Detecting a transaction volume anomaly
A server device obtains historical transaction data regarding transactions involving a network service, obtains historical calendar data regarding static date information for a historical time period that corresponds with the historical transaction data, and processes the historical transaction data and historical calendar data to train a machine learning model using a gradient boosting machine learning technique to predict a normal transaction volume for a period of time and confidence bands associated with the normal transaction volume. The server device generates the normal transaction volume for the period of time and confidence bands using the machine learning model, obtains real-time data concerning a transaction volume during the period of time, detects a transaction volume anomaly based on comparing the real-time data and normal transaction volume and confidence bands, and sends an alert, based on the transaction volume anomaly, to cause a remote device to display the alert and perform an action.
1. A method associated with a network service related to one or more point of sale (POS) terminals or one or more automated teller machines (ATMs), the method comprising:
obtaining, by a device, historical transaction data regarding a plurality of transactions,
wherein the plurality of transactions are associated with the network service;
training, by the device and based on processing the historical transaction data and historical calendar data, a machine learning model;
predicting, by the device and based on the machine learning model, a first transaction volume for a period of time;
generating, by the device and based on the machine learning model, one or more confidence bands associated with the first transaction volume;
detecting, by the device, a transaction volume anomaly based on real-time data, concerning a transaction volume for the period of time, based on the first transaction volume, and based on the one or more confidence bands;
determining, by the device, a first point in time associated with detecting the transaction volume anomaly;
determining, by the device, a second point in time associated with determining that the detected transaction volume anomaly is erroneous;
discarding, by the device, real-time data associated with a time period between the first point in time and the second point in time; and
updating, by the device, the machine learning model based on real-time data that was not discarded.
2. The method of claim 1 , wherein the historical transaction data comprises transactions associated with the network service.
3. The method of claim 1 , wherein the one or more confidence bands indicate, for the first transaction volume, at least one of:
a standard deviation, or
a margin of error.
4. The method of claim 1 , further comprising:
determining a first confidence band based on the one or more confidence bands,
the first confidence band indicating a first transaction volume threshold for determining transaction volume anomalies; and
determining a second confidence band based on the one or more confidence bands,
the second confidence band indicating a second transaction volume threshold for determining transaction volume anomalies.
5. The method of claim 1 , further comprising:
providing, to another device, data that causes display of:
the real-time data,
the first transaction volume, and
the one or more confidence bands.
6. The method of claim 1 , wherein the plurality of transactions include different types of transactions associated with the network service and other network services.
7. The method of claim 6 , wherein the first transaction volume is predicted for a particular transaction type, of the different types of transactions.
8. A device associated with a network service related to one or more point of sale (POS) terminals or one or more automated teller machines (ATMs), the device comprising:
one or more memories; and
one or more processors communicatively coupled to the one or more memories, configured to:
obtain historical transaction data regarding a plurality of transactions,
wherein the plurality of transactions are associated with the network service;
train, based on processing the historical transaction data and historical calendar data, a machine learning model;
predict, based on the machine learning model, a first transaction volume for a period of time;
generate, based on the machine learning model, one or more confidence bands associated with the first transaction volume;
detect a transaction volume anomaly based on real-time data, concerning a transaction volume for the period of time, based on the first transaction volume, and based on the one or more confidence bands;
determine a first point in time associated with detecting the transaction volume anomaly;
determine a second point in time associated with determining that the detected transaction volume anomaly is erroneous;
discard real-time data associated with a time period between the first point in time and the second point in time; and
update the machine learning model based on real-time data that was not discarded.
9. The device of claim 8 , wherein the historical transaction data comprises transactions associated with the network service.
10. The device of claim 8 , wherein the one or more confidence bands indicate, for the first transaction volume, at least one of:
a standard deviation, or
a margin of error.
11. The device of claim 8 , wherein the one or more processors are further configured to:
determine a first confidence band based on the one or more confidence bands,
the first confidence band indicating a first transaction volume threshold for determining transaction volume anomalies; and
determine a second confidence band based on the one or more confidence bands,
the second confidence band indicating a second transaction volume threshold for determining transaction volume anomalies.
12. The device of claim 8 , wherein the one or more processors are further configured to:
provide, to another device, data that causes display of:
the real-time data,
the first transaction volume, and
the one or more confidence bands.
13. The device of claim 8 , wherein the plurality of transactions include different types of transactions associated with the network service and other network services.
14. The device of claim 13 , wherein the first transaction volume is predicted for a particular transaction type, of the different types of transactions.
15. A non-transitory computer-readable medium storing one or more instructions, the one or more instructions being associated with a network service related to one or more point of sale (POS) terminals or one or more automated teller machines (ATMs), the one or more instructions, when executed by one or more processors, cause the one or more processors to:
obtain historical transaction data regarding a plurality of transactions,
wherein the plurality of transactions are associated with the network service;
train, based on processing the historical transaction data and historical calendar data, a machine learning model;
predict, based on the machine learning model, a first transaction volume for a period of time;
generate, based on the machine learning model, one or more confidence bands associated with the first transaction volume;
detect a transaction volume anomaly based on real-time data, concerning a transaction volume for the period of time, based on the normal transaction volume, and based on and the one or more confidence bands;
determine a first point in time associated with detecting the transaction volume anomaly;
determine a second point in time associated with determining that the detected transaction volume anomaly is erroneous;
discard real-time data associated with a time period between the first point in time and the second point in time; and
update the machine learning model based on real-time data that was not discarded.
16. The non-transitory computer-readable medium of claim 15 , wherein the historical transaction data comprises transactions associated with the network service.
17. The non-transitory computer-readable medium of claim 15 , wherein the one or more confidence bands indicate, for the first transaction volume, at least one of:
a standard deviation, or
a margin of error.
18. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
determine a first confidence band based on the one or more confidence bands,
the first confidence band indicating a first transaction volume threshold for determining transaction volume anomalies; and
determine a second confidence band based on the one or more confidence bands,
the second confidence band indicating a second transaction volume threshold for determining transaction volume anomalies.
19. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
provide, to another device, data that causes display of:
the real-time data,
the first transaction volume, and
the one or more confidence bands.
20. The non-transitory computer-readable medium of claim 15 , wherein the plurality of transactions include different types of transactions associated with the network service and other network services.