable medium of claim 17 , wherein the one or more instructions further cause the first device to:
query the external account system using the converted request.
19 . The non-transitory computer-readable medium of claim 16 , wherein an underlying non-public API is abstracted away for the normalized API.
20 . The non-transitory computer-readable medium of claim 15 , wherein estimating the estimated attribute value is based on a confidence value associated with the group satisfying a threshold value, and
wherein the confidence value is determined based on a distribution of the transaction amounts over a time period associated with the transaction group.