Task multi-assignment method, apparatus, and computer program using tier data structure of crowdsourcing-based project

A task multi-assignment method using a tier data structure of a crowdsourcing-based project,

• the method being performed by a computer, being a task assignment method using a tier data structure of a crowdsourcing-based project (hereinafter, a project) of a multi-assignment method of duplicately assigning each task to a plurality of n different workers (wherein n is a natural number greater than 1), and comprising:
• configuring an assignment tier storage including a queue into which task data corresponding to a task that is not assigned to the worker is enqueued, and one or more jth stacks into which task data corresponding to tasks that are each assigned j times (wherein j is a natural number less than n) is pushed;
• and
• assigning a task of the project to a worker (hereinafter, a target worker) who requests task assignment and requesting the target worker to perform the task, by using the assignment tier storage;
• configuring a rejection tier storage including one or more pieces of rejection data corresponding to a worker who has a work result (hereinafter, a rejected work result) that is rejected according to an inspection result;
• and,
• when the target worker has the rejected work result, requesting the target worker to perform rework by using the rejection tier storage, wherein
• the task data includes a key-value pair including an identifier of the task and n pieces of multi-assignment data,
• the multi-assignment data includes a key-value pair including an identifier of a worker to which the task is assigned, and a multi-assignment index,
• the assigning the task of the project and requesting to perform the task includes
• extracting task data at the top of a current stack in a descending order with respect to the one or more jth stacks,
• identifying whether or not the task data at the top of the current stack includes an identifier of the target worker,
• when the task data at the top of the current stack does not include the identifier of the target worker, popping the task data from the current stack, assigning a task corresponding to the task data popped from the current stack to the target worker, requesting the target worker to perform the task, inputting the identifier of the target worker into the multi-assignment data of the task data popped from the current stack, and pushing the task data popped from the current stack into a previous stack,
• the rejection data includes a key-value pair including an identifier of a worker who has the rejected work result, and a rejection task queue, an identifier of a task corresponding to the rejected work result is enqueued into the rejection task queue,
• the requesting to perform the rework includes,
• when a rejected work result occurs, enqueuing an identifier of a task corresponding to the rejected work result into a rejection task queue corresponding to a worker who inputs the rejected work result,
• identifying whether or not the one or more pieces of rejection data of the rejection tier storage include the identifier of the target worker, and
• when the one or more pieces of rejection data of the rejection tier storage include the identifier of the target worker, dequeuing an identifier of a task corresponding to the rejected work result from a rejection task queue corresponding to the target worker, and requesting the target worker to perform rework for a task corresponding to the dequeued identifier of the task.