Neural embeddings of transaction data

Patent metadata

Field	Value
Publication number	US-10789530-B2
Application number	US-201916246911-A
Country	US
Kind code	B2
Filing date	Jan 14, 2019
Priority date	Jan 14, 2019
Publication date	Sep 29, 2020
Grant date	Sep 29, 2020

Abstract

Systems, methods, and computer program products to provide neural embeddings of transaction data. A network graph of transaction data based on a plurality of transactions may be received. The network graph of transaction data may define relationships between the transactions, each transaction associated with at least a merchant and an account. A neural network may be trained based on training data comprising a plurality of positive entity pairs and a plurality of negative entity pairs. An embedding function may then encode transaction data for a first new transaction. An embeddings layer of the neural network may determine a vector for the first new transaction based on the encoded transaction data for the first new transaction. A similarity between the vectors for the transactions may be determined. The first new transaction may be determined to be related to the second transaction based on the similarity.

First claim

What is claimed is:

1. A system, comprising:
a processor circuit; and
a memory storing instructions which when executed by the processor circuit cause the processor circuit to:
receive a network graph of transaction data based on a plurality of transactions, the network graph of transaction data defining relationships between the plurality of transactions, each transaction associated with at least a merchant and one account of a plurality of accounts, the network graph based on a transaction log for a plurality of prior transactions, the transaction log specifying, for each prior transaction, a customer account, a merchant, a timestamp, and a transaction amount;
train a neural network based on training data comprising a plurality of positive entity pairs from the network graph of transaction data and a plurality of negative entity pairs not present in the network graph of transaction data, the negative entity pairs comprising artificially generated relationships between each entity in the negative entity pair, the neural network comprising an embeddings layer, a time between the timestamps of the transactions of each positive entity pair being less than a time threshold;
encode, by an embedding function, transaction data for a first new transaction;
determine, by the embeddings layer of the neural network based on the encoded transaction data for the first new transaction, a vector for the first new transaction;
determine a similarity between the vector for the first new transaction and a vector for a second transaction; and
determine, based on the similarity between the vector for the first new transaction and the vector for the second transaction, that the first new transaction is related to the second transaction.

2. The system of claim 1 , the memory storing instructions which when executed by the processor circuit cause the processor circuit to:
train a model based on the embeddings layer of the neural network, the model associated each of the plurality of accounts.

3. The system of claim 2 , the memory storing instructions which when executed by the processor circuit cause the processor circuit to:
determine, by the trained model, that the first new transaction comprises: (i) a fraudulent transaction, (ii) an anomalous transaction, and (iii) a money laundering transaction, wherein a first positive entity pair of the plurality of positive entity pairs comprises a first transaction with a first merchant and a second transaction with a second merchant, the time between the timestamps of the first and second transactions less than the time threshold.

4. The system of claim 1 , the memory storing instructions which when executed by the processor circuit cause the processor circuit to:
receive the transaction log; and
generate the network graph of transaction data based on one or more extract transform load (ETL) operations applied to the transaction log, the one or more ETL operations comprising: (i) standardizing the transaction log according to one or more formats, and (ii) assigning a unique identifier to each unique customer account in the transaction log, wherein a first negative entity pair of the plurality of negative entity pairs comprises a first transaction with a first merchant and a second transaction with a second merchant, the time between the timestamps of the first and second transactions greater than the time threshold.

5. The system of claim 4 , the memory storing instructions which when executed by the processor circuit cause the processor circuit to:
select a predefined number of the positive entity pairs from the network graph of transaction data;
generate, for each positive entity pair, a predefined number of negative entity pairs; and
generate the embeddings layer comprising a plurality of embedding values based on the training of the neural network using the selected positive entity pairs and the generated negative entity pairs.

6. The system of claim 5 , wherein the training of the neural network refines the plurality of embedding values of the embeddings layer such that a respective distance between each positive entity pair is minimized relative to initial embedding values for the embeddings layer and a respective distance between each negative entity pair is minimized relative to the initial embedding values for the embeddings layer, wherein the embeddings layer associates each embedding value with one of the unique identifiers, wherein the vector for the first new transaction is further determined based on a unique identifier in the transaction data for the first new transaction, the memory storing instructions which when executed by the processor circuit cause the processor circuit to:
generate a recommendation based on a location of the second transaction and a merchant of the second transaction, the recommendation specifying a different merchant in a different location relative to the second transaction.

7. The system of claim 1 , wherein the similarity is based on one or more of: (i) a computed difference between the vector for the first new transaction and the vector for the second transaction, (ii) a cosine similarity of the vector for the first new transaction and the vector for the second transaction, and (iii) an inner product of the vector for the first new transaction and the vector for the second transaction, wherein the second transaction is associated with a first account of the plurality of accounts, wherein the similarity reflects that the first new transaction is associated with the first account.

8. A non-transitory computer-readable storage medium storing instructions that when executed by a processor of a computing device, cause the processor to:
receive a network graph of transaction data based on a plurality of transactions, the network graph of transaction data defining relationships between the plurality of transactions, each transaction associated with at least a merchant and one account of a plurality of accounts, the network graph based on a transaction log for a plurality of prior transactions, the transaction log specifying, for each prior transaction, a customer account, a merchant, a timestamp, and a transaction amount;
train a neural network based on training data comprising a plurality of positive entity pairs from the network graph of transaction data and a plurality of negative entity pairs not present in the network graph of transaction data, the negative entity pairs comprising artificially generated relationships between each entity in the negative entity pair, the neural network comprising an embeddings layer, a time between the timestamps of the transactions of each positive entity pair being less than a time threshold;
encode transaction data for a first new transaction;
determine, by the embeddings layer of the neural network based on the encoded transaction data for the first new transaction, a vector for the first new transaction;
determine a similarity between the vector for the first new transaction and a vector for a second transaction; and
determine, based on the similarity between the vector for the first new transaction and the vector for the second transaction, that the first new transaction is related to the second transaction.

9. The non-transitory computer-readable storage medium of claim 8 , further storing instructions that when executed by the processor cause the processor to:
train a model based on the embeddings layer of the neural network, the model associated with each of the plurality of accounts.

10. The non-transitory computer-readable storage medium of claim 9 , further storing instructions that when executed by the processor cause the processor to:
determine, by the trained model, that the first new t

Assignees

• Capital One Services Llc

Inventors

• Bruss Christopher
• Hines Keegan

Classifications

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/098

  Distributed learning, e.g. federated learning · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06Q20/22

  Payment schemes or models · CPC title

• G06Q20/405

  Establishing or using transaction specific rules · CPC title