Directed trajectories through communication decision tree using iterative artificial intelligence

A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause one or more data processors to perform actions including:

• accessing a progressive-filtering decision tree for assigning a job, wherein the progressive-filtering decision tree includes a sequenced set of decision nodes and is configured to perform iterative decisions that correspond to progressive filtering of workers, wherein at least one decision node of the sequenced set of decision nodes of the progressive-filtering decision tree:
• (i) corresponds to an iterative decision of the iterative decisions;
• and (ii) is configured to identify, from the workers, an incomplete subset of the workers for performing the job, the incomplete subset of the workers being inputted for a subsequent decision node of the sequenced set of decision nodes, and wherein, for each decision node of the sequenced set of decision nodes, each of one or more trajectory extensions to the decision node triggers an intermediate or final decision pertaining to configuring a worker request to perform the job;
• initiating a trajectory of the progressive-filtering decision tree, the trajectory being associated with a set of workers;
• detecting that the trajectory has reached a first decision node of the sequenced set of decision nodes in the progressive-filtering decision tree;
• accessing a plurality of worker-associated unstructured data sets, each of the plurality of worker-associated unstructured data sets corresponding to a worker of the set of workers and including unstructured data;
• and
• executing one or more first iterations of a first machine-learning model configured with a first set of node-specific learned parameters and using the plurality of worker-associated unstructured data sets, wherein the execution of the one or more first iterations of the first machine-learning model generates a first result that includes a plurality of worker feature data sets, and wherein each of the plurality of worker feature data sets:
• (i) corresponds to a worker of the set of workers;
• and (ii) includes a set of values generated by applying the first machine-learning model to the worker-associated unstructured data set of the worker, the set of values predictive of an extent to which the worker qualifies for performing the job;
• identifying, based on the first result, a subset of the set of workers for a query for the job;
• triggering a communication action that requests each of the subset of workers to participate in the query;
• accessing a set of responses from at least one worker of the subset of workers;
• identifying, based on the set of responses, one or more workers selected from the subset of workers for continued evaluation for the job;
• detecting that the trajectory has reached a second decision node of the sequenced set of decision nodes in the progressive-filtering decision tree;
• executing one or more second iterations of a second machine-learning model configured with a second set of node-specific learned parameters and using a subset of the plurality of worker feature data sets, wherein each of the subset of the plurality of worker feature data sets corresponds to a worker in the one or more selected workers, and wherein execution of the one or more second iterations of the second machine-learning model generates a second result that indicates, for each worker of the one or more selected workers, an estimated absolute or relative degree of correspondence between the job and the selected worker;
• and
• outputting the second result.

A system comprising:

• one or more data processors;
• and
• a non-transitory computer readable storage medium containing instructions which when executed on the one or more data processors, cause the one or more data processors to perform actions including:
• accessing a progressive-filtering decision tree for assigning a job, wherein the progressive-filtering decision tree includes a sequenced set of decision nodes and is configured to perform iterative decisions that correspond to progressive filtering of workers, wherein at least one decision node of the sequenced set of decision nodes of the progressive-filtering decision tree:
• (i) corresponds to an iterative decision of the iterative decisions;
• and (ii) is configured to identify, from the workers, an incomplete subset of the workers for performing the job, the incomplete subset of the workers being inputted for a subsequent decision node of the sequenced set of decision nodes, and wherein, for each decision node of the sequenced set of decision nodes, each of one or more trajectory extensions to the decision node triggers an intermediate or final decision pertaining to configuring a worker request to perform the job;
• initiating a trajectory of the progressive-filtering decision tree, the trajectory being associated with a set of workers;
• detecting that the trajectory has reached a first decision node of the sequenced set of decision nodes in the progressive-filtering decision tree;
• accessing a plurality of worker-associated unstructured data sets, each of the plurality of worker-associated unstructured data sets corresponding to a worker of the set of workers and including unstructured data;
• and
• executing one or more first iterations of a first machine-learning model configured with a first set of node-specific learned parameters and using the plurality of worker-associated unstructured data sets, wherein the execution of the one or more first iterations of the first machine-learning model generates a first result that includes a plurality of worker feature data sets, and wherein each of the plurality of worker feature data sets:
• (i) corresponds to a worker of the set of workers;
• and (ii) includes a set of values generated by applying the first machine-learning model to the worker-associated unstructured data set of the worker, the set of values predictive of an extent to which the worker qualifies for performing the job;
• identifying, based on the first result, a subset of the set of workers for a query for the job;
• triggering a communication action that requests each of the subset of workers to participate in the query;
• accessing a set of responses from at least one worker of the subset of workers;
• identifying, based on the set of responses, one or more workers selected from the subset of workers for continued evaluation for the job;
• detecting that the trajectory has reached a second decision node of the sequenced set of decision nodes in the progressive-filtering decision tree;
• executing one or more second iterations of a second machine-learning model configured with a second set of node-specific learned parameters and using a subset of the plurality of worker feature data sets, wherein each of the subset of the plurality of worker feature data sets corresponds to a worker in the one or more selected workers, and wherein execution of the one or more second iterations of the second machine-learning model generates a second result that indicates, for each worker of the one or more selected workers, an estimated absolute or relative degree of correspondence between the job and the selected worker;
• and
• outputting the second result.