Abstract: In one embodiment, a method includes receiving, by one or more processors of an information processing system, one or more results of a task. The method includes determining, by the one or more processors, an accuracy confidence score for each result of the task. The method includes determining, by the one or more processors, that the results satisfy an accuracy quality threshold set for the task based on the accuracy confidence score for each result of the task. The method includes providing, by the one or more processers, a determination for the task to a customer system in response to the determination that the results satisfy the accuracy quality threshold.

Abstract: In one embodiment, a method for providing recommendations for workflow alteration is disclosed. Task results for completion of a first set of iterations of a workflow are received. Training data may be extracted from the task results. The training data may be used to build a machine learning model for altering at least a portion of the workflow. An automation forecast that assesses the effects of altering the workflow for a second set of the iterations of the task may be generated, and a workflow alteration recommendation may be provided. Based on automation parameters, such as a minimum required level of accuracy, and the automation forecast, a recommendation regarding whether to automate the task may be included in the workflow alteration recommendation. Finally, based on the recommendation, an automated process may be generated to handle at least a portion of the task.

Abstract: Particular embodiments manage distribution of tasks to heterogeneous group of computing devices is disclosed. Computing devices in the group may have disparate attributes that may affect handling of a specific task (e.g., processing speed, network bandwidth, GPU availability, utilization, type of operating system, availability of certain utilities). Tasks may be received from task-requester computer systems, and each task may be associated with an information set. This information set may comprise a description of the task, a token to be provided for completion of the task, and at least one task-processing rule for a task result provided by one or more computing devices involved in handling the task. Task results are received and compared to the task-processing rule until the task-processing rule is satisfied.

Abstract: In one embodiment, a method for providing recommendations for workflow alteration is disclosed. Task results for completion of a first set of iterations of a workflow are received. Training data may be extracted from the task results. The training data may be used to build a machine learning model for altering at least a portion of the workflow. An automation forecast that assesses the effects of altering the workflow for a second set of the iterations of the task may be generated, and a workflow alteration recommendation may be provided. Based on automation parameters, such as a minimum required level of accuracy, and the automation forecast, a recommendation regarding whether to automate the task may be included in the workflow alteration recommendation. Finally, based on the recommendation, an automated process may be generated to handle at least a portion of the task.

Abstract: In one embodiment, a method includes receiving workflow results from completion of a first set of iterations of a workflow. The workflow is associated with a workflow structure including one or more tasks within the workflow. The method includes producing a recommended alteration to the workflow structure including a percentage of the one or more tasks within the workflow to automate. The method includes extracting, from the workflow results, training data for training a machine learning model to automate one or more tasks within the workflow in accordance with the recommended alteration. The method includes causing a machine learning model to be trained to automate the one or more tasks within the workflow structure using the extracted training data. The method includes causing the machine learning model to perform the one or more tasks during a subsequent iteration of the workflow.

Abstract: In one or more embodiments, one or more methods, processes, and/or systems may receive data associated with completion of tasks by agents. Each task corresponds to a category of tasks and is associated with an outcome relative to satisfaction of a specification of performance by a work distributor. An aptitude prediction model is trained to map, for each category of task, a correlation between an outcome corresponding to satisfaction of the specification of performance and one or more aspects of each task and one or more attributes of each agent that has completed the task. An aptitude of each agent towards a category of tasks is determined. A probability that a first agent will complete a first task in a manner specified by the work distributor is predicted using the trained aptitude prediction model. Identification information of the first agent is provided for display in association with the determined probability.

Abstract: In one or more embodiments, one or more methods, processes, and/or systems may receive data associated with effective completion of tasks by agents and determine a positive correlation within the data between first particular feature values of feature vectors associated with the tasks and second particular feature values of feature vectors associated with the agents. A first agent associated with a feature vector that matches, within a first threshold, the second particular feature values may be selected, and a probability that the first agent will effectively complete a first task based on a feature vector associated with the first task matching, within a second threshold, the first particular feature values may be determined.

Abstract: Particular embodiments may receive, by one or more processors of an information processing system, results of a benchmark task performed by a set of one or more workers. Performance of each of the workers in the set on the benchmark task may be determined. One or more best workers of the workers in the set may be selected. Particular embodiments may then determine whether at least one of the best workers is available for a new task. If one of the best workers is available, the new task may be assigned to the available best worker; else the new task may be assigned to a random worker.

Abstract: Particular embodiments may receive, by one or more processors of an information processing system, results from completion of a plurality of iterations of a task performed by a worker. The volume of the results for the worker may be determined to be insufficient. A set of similar workers may be identified, wherein each of the similar workers has a sufficient volume of results for the task. A weighted score of similarity between the worker and the similar workers may be assigned based on the results of the task as completed by the similar workers. Estimates based on the weighted similarity score may be provided for the worker's behavior for the task.

Abstract: Particular embodiments may receive results from completion of a first set of a plurality of iterations of a workflow, the workflow comprising one or more tasks performed by one or more workers. A result for a task may be received from a worker. An outlier model may be determined based on a vector of behavioral features of the worker. A determination may be made regarding whether the result is fraudulent, based on the outlier model and a global normal feature vector.

Abstract: Particular embodiments manage distribution of tasks to heterogeneous group of computing devices is disclosed. Computing devices in the group may have disparate attributes that may affect handling of a specific task (e.g., processing speed, network bandwidth, GPU availability, utilization, type of operating system, availability of certain utilities). Tasks may be received from task-requester computer systems, and each task may be associated with an information set. This information set may comprise a description of the task, a token to be provided for completion of the task, and at least one task-processing rule for a task result provided by one or more computing devices involved in handling the task. Task results are received and compared to the task-processing rule until the task-processing rule is satisfied.

Abstract: In one embodiment, a method includes receiving workflow results from completion of a first set of iterations of a workflow. The workflow is associated with a workflow structure including one or more tasks within the workflow. The method includes producing a recommended alteration to the workflow structure including a percentage of the one or more tasks within the workflow to automate. The method includes extracting, from the workflow results, training data for training a machine learning model to automate one or more tasks within the workflow in accordance with the recommended alteration. The method includes causing a machine learning model to be trained to automate the one or more tasks within the workflow structure using the extracted training data. The method includes causing the machine learning model to perform the one or more tasks during a subsequent iteration of the workflow.

Abstract: In one embodiment, a method includes receiving, by one or more processors of an information processing system, one or more results of a task. The method includes determining, by the one or more processors, an accuracy confidence score for each result of the task. The method includes determining, by the one or more processors, that the results satisfy an accuracy quality threshold set for the task based on the accuracy confidence score for each result of the task. The method includes providing, by the one or more processers, a determination for the task to a customer system in response to the determination that the results satisfy the accuracy quality threshold.

Abstract: Particular embodiments may receive results from completion of a first set of a plurality of iterations of a workflow, the workflow comprising one or more tasks performed by one or more workers. A result for a task may be received from a worker. A fraud candidate model may be determined based on attributes of one or more confirmed instances of fraud assessment and a first vector of behavioral features of the worker. Particular embodiments may then determine whether the result is fraudulent based on the fraud candidate model.

Abstract: Particular embodiments may receive results of a task performed by a plurality of workers, wherein fraud analysis information is provided for each of the results. An accuracy confusion matrix may be determined based on the results. For each of the results, an accuracy confidence score may be determined based on the fraud analysis information. Based on the accuracy confusion matrix and the accuracy confidence score, accuracy of workers performing the task may be predicted.

Abstract: Particular embodiments may receive results of a benchmark task performed by a worker, wherein fraud analysis information is provided for each of the results. An accuracy confusion matrix may be determined based on the results. For each of the results, an accuracy confidence score may be determined based on the fraud analysis information. Based on the accuracy confusion matrix and the accuracy confidence score, an accuracy of the worker when performing the benchmark task may be predicted.

Abstract: Particular embodiments may receive, by one or more processors of an information processing system, results of one or more tasks performed by a worker. The volume of the results for one of the tasks may be determined to be insufficient. The volume of the results of similar tasks completed by the worker may be determined to be sufficient. A weighted similarity score may be assigned based on the results of the similar tasks completed by the worker. Estimates may be provided based on the weighted similarity score of the worker's behavior for a task type of the one of the tasks.

Abstract: In one embodiment, a method for providing recommendations for workflow alteration is disclosed. Task results for completion of a first set of iterations of a workflow are received. Training data may be extracted from the task results. The training data may be used to build a machine learning model for altering at least a portion of the workflow. An automation forecast that assesses the effects of altering the workflow for a second set of the iterations of the task may be generated, and a workflow alteration recommendation may be provided. Based on automation parameters, such as a minimum required level of accuracy, and the automation forecast, a recommendation regarding whether to automate the task may be included in the workflow alteration recommendation. Finally, based on the recommendation, an automated process may be generated to handle at least a portion of the task.

Abstract: In one or more embodiments, one or more methods, processes, and/or systems may receive data associated with effective completion of tasks by agents and determine a positive correlation within the data between first particular feature values of feature vectors associated with the tasks and second particular feature values of feature vectors associated with the agents. A first agent associated with a feature vector that matches, within a first threshold, the second particular feature values may be selected, and a probability that the first agent will effectively complete a first task based on a feature vector associated with the first task matching, within a second threshold, the first particular feature values may be determined.

Abstract: Particular embodiments may receive, by one or more processors of an information processing system, results of a benchmark task performed by a set of one or more workers. Performance of each of the workers in the set on the benchmark task may be determined. One or more best workers of the workers in the set may be selected. Particular embodiments may then determine whether at least one of the best workers is available for a new task. If one of the best workers is available, the new task may be assigned to the available best worker; else the new task may be assigned to a random worker.