Abstract: Systems and methods are disclosed for determining optimal resource allocation options for network systems. The method includes receiving real-time data associated with a plurality of systems; generating one or more features associated with a first system of the plurality of systems based on at least a portion of the real-time data; generating, via input of one or more features into a machine learning model, a prediction that the first system is approaching a capacity threshold; determining one or more probabilities associated with respective one or more second systems of the plurality of systems, each of the one or more probabilities indicating a preferability of the respective second system for reallocating one or more resources from the first system; and simulating the reallocation across the plurality of systems based on the one or more probabilities associated with the respective one or more second systems.

Abstract: Systems and methods are disclosed for determining optimal resource allocation options for network systems. The method includes receiving real-time data associated with a plurality of systems; generating one or more features associated with a first system of the plurality of systems based on at least a portion of the real-time data; generating, via input of one or more features into a machine learning model, a prediction that the first system is approaching a capacity threshold; determining one or more probabilities associated with respective one or more second systems of the plurality of systems, each of the one or more probabilities indicating a preferability of the respective second system for reallocating one or more resources from the first system; and simulating the reallocation across the plurality of systems based on the one or more probabilities associated with the respective one or more second systems.

Abstract: Various embodiments of the present disclosure are directed to model feature classification, analysis, and updating for analyzing one or more risk determination model(s). Embodiments include an improved apparatus configured to generate an actionable feature data object subset of a risk model feature set for a risk determination model, for example using an actionable determination model. The apparatus may provide the actionable feature data object subset for rendering to an interface of a client device. The apparatus may additionally or alternatively be configured to utilize user feedback, provided either directly or identified by the apparatus based on user interactions, to update the actionable determination model. The apparatus may additionally or alternatively be configured to maintain and utilize linked claim data object(s) for use in rendering a linked claim scores analysis interface that provides additional insight regarding the risk level of a particular entity.

Abstract: Methods, apparatus, systems, computing devices, computing entities, and/or the like for programmatically prioritizing a data processing queue are provided. An example method may include retrieving a plurality of data objects in the data processing queue, generating a base data table based at least in part on the plurality of data objects, determining a predictive data model based at least in part on the base data table, and adjusting a queue order of the plurality of data objects in the data processing queue based at least in part on a risk score calculated by the predictive data model.

Abstract: Embodiments herein relate to resource allocation optimization. An example method includes receiving a resource allocation optimization request, the resource allocation optimization request comprising a plan identifier and a member data structure population identifier. The example method may further include retrieving a plurality of member data structures based at least in part on the member data structure population identifier. The example method may further include retrieving a plurality of measure data structures based at least in part on the plan identifier. The example method may further include, for each measure data structure of the plurality of measure data structures, generating an optimization score.

Abstract: To detect anomalous activity within a multi-provider environment of transactional data, a particular target entity of a plurality of entities, such as a provider, is identified; multiple relationships associated with the multi-provider environment are determined, wherein each relationship is associated with a relationship score, and wherein the relationship score is determined based on a relational proximity criterion satisfied by the relationship; one or more risk scores are generated; a network risk score is generated for the target entity, multiple levels of related entities are identified, and anomalous activity detection is performed based on the network risk score. Anomalous activity and entity relationships are presented in a graph interface.

Abstract: Various embodiments of the present disclosure are directed to model feature classification, analysis, and updating for analyzing one or more risk determination model(s). Embodiments include an improved apparatus configured to generate an actionable feature data object subset of a risk model feature set for a risk determination model, for example using an actionable determination model. The apparatus may provide the actionable feature data object subset for rendering to an interface of a client device. The apparatus may additionally or alternatively be configured to utilize user feedback, provided either directly or identified by the apparatus based on user interactions, to update the actionable determination model. The apparatus may additionally or alternatively be configured to maintain and utilize linked claim data object(s) for use in rendering a linked claim scores analysis interface that provides additional insight regarding the risk level of a particular entity.

Abstract: Methods, apparatus, systems, computing devices, computing entities, and/or the like for programmatically prioritizing a data processing queue are provided. An example method may include retrieving a plurality of data objects in the data processing queue, generating a base data table based at least in part on the plurality of data objects, determining a predictive data model based at least in part on the base data table, and adjusting a queue order of the plurality of data objects in the data processing queue based at least in part on a risk score calculated by the predictive data model.

Abstract: To detect anomalous activity within a multi-provider environment of transactional data, a particular target entity of a plurality of entities, such as a provider, is identified; multiple relationships associated with the multi-provider environment are determined, wherein each relationship is associated with a relationship score, and wherein the relationship score is determined based on a relational proximity criterion satisfied by the relationship; one or more risk scores are generated; a network risk score is generated for the target entity, multiple levels of related entities are identified, and anomalous activity detection is performed based on the network risk score. Anomalous activity and entity relationships are presented in a graph interface.

Abstract: Machine-learning concepts for detecting and visualizing healthcare fraud, waste, and abuse risk using a data driven decision and investigations support system. The concepts comprise an analytic core that processes a large amount of data, generates an overall risk score, and ranks healthcare providers and/or members. The overall risk score is an integrated score encompassing multiple categories of risk. The multiple categories of risk factors include multiple definitions of what defines risk, allowing for a synergistic effect between the risk analytics, where the overall effect of the combination is greater than the sum of the effects of any one definition of risky behavior by a provider or member. Utilizing this approach a unique risk profile of healthcare providers and members is generated and visualized to the user. Various embodiments further encompass a user interface that comprises linked panels that display targeted information regarding providers and members.

Abstract: Machine-learning concepts for detecting and visualizing healthcare fraud, waste, and abuse risk using a data driven decision and investigations support system. The concepts comprise an analytic core that processes a large amount of data, generates an overall risk score, and ranks healthcare providers and/or members. The overall risk score is an integrated score encompassing multiple categories of risk. The multiple categories of risk factors include multiple definitions of what defines risk, allowing for a synergistic effect between the risk analytics, where the overall effect of the combination is greater than the sum of the effects of any one definition of risky behavior by a provider or member. Utilizing this approach a unique risk profile of healthcare providers and members is generated and visualized to the user. Various embodiments further encompass a user interface that comprises linked panels that display targeted information regarding providers and members.