KR102315396B1 - Project task management method that provides an online work-based optimized task operation and apparatus of the same - Google Patents

Abstract

The present invention relates to a project-based task management method for a crowdsourcing-based project executed by a device. The method includes: a step of grouping tasks executing the project into task groups based on the number of the tasks; a step of evenly distributing source data executing the project to the task groups; a step of calculating an evaluation score processing the source data of each of the task groups from the execution of the project to the evaluation using an evaluation function determined based on the type of the project; and a step of distributing the remaining source data not processed at the evaluation time point to the task group corresponding to the highest evaluation score among the task groups. The evaluation function is determined based on at least one evaluation element and a weight corresponding to the at least one evaluation element. The at least one evaluation element includes at least one of the labor cost required for the source data processing, the time consumed per result data in the data redistribution period, the quality of the result data, the normalized progress rate of outputting the result data, and the level of difficulty of processing the source data. The weight is determined based on the type of the project.

Description

A project task management method that provides an online work-based optimized task operation method and a device for performing the same

The present invention relates to a method for managing a task included in a project, and more particularly, to an online work-based project task management method.

An increasing number of companies are collecting and processing large amounts of data based on crowdsourcing that engages the general public in some processes of business activities.

Crowdsourcing is a compound word of crowd and outsourcing, and refers to involving the public in some process of business activities.

This crowdsourcing is a structure suitable for the gig economy, which has the meaning of fixed-term work traded in the recent digital marketplace, and is optimized for on demand, in which companies provide services immediately to the needs of consumers. may be provided.

Through this, if the public is involved in the company's activities, new ideas and practical opinions can be heard from the company's point of view, and the public can be paid for participating in feedback, and RPA technology based on cloud and AI, cognitive and Analysis) combined with Digital Workforce can be used to provide high-quality project results to consumers.

In addition, using crowdsourcing has the advantage of lowering development costs and obtaining potential customers if the public participates directly to produce the desired results rather than entrusting it to an external professional to make or service a product.

Recently, crowdsourcing has been used for data labeling services such as data labeling automation, distributed processing design, online non-face-to-face management, and human resource management (HR Tech). It is being actively used in industries that require learning data.

However, when performing a project using crowdsourcing, the difficulty of the project is too high, and the public cannot participate in the project.

On the other hand, in the case of a specific project, even if the worker wants to participate, it is difficult to perform the task due to the high difficulty of the content included in the source data used for each project. Therefore, there is a need for a method for solving these problems.

It may also be desirable to have a way to divide the project into appropriate tasks so that a large number of the public can participate in the project.

Furthermore, in order to properly manage the tasks on the project, a method for properly evaluating the tasks is required, but research on this is insufficient.

Laid-Open Patent Publication No. 10-2014-0095956, 2014.08.04

The problem to be solved by the present invention is to provide an online work-based project division method that analyzes the difficulty of content for each project and divides it into an appropriate level of difficulty when the difficulty of the content is high so that many workers can participate in the project, and an apparatus for performing the same will do

The problem to be solved by the present invention is to provide an online work-based project task management method in which the tech on the project is appropriately divided by difficulty to induce a large number of workers to participate in the corresponding project.

The problem to be solved by the present invention is to provide an online work-based project task management method for evaluating each task in order to manage the tasks on the project in an optimal state in carrying out the project, and an apparatus for performing the same.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

A project-based task management method according to an embodiment of the present invention, in a method for managing a crowdsourcing-based project task performed by an apparatus,

grouping a plurality of tasks for performing the project into a plurality of task groups based on the number of the plurality of tasks; evenly distributing the source data for performing the project to the plurality of task groups; calculating an evaluation score for processing the source data of each of the plurality of task groups from the project execution time to the evaluation time point by using an evaluation function determined based on the type of the project; Distributing the remaining source data not processed at the time of the evaluation to a task group corresponding to the highest evaluation score among the plurality of task groups;

The evaluation function is determined based on at least one evaluation factor and a weight corresponding to the at least one evaluation factor,

The at least one evaluation factor includes a labor cost consumed for processing the source data, a time consumed per result data in the data redistribution period, a quality of the result data, a normalized progress rate for outputting the result data, and the at least one of the difficulty levels of processing the source data;

The weight may be determined based on the type of the project.

The evaluation time is determined by the sum of the data redistribution periods having a period,

The period of the data redistribution period may be determined as a maximum value from a time point when the source data is equally distributed to the plurality of task groups, and a time during which each of the plurality of task groups processes the source data at a predetermined ratio.

The task management method according to an embodiment of the present invention includes: determining evaluation scores of the plurality of task groups for each data redistribution period from the project start time; and redistributing the remaining source data based on a ratio of the evaluation score of a specific task group to the sum of the evaluation scores of each of the plurality of task groups for each data redistribution period.

The redistributing of the remaining source data may be repeatedly performed until a point in time when a sum of evaluation scores corresponding to each of the plurality of task groups becomes a maximum value.

The evaluation time may be determined as a time when the sum of evaluation scores corresponding to each of the plurality of task groups becomes a maximum value.

The consumed cost may be determined as the sum of the total points paid for outputting the result data with respect to the number of result data determined based on the distributed source data.

The time consumed per result data in the data redistribution period is,

It may be determined as the sum of the work time and inspection time of the result data with respect to the number of result data output during the data redistribution period.

The quality of the result data may be determined as a sum of the number of rejections of the result data to the number of result data determined based on the distributed source data.

The normalized progress rate for outputting the result data is based on the number of result data output from the source data for the time each of the plurality of task groups actually processed the source data during the plurality of data redistribution periods can be decided.

The difficulty of processing the source data may be determined as the number of workers corresponding to a predetermined level or less with respect to the number of workers participating in each of the plurality of task groups.

An online work-based project task management method and an apparatus for performing the same according to an embodiment of the present invention analyze the difficulty of the content for each project, and if the difficulty of the content is high, divide it into an appropriate difficulty level so that many workers can participate in the project. can

The online work-based project task management method according to an embodiment of the present invention can induce a large number of workers to participate in the project by appropriately dividing the tech on the project by difficulty level.

The online work-based project task management method and the apparatus for performing the same according to an embodiment of the present invention may evaluate each task in order to manage the tasks on the project in an optimal state in carrying out the project.

The online work-based project task management method according to an embodiment of the present invention can minimize the influence of unexpected situations or variables in advance by dividing or adjusting the task using the results of actual execution.

The online work-based project task management method according to an embodiment of the present invention optimizes the initial work division configuration by using the data accumulated through the described process, thereby reducing the time required to find the optimized configuration. .

The online work-based project division method and the apparatus for performing the same according to an embodiment of the present invention may be capable of calculating an appropriate unit price by dividing the content into content having an appropriate difficulty for each project.

The online work-based project division method and apparatus for performing the same according to an embodiment of the present invention can reduce the time required for many workers to participate in the project and receive result data on the project.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a block diagram of an online work-based project task management apparatus according to an embodiment of the present invention.
2 is a diagram illustrating an operation of grouping tasks constituting a project according to an embodiment of the present invention.
3A and 3B are diagrams for explaining an operation of distributing source data according to an embodiment of the present invention.
4A and 4B are diagrams for explaining an operation of distributing the remaining source data to each task group according to an embodiment of the present invention.
5 is a diagram for explaining an operation of processing source data by further subdividing a task according to an embodiment of the present invention and calculating an evaluation score for the same.
6 is a diagram for explaining an operation of calculating an evaluation score of each task according to an embodiment of the present invention.
7 is a diagram for explaining an operation of distributing source data to a plurality of task groups formed by the number of tasks according to an embodiment of the present invention.
8 is a diagram for explaining an operation related to an evaluation time point and a data redistribution period according to an embodiment of the present invention.
9 is a diagram for explaining an operation of distributing source data based on an evaluation score of each task according to an embodiment of the present invention.
10 is a diagram for explaining an operation of determining an operation of each task group based on an evaluation score according to an embodiment of the present invention.
11 is a diagram for explaining the contents of an evaluation function according to an embodiment of the present invention.
12A is a diagram illustrating a project and contents constituting the project according to an embodiment of the present invention, and FIG. 12B is a diagram illustrating an overall operation of dividing a project based on content according to an embodiment of the present invention.
13 is a diagram for explaining an operation of dividing a content group into sub content groups according to an embodiment of the present invention.
14 is a diagram for explaining an operation of calculating a working time according to a data checklist according to an embodiment of the present invention.
15 is a view for explaining an operation of calculating a working time according to the participation of an operator according to an embodiment of the present invention.
16 is a diagram for explaining an operation of dividing content according to work time according to an embodiment of the present invention.
17 to 21 are flowcharts according to an embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

Before the description, the meaning of the terms used herein will be briefly described. However, it should be noted that the description of terms is for the purpose of helping the understanding of the present specification, and is not used in the meaning of limiting the technical idea of the present invention unless explicitly described as limiting the present invention.

Meanwhile, in the present specification, the term 'device' includes various devices capable of providing a result to a user by performing an arithmetic process. For example, devices may be in the form of computers and mobile terminals. The computer may be in the form of a server that receives a request from a client and processes information. The mobile terminal includes a mobile phone, a smart phone, a personal digital assistant (PDA), a portable multimedia player (PMP), a navigation system, a notebook PC, a slate PC, a tablet PC, and an ultrabook. ), a wearable device, for example, a watch-type terminal (smartwatch), a glass-type terminal (smart glass), a head mounted display (HMD), and the like.

The present invention relates to a method for efficiently operating a project in which many participants participate in a project performed based on crowding sourcing, and an apparatus for performing the same.

To this end, according to an embodiment of the present invention, as will be described later, source data having a large deviation in difficulty before or during the project may be filtered and processed through a different path.

According to an embodiment of the present invention, it is possible to divide and adjust the task related to the source data filtered before or during the progress of the project into several sub-tasks of appropriate difficulty.

Furthermore, according to an embodiment of the present invention, it is possible to compare and evaluate an advantageous method among one big task and a plurality of sub-tasks.

Hereinafter, in FIGS. 1 to 11, a method of managing the task of the project based on the evaluation of each task will be described,

A method of managing a project task through grouping by difficulty of content will be described in FIGS. 12A to 16 .

17 to 21, the operation of the present invention will be described chronologically.

In the present specification, the worker terminal 30 may be a device used by the worker performing the project after the unit price prediction of the project is completed in the device 10 . Here, the worker terminal 30 may participate in the project opened on the platform through an application or website provided by the device 10 .

In the present specification, the request server may refer to a computing device that requests the device performing the project to perform a specific project.

The request server 20 may be a server that requests an online work-based project for the purpose of collecting source data or data annotation for the generation of artificial intelligence learning data.

In this specification, a task may mean a set of sub-operations for performing the entire project.

In the present specification, the evaluation function and evaluation score may refer to a method for evaluating a task for managing the task and an output score derived through evaluation.

In this specification, source data may refer to data input to perform a project.

In the present invention, content may mean data for performing a project included in source data on a corresponding project.

In the present specification, the search window may refer to a component of a project that evaluates each task using an evaluation function and an evaluation score.

In the present specification, result data may refer to data output when source data is input to a project.

1 is a block diagram of an online work-based project task management apparatus 10 according to an embodiment of the present invention.

Referring to FIG. 1 , an apparatus 10 for performing the operation of the present invention may include a memory 120 , a communication unit 110 , and a control unit 130 .

Also, the device 10 may communicate with the request server 20 and the operator terminal 30 .

The memory 120 may store various data for performing a project.

Specifically, the memory 120 may store source data for performing a project and an evaluation score for evaluating each task.

The memory 120 is a non-volatile memory device or RAM (such as a cache, read only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), and flash memory (Flash memory). It may be implemented as at least one of a volatile memory device such as a random access memory, a hard disk drive (HDD), or a storage medium such as a CD-ROM, but is not limited thereto. The memory 120 may be a memory implemented as a chip separate from the processor described above with respect to the controller 130 , or may be implemented as a single chip with the processor.

The communication unit 110 may communicate with the request server 20 and the worker terminal 30 .

The communication unit 110 may include one or more components that enable communication with an external device, and may include, for example, at least one of a short-range communication module, a wired communication module, and a wireless communication module.

In particular, the wireless communication module includes, in addition to the Wifi module and the Wireless broadband module, Global System for Mobile Communication (GSM), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), and universal mobile telecommunications (UMTS). system), Time Division Multiple Access (TDMA), Long Term Evolution (LTE), etc. may include a wireless communication module supporting various wireless communication methods.

The control unit 130 may include a task management unit 131 , a central processor 133 , a source data distribution unit 132 , a search window management unit 134 , and an optimization task determination unit 135 .

The search window manager 134 may additionally include an evaluation function determiner 136 .

The controller 130 may form a plurality of tasks for performing a project.

According to an embodiment of the present invention, a project may mean performing a specific operation on an object of an image provided as source data.

The work type may include a bounding box work, a segmentation work, a sentiment work, and the like.

Here, first, the bounding box task may be a task of selecting a work target as a bounding box in the image.

Next, the segmentation task may be a task of classifying the work object along a border in the image.

Next, it may be a task of classifying a sentence or at least one word in the sentiment text according to the work target condition.

The controller 130 may group the plurality of tasks into a first group and a second group including the plurality of tasks.

According to an embodiment of the present invention, the first unit corresponding to the first group may mean a big task in which most tasks for performing a project are grouped.

On the other hand, the second unit corresponding to the second group may mean a sub task grouped by dividing a plurality of tasks for performing a project.

That is, the first unit may mean a large number of tasks for performing a project, that is, a big task, and the second unit may mean a sub-task grouped into a smaller number of tasks than the first unit.

The controller 130 may distribute some of the source data used to perform the project to the first task group and the second group.

The distributed source data may be processed in each group for a predetermined time.

In order to evaluate each group before carrying out the project in earnest, the controller 130 may decompose some of the source data into each group to process the source data.

According to an embodiment, the source data distribution unit 132 of the control unit 130 may distribute half of the source data performed on the project to the first group and the other half to the first group.

The controller 130 may calculate an evaluation score for processing the source data of each of the first task group and the second task group by using the evaluation function determined in response to the project.

Meanwhile, in calculating the evaluation score, the processor may control the plurality of pre-formed search windows to match the first task group and the second task group, respectively, to calculate the evaluation score.

Specifically, the control unit 130 is provided with a search window management unit 134 , and each search management unit management unit may include an evaluation function determination unit.

The control unit 130 may evaluate the source data processing result of each group by using the evaluation function through this functional structure.

The evaluation function used by the controller 130 may consider cost and time required to process the source data, the quality of output data, and the speed of the project.

The controller 130 may distribute the remaining unprocessed source data among the source data to the first task group or the second task group based on the evaluation score.

Here, the remaining source data may refer to the remaining source data excluding the source data used to calculate the evaluation score.

Meanwhile, the control unit 130 may use the evaluation function to evaluate each task and the group including the task.

The evaluation function considers the cost consumed for processing the source data, the time consumed for outputting the result data based on the source data, the quality of the result data, the progress speed of outputting the result data, and the difficulty of performing the task of each task group can be formed by

Meanwhile, in addition to the above-described operation, the controller 130 may determine the optimization task group by differently determining the task group based on the number of tasks for performing the project from the beginning.

Specifically, the controller 130 may group a plurality of tasks for performing a project into a plurality of task groups based on the number of the plurality of tasks.

That is, the control unit 130 may group each task differently based on the number of tasks for performing the project from the start of performing the pilot of the project.

According to an embodiment of the present invention, if there are 10 tasks, the task can be grouped by dividing it into 10 methods.

The controller 130 may equally distribute the source data for performing the project to the plurality of task groups. As will be described later, the source data may be redistributed, but at the first point in time before the redistribution is performed, the controller may evenly distribute the data.

Meanwhile, the controller 130 may calculate an evaluation score for processing the source data of each of the plurality of task groups from the time of project execution to the time of evaluation by using the evaluation function determined based on the type of the project.

The evaluation function may be determined differently depending on the project being performed.

The evaluation time point may mean a time point at which the control unit determines an optimal task structure for performing a project.

According to an embodiment, the evaluation time point may mean a time point at which the maximum value of the evaluation score of each task group is formed.

The controller 130 may distribute the remaining source data that has not been processed at the time of evaluation to a task group corresponding to the highest evaluation score among the plurality of task groups. That is, the controller may distribute the most source data to the optimized task group by distributing all remaining source data to the most efficient task group at the time of evaluation.

Meanwhile, the above-described evaluation function may be determined based on at least one evaluation factor and a weight corresponding to the at least one evaluation factor.

Specifically, the at least one evaluation factor includes a labor cost consumed for processing the source data, a time consumed per result data in the data redistribution period, the quality of the result data, a normalized progress rate for outputting the result data, and It may include at least one of the difficulty of processing the source data.

In addition, the weight may be determined based on the type of the project. Hereinafter, the operation of the above-described control unit will be described in detail.

At least one component may be added or deleted according to the performance of the components of the device 10 shown in FIG. 1 . In addition, it will be readily understood by those skilled in the art that the mutual positions of the components may be changed corresponding to the performance or structure of the system.

Meanwhile, each component shown in FIG. 1 means a hardware component such as software and/or a Field Programmable Gate Array (FPGA) and an Application Specific Integrated Circuit (ASIC).

2 is a diagram illustrating an operation of grouping tasks constituting a project according to an embodiment of the present invention.

Referring to FIG. 2 , several tasks may exist on the project when the project is performed. In FIG. 2, for convenience of description, it is assumed that there are six total tasks for performing the project.

Meanwhile, the controller may group all tasks T21 into one group G21. Specifically, in FIG. 2 , all tasks for which T21 performs a project may form a group in which one unit is grouped.

In addition, the processor may group several tasks (T22, T23) into different units (G22).

That is, the controller may group the plurality of tasks for performing the project into a first unit T21 and a second unit T22 and T23. In FIG. 2 , the first unit T21 may mean a unit of a big task in which six tasks are grouped.

On the other hand, the second units T22 and T23 may mean a unit of a sub-task grouped as a unit of three tasks among six tasks for performing a project.

In addition, the controller may determine the first task group G21 for processing some source data by using the task grouped in the first unit.

A second task group G22 that processes some other source data may be determined by using the task grouped in the second unit.

Meanwhile, the controller may calculate an evaluation score for the processing result of the source data of each of the first and second task groups based on the evaluation function determined in response to the project through the search window as described above. A detailed description related thereto will be provided later.

Meanwhile, as shown in FIG. 2 , the number of tasks grouped in the first unit among the plurality of tasks may be provided to exceed the number of tasks grouped in the second unit among the plurality of tasks.

That is, in FIG. 2 , the number of tasks in the first unit is six, and the number of tasks in the second unit is three.

In FIG. 2 , the task included in the project is merely an embodiment of the present invention, and there is no limitation thereto in the type and number of tasks included in the project and the grouping method.

3A and 3B are diagrams for explaining an operation of distributing source data according to an embodiment of the present invention.

Referring to FIG. 3A , the controller may distribute some of the source data D31 used for the project D31-1 and D31-2 to each group.

The controller may distribute the source data D31 corresponding to some of the source data to each group. Through some source data D31, the control unit may perform a pilot for evaluating each group performing a project.

Referring to FIG. 3A , the control unit corresponds half of the source data to the first task group G311 grouped as the big task T311 and the remaining source data to the second task group T312 grouped as the sub tasks T312 and T313. ) can be matched.

The controller may perform a pilot operation on a task group to which each source data is distributed.

Meanwhile, as described above, the controller may evaluate the result of processing the source data of each group through the search window corresponding to each group.

The control unit may calculate the evaluation score of the data processing result of each group performed based on the source data through the search windows W311 and W312 formed in advance.

That is, the search windows W311 and W312 are previously formed in the device, and the result of processing the source data of each group can be monitored.

According to an embodiment of the present invention, when each group processes source data as shown in FIG. 3 , the first search window W311 evaluates the operation of the task included in the first task group G311, The second search window W312 may derive an evaluation score by evaluating the operation of the sub-task included in the second task group G312.

Although FIG. 3A shows that a group is formed by dividing the task into two types of units, the operation of dividing the type of the group is not limited thereto.

Referring to Figure 3b, the control unit is a task group (G321) grouped with six tasks (T321), a task group (G322) grouped with three tasks (T322, T323) and two tasks (T324, T325, T326) It shows that a grouped group (G323) is formed.

In the case of FIG. 3B , the first source data D21 may be divided into thirds and transmitted to each group.

In addition, each group may process source data through differently grouped tasks, and in this case, three search windows (W321, W322, W333) may be provided.

That is, as long as the control unit determines the grouping form of the optimized task by grouping each task performing the project in a different form, there is no limitation on the operation.

4A and 4B are diagrams for explaining an operation of distributing the remaining source data to each task group according to an embodiment of the present invention.

Referring to FIG. 4A , when the evaluation score of the first task group G411 is higher than the evaluation score of the second task group G412 in the evaluation score calculated based on the above-described operation, the controller determines the remaining source data D41 ) can be distributed to the first task group.

Here, the remaining source data may mean the remaining data except for the source data processed in each group for a predetermined time in order to evaluate the source data processing capability of each group.

4A shows a case in which performing a project in one big task is more efficient than dividing the task, and the processor may distribute the remaining source data D41 to the first task group G411. .

The controller may distribute the remaining source data D41 to the first task group G411 in order to process the source data as a big task T411 grouped into a large number of tasks in order to perform the corresponding project.

Meanwhile, referring to FIG. 4B , when the evaluation score of the second task group G422 is higher than the evaluation score of the first task group G421, the operation of distributing the remaining source data D42 to the second task group G422 is performed. is indicating

The control unit is grouped into six tasks T432 to process source data in a first task group G421 and three tasks T422 and T423 to process source data of a second task group G422 The result may be evaluated, and as a result, if the evaluation score of the second task group is high, the remaining source data D42 may be distributed to the second task group G422.

In this case, the controller may stop the operation of the task included in the first task group G421 corresponding to the remaining source data.

Thereafter, the controller may determine an optimized task group by grouping the tasks for the second task group G422 differently, which will be described later.

5 is a diagram for explaining an operation of processing source data by further subdividing a task according to an embodiment of the present invention and calculating an evaluation score for the same.

When the source data evaluation score of the task group grouped by dividing the task as shown in FIG. 4B is high, the control unit may additionally divide and group the task and process the source data.

4B and 5 together, when the evaluation score of the second task group G422 is higher than the evaluation score of the first task group G421, the control unit reduces the number of tasks included in the second task group G422. It is possible to form different groups G51 and G52 by dividing by numbers.

That is, in the above-described operation, a high evaluation score of a task group that is grouped into a plurality of tasks and processes source data means that it is efficient to divide the operation into several tasks in the corresponding project.

Therefore, the control unit can process the source data by dividing the task more than before and dividing each task into smaller units.

In FIGS. 4B and 5 , as a case in which there are six tasks for performing the project, the second group G422, which is grouped by three existing tasks and processed source data, is grouped by three tasks, and a third task for processing source data The group G51 and the task are grouped by two and are changed to a fourth task group G52 that processes source data, indicating that each source data is processed.

In this case, the controller may compare the evaluation scores of the third task group G51 and the fourth task group G52.

When the evaluation score of the third task group G51 is higher than the evaluation score of the fourth task group G52, it is determined that the task grouping of the third task group G51 is optimized. The project can be carried out in the form of task grouping of

On the other hand, when the evaluation score of the fourth task group G52 is high, the control unit may further subdivide the tasks constituting the fourth task group G52 to form a grouped task group to evaluate the processing result of the source data. .

The controller may determine an optimized task grouping form by continuously repeating this operation.

In addition, when the evaluation score of the task group grouped by two is high, the controller may change the task group grouped by two into a task group grouped by one, and compare the source data processing results.

Like this operation, each task is grouped one by one, and the sum of the evaluation scores of each task group is compared, and a task group optimized for carrying out the project can be determined.

The processor may repeat the subdivision of the task until the evaluation acceptance of each task group reaches a maximum value.

6 is a diagram for explaining an operation of calculating evaluation scores ( S61 , S62 , and S63 ) of each task according to an embodiment of the present invention.

The control unit may determine an evaluation function corresponding to the project to be performed.

The controller may calculate the evaluation scores S61, S62, and S63 for each task after the project is performed for a certain unit period of each task group.

The evaluation function used by the control unit includes the cost consumed for processing the source data, the time consumed for outputting the result data based on the source data, the quality of the result data, the progress speed of outputting the result data, and the value of each task group. It may be formed in consideration of the difficulty of performing the task.

6 , it may be shown that the processor calculates the evaluation score S61 of the first task group composed of the big task unit and the evaluation score S62 and S63 of the second task group composed of each sub-task unit.

The processor may compare the source data processing evaluation score S61 of the big task unit with the evaluation score of the sub-task unit constituting the second task group to determine which grouping form of the task is effective in performing the corresponding project.

7 is a diagram for explaining an operation of distributing source data to a plurality of task groups formed by the number of tasks according to an embodiment of the present invention.

Referring to FIG. 7 , the controller may group a plurality of tasks for performing a project into a plurality of task groups G71, G72, G73, G74, and G75.

7 shows an embodiment when the number of tasks for performing a project is six, and since there are six tasks, it is shown that the group is grouped into six task groups.

7 shows that task grouping is performed according to the number of tasks, but there is no limit to the number of grouped tasks. The number of existing tasks may be formed by N!

The controller may equally distribute the source data D7 for performing the project to the plurality of task groups for evaluation of each task.

Evaluation of task group Initially, the control unit may start task evaluation by equally distributing each source data to each task group.

Meanwhile, the search window provided in the control unit may evaluate source data processing of each task group.

Meanwhile, the controller may calculate an evaluation score for processing the source data of each of the plurality of task groups from a project execution time to an evaluation time point by using an evaluation function determined based on the type of the project.

Meanwhile, when the evaluation of each task group is completed by the evaluation time, the remaining source data not processed at the evaluation time may be distributed to the task group D76 corresponding to the highest evaluation score among the plurality of task groups.

In FIG. 7 , when the evaluation score of the task group grouped into five tasks is the highest, the control unit may distribute source data not processed until the evaluation time point to the task group D76 grouped with all five tasks.

Meanwhile, grouping tasks and distributing source data by the control unit is only a basic embodiment of the present invention, and the operation of the present invention described with reference to FIG. 7 is not limited.

8 is a diagram for explaining an operation related to an evaluation time point and a data redistribution period according to an embodiment of the present invention.

As described with reference to FIG. 7 , the control unit may evaluate the processing capability of each task until the time of evaluation.

The evaluation time may be determined as a time point T at which the sum of evaluation scores corresponding to each of the plurality of task groups becomes the maximum value.

Specifically, when the control unit distributes the source data to each task at a specific ratio, the time point T at which the sum of the evaluation scores of each task becomes the maximum may be calculated, and the control unit may determine the time point as the evaluation time point.

In addition, the evaluation time may be determined as the sum of data redistribution periods having periods.

8 shows a plurality of data redistribution periods provided with a period of t.

8 shows that the evaluation time T is determined by the sum of the six redistribution periods.

On the other hand, the period of the data redistribution period may be determined as the maximum value from the time when the source data is equally distributed to the plurality of task groups, and the time during which each of the plurality of task groups processes the source data at a predetermined ratio.

For example, when the ratio is 0.1, the controller may measure a time for outputting source data of 0.1 among the distributed source data as result data.

Depending on the task group, the processing time of 0.1 source data may be determined differently. Here, the controller may determine the longest time consumed in each task group as the data redistribution time.

As described above, the control unit may distribute each of the source data to each task group, and may determine the amount of source data consumed to be used for evaluation of each task group.

Meanwhile, the control unit may determine the evaluation scores of the plurality of task groups for each data redistribution period from the start of the project.

The control unit may redistribute each source data based on the score for each data redistribution period, and the redistribution of the remaining source data is performed when the sum of the evaluation scores corresponding to each of the plurality of task groups becomes the maximum value, that is, the evaluation It can be repeated up to a point in time.

9 is a diagram for explaining an operation of distributing source data based on an evaluation score of each task according to an embodiment of the present invention.

The controller may redistribute the source data to a plurality of groups based on the evaluation score.

The controller may redistribute the sword data to each task group for each redistribution period. Specifically, the control unit may equally distribute each source data to each task group at the initial time point for evaluating each task group, but when the redistribution period arrives after the time point at which the source data is equally distributed, the evaluation score of each task group The remaining source data can be distributed in proportion to .

Specifically, the controller may redistribute the source data based on Equation 1 below.

Referring to Equation 1, Ts may mean the total amount of redistributed source data, Tl may mean the amount of source data distributed to a specific task group, and Pl may mean an evaluation score of a specific task group. N may mean the total number of task groups.

That is, the control unit distributes the source data as much as the evaluation score corresponding to the task group from the sum (P1+P2+P3+P4) of the evaluation scores of each task group (G71, G72, G73, G74). A large amount of source data can be assigned to a group and a small amount of source data can be assigned to a task group with a small amount of source data.

That is, the control unit determines the evaluation score of the plurality of task groups for each data redistribution period from the start of the project, and the ratio of the evaluation score of a specific task group to the sum of the evaluation scores of each of the plurality of task groups for each data redistribution period Based on You can distribute the remaining source data that has not been redistributed with

Since detailed descriptions of the data redistribution period and evaluation time have been described above, detailed descriptions thereof will be omitted.

Meanwhile, the operation of redistributing the remaining source data described with reference to FIG. 9 is only an embodiment of the present invention, and there is no limitation on the operation of redistributing the source data.

10 is a diagram for explaining an operation of determining an operation of each task group (G81, G82, G83) based on an evaluation score according to an embodiment of the present invention.

Specifically, the control unit may calculate an evaluation score for processing the source data of each of the plurality of task groups (G101, G102, G103) from the time of project execution to the time of evaluation by using the evaluation function determined based on the type of the project.

The controller may distribute the remaining source data D10 that is not processed at the time of evaluation to a task group corresponding to the highest evaluation score among the plurality of task groups.

10 illustrates a case in which the evaluation score of the first task group G101 is the highest. That is, in performing the project, the control unit forms each task group G101, G102, and G103, and at the time of evaluation, the evaluation score of the first task group G101 is calculated as the highest.

Accordingly, the controller may distribute the remaining source data D10 to the first task group G101.

Based on the above-described operation, when a corresponding project is performed with a plurality of tasks, the control unit may perform the project in an optimized method by driving source data to a task having a high evaluation score.

11 is a diagram for explaining the contents of an evaluation function according to an embodiment of the present invention.

The evaluation function F11 used in the present invention includes the cost P11 consumed for processing the source data, the time consumed for outputting the result data based on the source data T11, and the quality of the result data Q11. , and a progress speed (V11) of outputting the result data, and a difficulty (D11) of the project, and a weight corresponding to each variable. Such an evaluation function may be determined based on Equation 2 below.

Referring to Equation 2, Pn may mean an evaluation score of each task, and w1 to w4 may mean weights determined based on a project.

In addition, NC means cost, NWT means time consumed, NR means quality, and NS means project progress speed. In addition, LWN may mean the difficulty of the task reflected in each task group.

Here, the cost consumed may be determined based on Equation 3 below.

Referring to Equation 3, rk may mean the number of result data, and vk may mean the number of points paid to output the result data in project execution.

That is, the cost may be determined as the sum of the total points paid for outputting result data for the number of result data determined based on the distributed source data.

Here, the consumption time may be determined based on Equation 4 below.

Referring to Equation 4, Ti1 may mean an output time of the result data, and Ti2 may mean an inspection time of the corresponding result data.

That is, the consumption time may be determined as the sum of the output time and the inspection time of the result data with respect to the number of result data determined based on the source data.

Also, the quality of the result data may be determined based on Equation 5 below.

Referring to Equation 5, rk may mean the number of result data, and jk may mean the number of rejections upon inspection of each result data.

That is, the quality of data may be determined as the sum of the number of rejections of the result data to the number of result data determined based on the distributed source data.

Meanwhile, the speed of outputting the result data may be determined based on Equation 6 below.

Referring to Equation 6, NS may mean a progress speed, and T _r may mean a time during which actual data is processed during a data redistribution period.

D _r may mean the number of result data output based on the source data.

That is, the progress rate may be determined based on the number of output result data for the time spent processing actual data in the period for evaluating the source data processing capability of each task group, that is, the data redistribution period.

Meanwhile, the difficulty level may be determined based on the following equation.

Referring to Equation 7, LWN means difficulty, P _beg means the number of low-level workers participating, and P _tot means the total number of participating workers.

That is, the control unit may determine the difficulty level as the number of workers corresponding to the number of workers participating in each of the plurality of task groups is less than or equal to a predetermined level.

On the other hand, the above-described evaluation function is only an example for performing the operation of the present invention, and there is no limitation on the form and operation of the evaluation function.

The above-described operation described the operation of maximizing the efficiency of the project through task management in which the configuration of the task for performing the project is different.

Hereinafter, an operation for maximizing the efficiency of the project by task management through changing the contents of the source data used in the project will be described.

12A is a diagram illustrating a project and contents constituting the project according to an embodiment of the present invention.

Referring to FIG. 12A , since source data is required to perform a corresponding project, each work content C121 , C122 , and C123 may mean that the source data C12 is divided into predetermined reference units.

For example, when the source data is an image, the work content of the reference unit may be determined as one image.

12A shows that a bounding box operation is performed on each work content based on an image as the source data C12 and a labeling operation is performed thereon.

On the other hand, the level of difficulty in performing the bounding box work and labeling may be determined to be high depending on the number of objects or the type of bounding box and labeling.

The controller may determine the difficulty score of each content based on the type of the source data C12 and the task to be processed on the source data. Additionally, the work target object may be considered in determining the difficulty score.

Each task difficulty may be determined based on the following equation.

Referring to Equation 8, diff means the difficulty of each task content, n _target means the number of objects that have performed the task, n _tot means the total number of objects included in the source data, and D is the processing task It can mean the number of kinds.

For example, when performing bounding and labeling operations on source data, if the number of objects included in the source data is 10, there are 2 objects that need to be bounded, and there is only 1 object that performs labeling, the difficulty is set to 1. can be decided.

Meanwhile, in calculating the difficulty, when the source data corresponds to an image, the task difficulty score of each task content may be calculated based on the type of the pre-processing library that pre-processes the source data.

For example, if the source data is an image, scikit-image, Numpy, Scipy, PIL/Pilow, OpenCV-Pyrthon, SimpleCV, Mahotas, SimpleITK, pgmagick, Pycairo, etc. can be used in the preprocessing process, but only when the source data is an image A difficulty score can be calculated by considering the library type.

When the number of objects to be processed based on the source data is large and the amount of objects on the work content is large, the difficulty may be determined to be high.

At least one content may be grouped into a plurality of groups G121 and G122 based on a difficulty score corresponding to each work content.

According to an embodiment of the present invention, the first content group G101 may be formed as a group of contents in which the difficulty score of each content exceeds a predetermined score, and the second content group G122 has a predetermined difficulty score. It may be formed into a group of content that is less than a score.

On the other hand, the second content group is evaluated with an appropriate level of difficulty, so that the project can be opened later and the work can be performed immediately.

On the other hand, the first content group may be divided to form subgroups as will be described later.

12B is a diagram illustrating an overall operation of dividing a project based on content according to an embodiment of the present invention.

The controller may extract the source data S12 used to perform the project.

Meanwhile, the controller may extract at least one content C12 processed based on the source data S12 from the request server.

In addition, the control unit may analyze the project. Also, the controller may extract an object Ob12 used for a corresponding project and content from the source data.

Meanwhile, the controller may determine the difficulty score of the at least one content based on the type of the at least one content C12 and the number of objects Ob12.

The controller may calculate a difficulty score D12 corresponding to each of the at least one content.

Also, the controller may group at least one content into a plurality of groups based on the difficulty score (G12).

Due to the division and grouping of the content itself, each content may be provided with a lower difficulty than when it is provided integrally.

The controller may determine a high difficulty group in which the sum of the difficulty scores of the at least one content among the plurality of groups exceeds a predetermined difficulty score.

This high difficulty group may be interpreted as having a high level of difficulty even though the contents are divided and grouped.

In this case, the controller may divide at least one content corresponding to the high difficulty group into a plurality of subgroups.

Meanwhile, when the source data is an image, an operation performed on the source data may be determined as at least one of a bounding box operation, OCR, labeling, and segmentation operation of the object.

The bounding box operation may be an operation of selecting a work target as a bounding box in the image.

The labeling operation may refer to an operation of matching a specific meaning or character to an object.

An optical character reader (OCR) may refer to an operation of reading typed characters on source data.

The segmentation operation may be an operation of classifying the work object along a border in an image.

The control unit may create a project by grouping the contents by difficulty based on the above-described operation (P12), and match a worker to each project (H12).

In addition, the control unit may calculate the working time required to perform each project based on the matched project and the worker (T12).

The object may be an object included in an image or text set by the request server 20 as a target for which work should be performed. For example, the object corresponding to the work target in the image may include a preset person, vehicle, cat, dog, cup, and the like. In addition, the object corresponding to the work target in the text may include some characters such as preset words, sentences, phrases, and keywords in the text.

13 is a diagram for explaining an operation of dividing a content group into sub content groups according to an embodiment of the present invention.

As described above, the controller performing the operation of the present invention may determine a high difficulty group in which the sum of the difficulty scores of at least one content among the plurality of content groups G131 and G132 exceeds a predetermined difficulty score.

13 shows that the first content group G131 is determined to be a high difficulty group.

For the content group G132, which is not of high difficulty, a project may be opened immediately and work may be performed.

Meanwhile, the content group G131 determined to be a high difficulty group with high difficulty may be divided into subgroups SG131 and SG132.

Meanwhile, the controller may calculate the difficulty level of the divided subgroups SG131 and SG132.

That is, the controller may calculate a task difficulty score corresponding to the subgroups SG131 and SG132 divided from the first content group G131.

On the other hand, if the task difficulty score corresponding to the subgroup exceeds the predetermined reference difficulty score, the processor may repeat the operation of dividing the subgroup until the task difficulty score corresponding to the subgroup is less than the predetermined reference difficulty score. have.

For example, referring to FIG. 13 , the processor groups each work content, and the sum of the difficulty scores of the first content group G131 may exceed a predetermined difficulty score.

In this case, it may mean a case in which it is difficult for many workers to participate because the difficulty of the group is high.

The processor may divide the content corresponding to the high difficulty group G131 into a plurality of subgroups SG131 and SG132.

When the subgroup is continuously divided and the difficulty score corresponding to the divided group (SG1311, SG1312, SG1313, SG1314) is less than the predetermined reference difficulty level, the processor may open the project and perform the work.

Meanwhile, each of the divided groups forms a separate sub-project, and the workers H1311, H1312, H1313, H1314, and H1315 can participate in each group to perform work.

Also, the processor may form result data corresponding to the project by merging the result data respectively output from the first content group and the second content group as described above.

Meanwhile, the controller may calculate a unit price corresponding to the plurality of groups and the plurality of subgroups based on the number of objects and the difficulty score. When the difficulty score of the corresponding group is low or the number of objects to be processed is small, the unit price can be calculated low.

In other words, since the content is divided, the worker can participate in the project by content. Even with the same content, if the number of objects to be processed is large or the difficulty score of the content is high, the processor can calculate the unit price of the group correspondingly higher. have.

On the other hand, the operation described in FIG. 13 is only an embodiment of the present invention, and there is no limitation on the type and number of contents or the operation of grouping such contents according to the present invention.

14 is a diagram for explaining an operation of calculating a working time according to a data checklist according to an embodiment of the present invention.

Referring to FIG. 14 , a client performing a project may input a type of source data, a processing task to be processed in the source data, and the like through a data checklist by the client before performing the task. In addition, the client can input human resource (HR) information suitable for carrying out the project through the data checklist. A detailed description related to HR information will be described later.

For example, the client may input whether the source data is an image through the data checklist DC14.

12 shows a case where the source data SD14 to be processed by the client is an image (I141), and operations performed on the source data are masking, labeling, and OCR (I142). The information input by the client may be collected by the control unit.

The control unit may determine the difficulty score of the corresponding source data and content constituting the source data based on the information input by the client into the data checklist.

For example, when the client does not input a task such as OCR on the source data, the processor may not evaluate the object of OCR in determining the difficulty score of each content.

That is, the processor may calculate the difficulty information of each content based on the information input by the user through the data checklist as shown in FIG. 14 .

15 is a view for explaining an operation of calculating a working time according to the participation of an operator according to an embodiment of the present invention.

Referring to FIG. 15 , the controller may calculate the number of workers participating in a plurality of groups using a predetermined human pool.

Specifically, the control unit may receive HR (Huma Resource) information suitable for the project from the client through the predetermined data checklist. For example, HR information may be determined under conditions such as "a worker who performed labeling work 5 or more times". On the other hand, a pool containing such HR information may be prepared in the database of the system performing the project, and the processor may determine the number of workers corresponding to the HR information input by the client in the pool.

That is, the control unit may calculate the workers participating in the plurality of groups by searching for workers corresponding to the HR information in a database stored in advance.

FIG. 15 shows that the worker searched in this way corresponds to each content.

Specifically, the first workers H151 corresponded to the first work content C151 shown in FIG. 15 .

Also, the first workers H152 correspond to the second work content C152.

Also, the first workers H153 corresponded to the third work content C153.

Meanwhile, in this case, the processor may calculate the working hours corresponding to the plurality of groups based on the data information input by the user and the number of workers.

According to an embodiment of the present invention, the working time may be calculated based on the following equation.

Referring to Equation 9, T denotes the working time of each group, N denotes the number of contents included in each group, R denotes the working time consumed per unit content, and H denotes the working time of the corresponding group. can mean

That is, the working time may be determined using the number of contents included in the plurality of groups, the reference time, and the number of workers.

On the other hand, this working time can also be used to divide the project, and a detailed description related to this operation will be described later.

16 is a diagram for explaining an operation of dividing content according to work time according to an embodiment of the present invention.

The controller may determine a delay group in which the working time exceeds a predetermined working time from among the plurality of groups.

In FIG. 16 , it is determined that C162 corresponds to the delayed content group and described.

16, the processor divides the entire project based on the difficulty of the work content in order to lower the difficulty, and the difficulty of each group means that a group that takes a lot of time to process the content of each group occurs. have.

For example, such a time-consuming group is a case in which many objects are included in the corresponding content, and the controller assigns the delay group C162 to the subgroups C162-1, C162-1, based on the number of objects included in the delay group C162. C162-2).

That is, in FIG. 16, since many objects are included in the work content included in the delay group C162, and it takes a lot of time to perform this, the delay group C162 is divided into subgroups C162-1 and C162-2. It can shorten the project execution time of the group.

Meanwhile, the control unit may calculate the optimization work time corresponding to the project based on each calculated work time and difficulty score.

The controller may divide the at least one content group C162 into sub content groups C162-1 and C162-2 so that the sum of the working hours of the plurality of groups reaches the optimization working time.

The control unit may continue the division of each group only when the condition that the working time of the group before division is greater than the sum of the working hours of each group after division is satisfied.

Such an operation may be expressed by the following equation.

_Referring to Equation 10, T _before indicates the working time of the group before division, and T after1 and T _after2 indicate the working time of each group after division.

The control unit may perform the division of each group only when the condition of Equation 2 is satisfied, and based on this, may derive the optimization working time of each group.

On the other hand, the content division of the present invention described with reference to FIG. 16 is only an embodiment for explaining the operation of the present invention, and there is no limitation on a value that is a basis for dividing each content.

Hereinafter, the overall operation of the present invention will be described in time series.

17 to 22 are flowcharts according to an embodiment of the present invention.

Referring to FIG. 17 , the controller may form a plurality of task groups for performing a project ( S1701 ).

In addition, the processor may group each task into a first task group and a second task group (S1702).

Thereafter, the control unit may calculate the evaluation score of each task group by using a search window prepared in advance (S1703).

The controller may distribute each source data based on the evaluation score calculated in this way (S1704).

Referring to FIG. 18 , after the processor calculates the evaluation score of the group including each task, the evaluation score of the second task group and the first task group may be compared ( S1801 ).

If the evaluation score of the first task group is greater than the evaluation score of the second task group (S1802), the processor may distribute the remaining source data to the first task group (S1803).

On the other hand, if the evaluation score of the second task group is greater than the evaluation score of the first task group, the remaining source data may be distributed to the second task group (S1804).

The control unit may perform a subsequent project based on the distributed source data (S1805).

Referring to FIG. 19 , the processor may extract source data (S1901). Also, it is possible to extract an object from the contents included in each source data thereafter (S1902).

Also, the processor may calculate a difficulty score based on the number of objects and the number of contents of each content (S1903).

In addition, the processor may group each content by using the difficulty score (S1904).

Meanwhile, when the difficulty score of the specific grouped content group exceeds the reference value, the difficulty score may be lowered by dividing the corresponding content group into sub content groups (S1905 and S1906). However, if the group grouped by content has an appropriate difficulty score, the project may be performed based on the group (S1907).

Referring to FIG. 20 , when the control unit opens a project (S2001), the project may calculate a worker who performs the corresponding project (S2002).

In addition, the processor may calculate the working time based on the number of workers, each content, and a predetermined reference time (S2003). However, even in this case, if the working time of a specific group exceeds a predetermined value (S2004), subgroups may be formed by dividing the group based on the number of objects included in each content (S2005).

However, if the working time of a specific group is less than the reference value, it is determined that the working time of the group is appropriate and the project can be performed (S2006).

Referring to FIG. 21 , an operation after a group is divided into sub content groups is shown.

Even after the group is divided into sub-content groups (S2101), if the difficulty level of the corresponding sub-group is greater than or equal to the reference value (S2102), the controller may continue to divide the group in order to lower the difficulty of the corresponding sub-group.

On the other hand, if the difficulty level of the sub content group corresponds to the difficulty level less than the reference value, the control unit may perform the project based on the corresponding grouping (S2103).

The steps of a method or algorithm described in relation to an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may contain random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can realize that the present invention can be embodied in other specific forms without changing its technical spirit or essential features. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

10: device
110: communication department
120: memory
130: control unit
20: request server
30: operator terminal

Claims (10)

A method for managing crowdsourcing-based project tasks performed by a project task management device, the method comprising:
grouping a plurality of tasks for performing the project into a plurality of task groups based on the number of the plurality of tasks;
evenly distributing the source data for performing the project to the plurality of task groups;
calculating an evaluation score for processing the source data of each of the plurality of task groups from the time of project execution to the time of evaluation by using an evaluation function determined based on the type of the project;
Distributing the remaining source data not processed at the time of the evaluation to a task group corresponding to the highest evaluation score among the plurality of task groups;
The evaluation function is
It is determined based on at least one evaluation factor and a weight corresponding to the at least one evaluation factor,
The at least one evaluation factor is
At least one of labor cost consumed for processing the source data, time consumed per result data in a data redistribution period, quality of the result data, a normalized progress rate for outputting the result data, and difficulty of processing the source data including,
The weight is determined based on the type of the project,
The evaluation time is
is determined as the sum of the data redistribution periods with the period,
The period of the data redistribution period is,
It is determined as the maximum value from the time when the source data is equally distributed to the plurality of task groups, during which time each of the plurality of task groups processes the source data of a predetermined ratio;
determining evaluation scores of the plurality of task groups for each data redistribution period from the project start time; and
Redistributing the remaining source data based on the ratio of the evaluation score of a specific task group to the sum of the evaluation scores of each of the plurality of task groups for each data redistribution period;
The step of redistributing the remaining source data is repeatedly performed until the sum of the evaluation scores corresponding to each of the plurality of task groups becomes a maximum value,
When the number of the plurality of tasks performing the project is N, the number of tasks included in each of the plurality of task groups is determined to be one of 1 to N!
delete delete delete According to claim 1,
The evaluation time is
It is determined as the time when the sum of the evaluation scores corresponding to each of the plurality of task groups becomes the maximum value,
How to manage project tasks
According to claim 1,
The cost consumed is
A project task management method determined as the sum of the total points paid for outputting the result data for the number of result data determined based on the distributed source data.
According to claim 1,
The time consumed per result data in the data redistribution period is,
Project task management method determined by the sum of the work time and inspection time of the result data with respect to the number of result data output during the data redistribution period
According to claim 1,
The quality of the result data is,
Project task management method determined by the sum of the number of rejections of the result data to the number of result data determined based on the distributed source data
According to claim 1,
The normalized progress rate of outputting the result data is,
determined based on the number of result data output from the source data for a time each of the plurality of task groups actually processed the source data during the plurality of data redistribution periods;
How to manage project tasks
According to claim 1,
The difficulty of processing the source data is,
Project task management method determined by the number of workers corresponding to the number of workers participating in each of the plurality of task groups below a predetermined level

Images