Sift Science, Inc.

Walmart Apollo, LLC

Rishabh Kothari

Kevin A Lee

Wei Liu

Hui Wang

Pradhan Bagur Umesh

Helen Marushchenko

Rajiv Veeraraghavan

Abbey Chaver

Zhiwei Qin

Arash Asadi-Shahmirzadi

Vvs Varaprasad Nagalla

Systems and methods for detecting digital abuse or digital fraud that involves malicious account testing includes implementing a machine learning threat model that predicts malicious account testing using misappropriate accounts, wherein a subset of a plurality of learnable variables of an algorithmic structure of the machine learning threat model includes one or more learnable variables derived based on feature data indicative of malicious account testing; wherein implementing the machine learning threat model includes: (i) identifying event data from an online event that is suspected to involve digital fraud or digital abuse, (ii) extracting adverse feature data from the event data that map to the one or more learnable variables of the subset, and (iii) providing the adverse feature data as model input to the machine learning threat model; and computing, using the machine learning threat model, a threat prediction indicating a probability that the online event involves malicious account testing.

Systems and methods for configuring and implementing a malicious account testing machine learning model in a machine learning-based digital threat mitigation platform

Systems and methods for configuring and implementing a card testing machine learning model in a machine learning-based digital threat mitigation platform

A system and method for generating an insult rate and reconfiguring an automated decisioning workflow includes configuring a testing group based on sampling from online events having an adverse disposal decision computed by an automated decisioning workflow computer that is configured with machine learning-based threat score thresholds that, if satisfied, causes a computation of a disallow decision or a block decision for a given online event; evaluating a performance and collecting performance data of distinct members of the testing group over a testing period; computing an insult rate for the testing group based on the performance data; computing an insult rate equilibrium for the automated decisioning workflow computer based on the performance data; evaluating the insult rate against the insult rate equilibrium; and reconfiguring adverse decisioning thresholds based on the evaluation of the insult rate of the testing group against the insult rate equilibrium for the automated decisioning workflow computer.

Systems and methods for insult rate testing and reconfiguring an automated decisioning workflow computer for improving a machine learning-based digital fraud and digital abuse mitigation platform

A method of reallocating inventory in a fulfillment network is disclosed herein. The fulfillment network can include a plurality of distribution centers. An allocation plan can be created in a one of a variety of different manners, where the allocation plan involves allocating an item to one or more distribution centers in the fulfillment network. Thereafter, the allocation plan can be analyzed for feasibility. If the allocation plan is not feasible, each distribution center in the allocation plan can be analyzed to determine if using the distribution center is feasible. If the distribution center cannot be used, another distribution in the same cluster of distribution centers is examined for feasibility. This process is repeated for each distribution center in the allocation plan. Once an alternative allocation plan has been developed in this manner, items can be allocated. Existing inventory can be taken into account in the allocation plan.

Optimal reallocation of inventory under capacity violations

Growing community of inventors

San Francisco, CA, United States of America

Rishabh Kothari