Machine learning techniques for predicting classification progression

A system comprising:

• one or more processors;
• and
• one or more memories storing processor-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
• inputting input data to a machine learning framework to generate a predictive output indicating a likelihood of a first severity level classification label of a plurality of severity level classification labels progressing to a second severity level classification label, different from the first severity level classification label, wherein (i) the machine learning framework comprises a classification machine learning model and a progression machine learning model, (ii) the classification machine learning model is configured to assign one of the plurality of severity level classification labels to the input data, and (iii) generating the predictive output comprises:
• inputting the input data to the classification machine learning model to generate a classification comprising the first severity level classification label of the plurality of severity level classification labels;
• inputting the classification to the progression machine learning model to generate the predictive output, wherein generating the classification prior to generating the predictive output reduces a number of operations required to be performed by the progression machine learning model and training the progression machine learning model comprises:
• receiving a model dataset comprising (a) a plurality of model features associated with a plurality of entities, and (b) a plurality of classifications of the plurality of entities, the plurality of classifications comprising an initial severity level classification label associated with a first time period, and a subsequent severity level classification label associated with a second time period, wherein the model dataset is generated by:
• (1) generating a base cohort from feature data associated with the plurality of entities, the base cohort comprising selected ones of the plurality of entities including initial severity level classification labels associated with the first time period that match a selected set of severity level classification labels,
• (2) assigning a selected one of a plurality of outcome labels to the selected ones of the plurality of entities based at least in part on a difference in the initial severity level classification label and the subsequent severity level classification label, and
• (3) generating the plurality of model features based at least in part on feature data associated with the selected ones of the plurality of entities, wherein the progression machine learning model is trained based on the model dataset;
• and
• initiating performance of a prediction-based action based at least in part on the predictive output.

One or more non-transitory computer-readable media storing processor-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

• inputting input data to a machine learning framework to generate a predictive output indicating a likelihood of a first severity level classification label of a plurality of severity level classification labels progressing to a second severity level classification label, different from the first severity level classification label, wherein (i) the machine learning framework comprises a classification machine learning model and a progression machine learning model, (ii) the classification machine learning model is configured to assign one of the plurality of severity level classification labels to the input data, and (iii) generating the predictive output comprises:
• inputting the input data to the classification machine learning model to generate a classification comprising the first severity level classification label of the plurality of severity level classification labels;
• inputting the classification to the progression machine learning model to generate the predictive output, wherein generating the classification prior to generating the predictive output reduces a number of operations required to be performed by the progression machine learning model and training the progression machine learning model comprises:
• receiving a model dataset comprising (a) a plurality of model features associated with a plurality of entities, and (b) a plurality of classifications of the plurality of entities, the plurality of classifications comprising an initial severity level classification label associated with a first time period, and a subsequent severity level classification label associated with a second time period, wherein the model dataset is generated by:
• (1) generating a base cohort from feature data associated with the plurality of entities, the base cohort comprising selected ones of the plurality of entities including initial severity level classification labels associated with the first time period that match a selected set of severity level classification labels,
• (2) assigning a selected one of a plurality of outcome labels to the selected ones of the plurality of entities based at least in part on a difference in the initial severity level classification label and the subsequent severity level classification label, and
• (3) generating the plurality of model features based at least in part on feature data associated with the selected ones of the plurality of entities, wherein the progression machine learning model is trained based on the model dataset;
• and
• initiating performance of a prediction-based action based at least in part on the predictive output.