Paypal, Inc.

Piyush Neupane

Aditya Jain

Yogesh Kandlur

Anirudh Singh Shekhawat

Prasiddha Malla

Visak Chockalingom

There are provided systems and methods for identifying transaction processing retry attempts based on machine learning models for transaction success. A service provider, such as an electronic transaction processor for digital transactions, may detect a failure of data processing for a transaction when processed with a separate data processing system, such as a card processing system for payment cards. In order to minimize cost and wasted resources for retrying transactions that are likely to further fail, a machine learning model may be implements that generates a predictive score for whether a failed transaction is likely to be successful if retried with the data processing system. The predictive score may be used to predict a probability of success, which may then be used with a cost function to determine a cost to retry the failed transaction and a cost to stop a retry of the failed transaction.

Identifying transaction processing retry attempts based on machine learning models for transaction success

A machine learning model, such as a Gradient Boosting Machine model, is trained using historical data associated with different types of operations. The machine learning model extracts features from the historical data and learns which features result in a predefined outcome, such as approval by a predefined entity. After the machine learning model is trained and validated as being accurate, it is used to predict the likelihood of a prospective operation achieving the predefined outcome when submitted right away versus a future date. If the likelihood of achieving the predefined outcome is low if submitted right away, the submission of the prospective operation is temporarily suspended. The machine learning model calculates a future time during which to submit the prospective operation for approval, where the future time has a greater likelihood of achieving the predefined outcome. This reduces waste of electronic resources including computer processing power and network communication bandwidth.

Training and using a two stage machine learning model to predict an outcome of a prospective transaction

Systems and techniques for facilitating card authentication technique selection via machine learning are provided. In various embodiments, a processor can access an attribute vector associated with a financial payment card. In various instances, the processor can determine, via execution of a machine learning model, whether to authenticate the financial payment card with a zero-dollar authentication technique or instead with a tokenization authentication technique, based on the attribute vector. In various cases, the processor can execute the zero-dollar authentication technique with respect to the financial payment card, when the processor determines that the financial payment card should be authenticated with the zero-dollar authentication technique. In various aspects, the processor can execute the tokenization authentication technique with respect to the financial payment card, when the processor determines that the financial payment card should be authenticated with the tokenization authentication technique.

Growing community of inventors

Hayward, CA, United States of America

Piyush Neupane