Computer-based management methods and systems

A processor-based method comprising:

• executing, by a processor, an event history machine learning model utilizing a discrete-time interval analysis configured to predict recovery data for each claimant record of a plurality of claimant records stored in a database utilizing survival analysis, by receiving disability income claim data representative of disability income insurance claims and claimants and outputting one or more claimant records including recovery data representative of a likelihood that respective claimant records will realize a claim termination event based at least on analyzing discrete-time intervals,
• wherein the claim termination event is one of a plurality of preselected recovery events representative of a claimant achieving return-to-work capacity,
• wherein the event history machine learning model is trained using updated disability income claim data representative of the disability income insurance claims and claimants, wherein the event history machine learning model is trained by transforming the updated disability income claim data into a longitudinal format that includes multiple person-time period records corresponding to each claimant record;
• and
• running, by the processor, the event history machine learning model on demand to update and display, by a display device in operative communication with the processor, a graphical user interface showing the recovery data representative of the likelihood that the respective claimants will realize a claim termination event.

A system comprising:

• non-transitory machine-readable memory that stores claimant records for a plurality of claimants for disability income insurance, and disability income claim data representative of disability income insurance claims and claimants;
• an event history machine learning model utilizing a discrete-time interval analysis configured to predict recovery data for each claimant record of the plurality of claimant records and to output one or more claimant records including recovery data representative of a likelihood that respective claimant records will realize a claim termination event based at least on analyzing discrete-time intervals, wherein the claim termination event is one of a plurality of preselected recovery events representative of a claimant achieving return-to-work capacity, wherein the event history machine learning model is trained using updated disability income claim data;
• and
• a processor in operative communication with a display device, wherein the processor in communication with the non-transitory, machine-readable memory executes a set of instructions instructing the processor to:
• receive the claimant records for the plurality of claimants and the disability income claim data representative of disability income insurance claims and claimants from the non-transitory machine-readable memory;
• apply the event history machine learning model to the claimant records for the plurality of claimants and the disability income claim data representative of disability income insurance claims and claimants to predict the recovery data for each claimant record of the plurality of claimant records and output the one or more claimant records including the recovery data representative of a likelihood that respective claimant records will realize a claim termination event;
• and
• run the event history machine learning model on demand to update and display, by the display device in operative communication with the processor, a graphical user interface showing the recovery data representative of the likelihood that the respective claimants will realize a claim termination event.