Abstract: A method includes receiving, by a server computer, data of a communication device; training, by the server computer, a neural network model based on the data of the communication device and communication device metadata from one or more additional communication devices, to generate a machine learning model configured to determine, based on a metadata associated with an application, a security value related to an indication of a security threat; and transmitting the machine learning model to the communication device. The communication device can use the machine learning model to determine the security value, by inputting the metadata associated with the application, as a vectorized data into the machine learning model. The communication device can determine whether to run or install the application based upon the security value.

Abstract: Embodiments are directed to a method for determining an interview script in a claims submission. The method may comprising receiving data relating to a claim being submitted, which may include claims submission data input by a user, information relating to the user, and one or more features. Data associated with the one or more features may be determined from an artificial intelligence model. A first score based on the data associated with the one or more features and data associated with the information relating to the user may be determined and used to determine an interview script. In one embodiment, questions in the interview script may continue to be provided to the interviewer computer if a continually updated score remains above a predetermined threshold. In another embodiment, the user may be routed to a live interview with a human representative if a continually updated score drops below a predetermined threshold.

Abstract: Embodiments are directed to a method of performing autonomous learning for updating input features used for an artificial intelligence model, the method comprising receiving updated data of an information space that includes a graph of nodes having a defined topology, the updated data including historical data of requests to the artificial intelligence model and output results associated with the requests, wherein different categories of input data corresponds to different input nodes of the graph. The method may further comprise updating edge connections between the nodes of the graph by performing path optimizations that each use a set of agents to explore the information space over cycles to reduce a cost function, each connection including a strength value, wherein during each path optimization, path information is shared between the rest of agents at each cycle for determining a next position value for each of the set of agents in the graph.

Abstract: One embodiment of the invention is directed to a method comprising, receiving an authentication request message for a transaction. The method further comprises determining that detailed device information is required to authenticate the transaction and generating a message including an identifier and a request for the detailed device information. The method further comprises retrieving the detailed device information from a remote server computer using the identifier and modifying the authentication request message to include the detailed device information. The method further comprises sending the modified authentication request message to an access control server computer. The method further comprises receiving an authentication response message from the access control server computer including a verification value for the transaction, where the verification value is generated based on a result of a risk analysis performed using the detailed device information.

Abstract: Client devices can send access request messages to resource management computers to request access to a resource. A data security hub can provide centralized routing between different client devices, resource management computers, and authentication data processing servers. The data security hub can reduce the risk of sensitive authentication information from leaking (e.g., due to a breach) by limiting the amount or types of authentication information distributed to the data processing servers. The data security hub can limited the authentication information being distributed based on its sensitivity, the trust level of the client device, and the security level of the requested resource. The data security hub can also evaluate the client devices and data processing servers to identify security breaches and can cancel or reroute access requests accordingly. Thus, the data security hub can maintain resource security while better preserving the privacy of the client device's authentication information.

Abstract: Client devices can send access request messages to resource management computers to request access to a resource. A data security hub can provide centralized routing between different client devices, resource management computers, and authentication data processing servers. The data security hub can reduce the risk of sensitive authentication information from leaking (e.g., due to a breach) by limiting the amount or types of authentication information distributed to the data processing servers. The data security hub can limited the authentication information being distributed based on its sensitivity, the trust level of the client device, and the security level of the requested resource. The data security hub can also evaluate the client devices and data processing servers to identify security breaches and can cancel or reroute access requests accordingly Thus, the data security hub can maintain resource security while better preserving the privacy of the client device's authentication information.

Abstract: Systems and methods are provided for transcompiling non-distributed source code for a non-distributed software program into a distributed software package for implementation on a distributed computing system. A transcompiler can identify loops within non-distributed source code written in a data-driven language. The transcompiler can generate MapReduce jobs using mapper keys based on grouping indicators associated with each of the loops. The MapReduce jobs can be linked together based on input-output connections of the loops in the non-distributed source code. Then, the transcompiler can generate a distributed software package including the generated MapReduce jobs to implement the same functionality as the non-distributed source code on the distributed computing system, thereby improving the speed of execution over very large datasets. The distributed software package can be optimized using machine learning searching algorithms.

Abstract: A server computing device receives authentication request messages formatted according to a first message format from a party via a computing device of a user involved in a transaction, translates the messages into modified authentication request messages formatted according to a second message format, and transmits the modified authentication request messages to an authorization computer. The server computing device may augment the modified authentication request messages by including data gathered from the received authentication request messages describing the user, user's computing device, and/or network path between the server computing device and user's computing device, as well as risk scores generated based upon the received authentication request message.

Abstract: A server computing device receives authentication request messages formatted according to a first message format from a party via a computing device of a user involved in a transaction, translates the messages into modified authentication request messages formatted according to a second message format, and transmits the modified authentication request messages to an authorization computer. The server computing device may augment the modified authentication request messages by including data gathered from the received authentication request messages describing the user, user's computing device, and/or network path between the server computing device and user's computing device, as well as risk scores generated based upon the received authentication request message.

Abstract: Methods for controlling genetic learning for predictive models using predefined strategies may include, for each of a plurality of agents, selecting a type of predictive model. A strategy may be selected from predefined strategies. Candidate genomes may be generated and may include a plurality of genes. Each gene may be associated with a feature of the agent predictive model. A fit of each candidate genome to the agent strategy may be determined. A candidate genome may be selected based on the fit. For each of a plurality of epochs, a plurality of training iterations may be performed for each agent. A fitness of each agent predictive model may be determined. A subset of agents with a highest fitness may be determined. For each agent of the subset, at least one new agent may be generated. The genomes of the new agents may be merged with some genomes of the subset.

Abstract: Embodiments are directed to a method for accelerating machine learning using a plurality of graphics processing units (GPUs), involving receiving data for a graph to generate a plurality of random samples, and distributing the random samples across a plurality of GPUs. The method may comprise determining a plurality of communities from the random samples using unsupervised learning performed by each GPU. A plurality of sample groups may be generated from the communities and may be distributed across the GPUs, wherein each GPU merges communities in each sample group by converging to an optimal degree of similarity. In addition, the method may also comprise generating from the merged communities a plurality of subgraphs, dividing each sub-graph into a plurality of overlapping clusters, distributing the plurality of overlapping clusters across the plurality of GPUs, and scoring each cluster in the plurality of overlapping clusters to train an AI model.

Abstract: Techniques for authorizing a transaction or interaction of a user that is modified by authentication information for the same user are described herein. In embodiments, an authorization request message for a transaction and a session identifier may be received from a transport computer or a resource provider computer. A portion of pre-analyzed data about the user and the one or more interactions may be obtained, from a database, based on the session identifier. A risk analysis for the transaction using the portion of the pre-analyzed data may be performed to generate a value. The authorization request message may be modified to include the portion of the pre-analyzed data and the value. The modified authorization request message may be transmitted to an authorizing computer.

Abstract: One embodiment of the invention is directed to a method comprising, receiving an authentication request message for a transaction. The method further comprises determining that detailed device information is required to authenticate the transaction and generating a message including an identifier and a request for the detailed device information. The method further comprises retrieving the detailed device information from a remote server computer using the identifier and modifying the authentication request message to include the detailed device information. The method further comprises sending the modified authentication request message to an access control server computer. The method further comprises receiving an authentication response message from the access control server computer including a verification value for the transaction, where the verification value is generated based on a result of a risk analysis performed using the detailed device information.

Abstract: Embodiments are directed to a method for accelerating machine learning using a plurality of graphics processing units (GPUs), involving receiving data for a graph to generate a plurality of random samples, and distributing the random samples across a plurality of GPUs. The method may comprise determining a plurality of communities from the random samples using unsupervised learning performed by each GPU. A plurality of sample groups may be generated from the communities and may be distributed across the GPUs, wherein each GPU merges communities in each sample group by converging to an optimal degree of similarity. In addition, the method may also comprise generating from the merged communities a plurality of subgraphs, dividing each sub-graph into a plurality of overlapping clusters, distributing the plurality of overlapping clusters across the plurality of GPUs, and scoring each cluster in the plurality of overlapping clusters to train an AI model.

Abstract: A method includes a) receiving, by a computer, network data comprising a plurality of transactions conducted by a plurality of actual users and a plurality of actual resource providers. The computer can b) generate a plurality of simulated users. Each simulated user based upon a set of the plurality of actual users. The computer can then c) generate a plurality of simulated resource providers. Each simulated resource provider based upon at least one actual resource provider. The computer can d) execute a simulation using the plurality of simulated users and the plurality of simulated resource providers, and in response to step d) can e) determine a plurality of simulated transactions conducted by the plurality of simulated users and the plurality of simulated resource providers.

Abstract: Systems and methods are provided for transcompiling non-distributed source code for a non-distributed software program into a distributed software package for implementation on a distributed computing system. A transcompiler can identify loops within non-distributed source code written in a data-driven language. The transcompiler can generate MapReduce jobs using mapper keys based on grouping indicators associated with each of the loops. The MapReduce jobs can be linked together based on input-output connections of the loops in the non-distributed source code. Then, the transcompiler can generate a distributed software package including the generated MapReduce jobs to implement the same functionality as the non-distributed source code on the distributed computing system, thereby improving the speed of execution over very large datasets. The distributed software package can be optimized using machine learning searching algorithms.

Abstract: A computer-implemented method of using past transaction declines to predict future fraudulent behavior. A transaction has been declined for an account is determined, and a risk score is determined. The risk score is compared to a risk threshold. The transaction is compared to one or more transactions in a transaction profile for a past fraudster in response to the risk score being determined to be over the risk threshold. A best fit of the transaction profiles of the past fraudster is determined and a measure of success for the best fit of the transaction profiles of the past fraudster is also determined. If the measure of success is over a threshold, the method updates profiles of past fraudsters based on the transaction to include the transaction that has been declined. The method predicts future fraudulent transactions and attempts to stop future fraudulent transactions based on the predicted future fraudulent transactions.

Abstract: One embodiment of the invention is directed to a method comprising, receiving an authentication request message for a transaction. The method further comprises determining that detailed device information is required to authenticate the transaction and generating a message including an identifier and a request for the detailed device information. The method further comprises retrieving the detailed device information from a remote server computer using the identifier and modifying the authentication request message to include the detailed device information. The method further comprises sending the modified authentication request message to an access control server computer. The method further comprises receiving an authentication response message from the access control server computer including a verification value for the transaction, where the verification value is generated based on a result of a risk analysis performed using the detailed device information.

Abstract: A method includes a processing computer querying a data store for plurality of network data since a last epoch of data. The processing computer can generate matrices based on the plurality of network data and then perform tensor factorization on the matrices, to obtain latent values in the network data. The processing computer can then determine that the latent values satisfy a predetermined criterion. The processing computer can input the latent values into a model associated with the signal, wherein the model generates output data. The processing computer can then generate an alert comprising the output data and then transmit the alert to a remote server computer.

Abstract: A method includes a processing computer receiving a processing request message comprising user data from a remote server computer. The processing computer can then determine latent values associated with the processing request message based on the user data and a multiplex graph. The processing computer can then normalize the latent values based on a community group in the multiplex graph. The community group can include at least a part of the user data. The processing computer can transmit a processing response message comprising at least one normalized latent value to the remote server computer.