Outcome-driven trajectory tracking

A computer-implemented method for predicting outcome risks for a plurality of entities based on trajectories, the method comprising:

• obtaining a dataset associated with the plurality of entities;
• determining, from the dataset, a set of parameters comprising a sample trajectory for each of the plurality of entities, a risk factor associated with one or more of the plurality of entities, and an outcome projected for each of the plurality of entities according to each entity's sample trajectory;
• calculating a maximum likelihood of each outcome using a likelihood function, the likelihood function being a mixture model of a trajectory model and an outcome model;
• updating the set of parameters according to the maximum likelihood of each outcome;
• building an updated trajectory model, using the dataset, according to the updated set of parameters;
• grouping the plurality of entities into subgroups according to the updated trajectory model, each subgroup containing one or more entities, and each of the one or more entities in a given subgroup having a same or similar trajectory to each other;
• and
• generating at least one alert associated with the entities in at least one of subgroups.

A computer program product for predicting outcome risks for a plurality of entities based on trajectories, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a processor to perform a method comprising:

• obtaining a dataset associated with the plurality of entities;
• determining, from the dataset, a set of parameters comprising a sample trajectory for each of the plurality of entities, a risk factor associated with one or more of the plurality of entities, and an outcome projected for each of the plurality of entities according to each entity's sample trajectory;
• calculating, using the set of parameters, a posterior probability for each outcome;
• applying gradient descent to a likelihood function, using the posterior probability, to determine a maximum likelihood of each outcome, the likelihood function being a mixture model of a trajectory model and an outcome model;
• updating the set of parameters according to the maximum likelihood of each outcome;
• building an updated trajectory model, using the dataset, according to the updated set of parameters;
• grouping the plurality of entities into subgroups according to the updated trajectory model, each subgroup containing one or more entities, and each of the one or more entities in a given subgroup having a same or similar trajectory to each other;
• and
• generating at least one alert associated with the entities in at least one of subgroups.