CN110852589A - Crowdsourcing task matching method based on capability evaluation - Google Patents

Crowdsourcing task matching method based on capability evaluation Download PDF

Info

Publication number
CN110852589A
CN110852589A CN201911058173.8A CN201911058173A CN110852589A CN 110852589 A CN110852589 A CN 110852589A CN 201911058173 A CN201911058173 A CN 201911058173A CN 110852589 A CN110852589 A CN 110852589A
Authority
CN
China
Prior art keywords
task
user
difficulty
crowdsourcing
technical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911058173.8A
Other languages
Chinese (zh)
Inventor
唐剑
唐艺
张翔
易比一
郝锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National Defense Technology Innovation Institute PLA Academy of Military Science
Original Assignee
National Defense Technology Innovation Institute PLA Academy of Military Science
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National Defense Technology Innovation Institute PLA Academy of Military Science filed Critical National Defense Technology Innovation Institute PLA Academy of Military Science
Priority to CN201911058173.8A priority Critical patent/CN110852589A/en
Publication of CN110852589A publication Critical patent/CN110852589A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G06Q10/063112—Skill-based matching of a person or a group to a task
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G06Q10/063118—Staff planning in a project environment

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a crowdsourcing task matching method based on capability assessment. According to the invention, the personal data of a registered user of a crowdsourcing task platform and the crowdsourcing task data which is participated and completed by the user in the past are used for dynamically analyzing the capability characteristics of the user, establishing a two-dimensional capability model of 'skill' and 'contribution degree', establishing a demand capability model of the task according to the crowdsourcing task demand, calculating the matching degree of the capability model and the demand characteristic model through an analytic hierarchy process, and sequencing the matched tasks in a user recommendation task list according to the activity degree of the technical field of the user.

Description

Crowdsourcing task matching method based on capability evaluation
Technical Field
The invention relates to the technical field of internet, in particular to a crowdsourcing task matching method based on capability assessment.
Background
The crowdsourcing task is also called as a crowd-sourced task, is a novel task solution under an internet mode, and outsources the work task which is supposed to be completed by the crowdsourcing task to unspecified social masses in a free and voluntary mode, so that the task is efficiently completed by the intelligence of the social masses. The task is usually undertaken by individuals, but as the number of tasks to be solved on a task issuing platform is increased, task responders are often buried in massive task demand information and cannot find tasks really suitable for being completed by themselves, and meanwhile, the task cannot be found by a larger range of people, so that the result quality of the collected task is not optimal.
Several main popular crowdsourcing platforms at present provide crowdsourcing task matching mechanisms, which can be summarized into the following 2 classes: the platform is an Innociceptive scientific research crowdsourcing platform which is known as the first open innovation platform in the world, and the platform is mainly recommended by disciplinary categories and task time; secondly, the famous popular task platform 'the pig eight-ring net' in China mainly adopts task categories and areas to match, and it can be seen that the task matching mechanisms are the active selection of task responders.
The mechanism of active selection by the user can help task responders to find crowdsourcing tasks suitable for the user from massive tasks to a great extent, but a great number of tasks still exist, which cannot be found by the user actively due to inaccurate definition of keywords during release or division into different technical fields, and the completion efficiency of the crowdsourcing tasks is also influenced to a certain extent.
Therefore, how to provide a crowdsourcing task matching method based on the capability evaluation of the accreditant for the platform user is to analyze the user data and the historical data to realize the accurate representation of the capability of the accreditant, so that a proper task can be found from a large number of tasks on the current platform in a more targeted manner and is actively pushed to the task accreditant, and the method is an important problem for improving the operation efficiency of the crowdsourcing task platform.
Disclosure of Invention
In view of the above problems, the present invention is directed to a crowdsourcing task matching method based on capability evaluation, which is provided to overcome the shortcomings of the existing crowdsourcing task matching technology. The invention solves two technical problems, namely, how to depict the capability of a platform user; and secondly, how to realize accurate matching of crowdsourcing tasks based on user capacity portrayal. By solving the two problems, users who are willing to sign for participation in the task can quickly find all crowdsourcing tasks meeting self ability conditions, so that the number of the payers of the crowdsourcing tasks is greatly increased, and the quality level of corresponding results is improved.
In order to achieve the purpose, the invention adopts the following technical scheme:
a crowdsourcing task matching method based on capability assessment comprises a crowdsourcing platform, wherein each crowdsourcing task in the crowdsourcing platform has an appreciation amount and can be obtained by user competition in the crowdsourcing platform, a user capability model database and a crowdsourcing task demand characteristic model database are built in the crowdsourcing platform, the user capability model database stores user capability model data, the user capability model data comprises two dimensions of skill and contribution degree, the skill dimension comprises three indexes of development tool preference, technical field preference and task completion difficulty distribution, and the contribution degree dimension comprises two indexes of user activity and task completion number; the crowdsourcing task demand characteristic model database stores crowdsourcing task demand characteristic model data, the crowdsourcing task demand characteristic model data comprises three indexes of required development tools, subdivision technical field and task difficulty, and the task matching working process comprises the following steps:
1) calling user capacity model data applying for participating in crowdsourcing tasks from a user capacity model database;
2) calling task demand characteristic model data of a crowdsourcing task to be matched from a crowdsourcing task demand characteristic model database;
3) matching the skill dimension in the called user capability model data with a requirement characteristic model of a crowdsourcing task, firstly matching the development tool preference in the skill dimension with a required development tool in the task requirement characteristic model, if the matching requirement is met, entering the next step, and otherwise, restarting the first step to call new task requirement characteristic model data;
4) matching technical field preference in the skill dimension in the user capability model data with subdivision technical fields in the task demand characteristic model, if the matching requirement is met, entering the next step, and otherwise, restarting the first step and calling new task demand characteristic model data;
5) judging according to technical field preference in skill dimension in user capability model data, task difficulty distribution completion, subdivision technical field in a task demand characteristic model and task difficulty, judging whether the task difficulty in the technical field is matched, if the matching requirement is met, marking the task matching and entering the matching of the next task, and if not, restarting the first step and calling new task demand characteristic model data;
6) judging whether the tasks in all the task databases are matched with the user capacity, if so, sequencing all the matched tasks from high to low according to the activity of each technical field in the user capacity model data, and otherwise, restarting the first step to call new task demand characteristic model data;
7) and pushing all the sequenced tasks matched with the user capability model to the user and displaying the tasks.
The requirement characteristic model is preferred, the skill dimension of the user capability model data is an M + N + 4-dimensional vector, wherein M is the number of development tools, and the M-dimensional vector is [ x ]1,x2,…xm,…xM]，xmIs a binary value, the user grasps one of the development tools and the corresponding number value xm1, otherwise 0; n is the number of the technical field, and the N-dimensional vector is [ y1,y2,…yn,…yN]，ynIs a binary numerical value, and the user has one of the technical field knowledge and the corresponding serial number value yn1, otherwise 0; the task difficulty is divided into 4 grades, and 4-dimensional vectors [ z ] are used1,z2,z3,z4]The task participated by the user is one level, the corresponding number value is 1, and the number values of the other three items are 0.
Preferably, the activity in the contribution dimension of the user capability model data is normalized by the participation rate of all tasks issued by the user in the current technical field relative to the mean value of all users, and the NumTaskiFor the useri number of all tasks involved in the technical field, NumTasktotalFor all the tasks the user is involved in the technical field,for the activity of the user in the technical field, I is the number of all participating users in the technical field, and the activity is calculated according to a formula
Figure BDA0002257116440000032
And (6) calculating.
Preferably, a quasi-weighting coefficient is used when judging that the task difficulty matches in the step 5, and the task completion number of the user in the contribution degree dimension of the user capability model data is set to be NumTask Compi，ωiFor the quasi-weighting coefficient, I is the number of all participating users in the technical field, and is calculated by formula
Figure BDA0002257116440000033
And (6) calculating.
Calculating an actual weighting coefficient of the task difficulty of the current user, wherein omega is the actual weighting coefficient and is expressed by a formula
Figure BDA0002257116440000034
And (6) calculating.
Let the task difficulty be UserTaskDiffiDIS and the task difficulty be
Figure BDA0002257116440000035
Wherein z is the reward amount of the task, a is a determined numerical value and is the highest reward amount of the task in the technical field, and the value range of the task difficulty is [1/e,1 ]; the difficulty of each task participated by the user is distributed on the [1/e,1] interval, and the task difficulty level can be obtained by the following formula:
Figure BDA0002257116440000041
wherein b, c and d areThe parameters are determined.
Counting the times of each task participated by the user falling into the corresponding difficulty level, and comparing the times with the total times to obtain the task completion difficulty distribution of the user, wherein the task completion difficulty distribution is UserTaskDiffiDIS [ gamma ]1,Γ2,Γ3,Γ4]
Wherein, gamma is1+Γ2+Γ3+Γ4＝1，Γ1Probability in the difficulty level easy interval, Γ2Probability in the normal interval of difficulty scale, Γ3Probability in the difficulty level hard interval, Γ4Probability in the difficulty class insane interval;
wherein, gamma is1+Γ2+Γ3+Γ4＝1，Γ1Probability in the difficulty level easy interval, Γ2Probability in the normal interval of difficulty scale, Γ3Probability in the difficulty level hard interval, Γ4Probability in the difficulty class insane interval; calculating the difficulty of the current task, judging that the task falls into a certain difficulty level area of the user TaskDffcity, and setting the probability of the interval as gammajJ is any integer from 1 to 4, gamma is set as the probability that the task difficulty of the current user participating in the completion is more than or equal to the current task difficulty,
Figure BDA0002257116440000042
weighting gamma by using an actual difficulty weighting coefficient omega to obtain a weighted value omega gamma of the current user in the current task difficulty, then comparing the weighted value omega gamma with a preset difficulty threshold gamma of the current task to be matched, if omega gamma is larger than gamma, matching the task difficulty, otherwise, mismatching the task difficulty.
Preferably, the skill dimension of the feature model data of the crowdsourcing task demand is an M + N + 4-dimensional vector, wherein M is the number of development tools, and the M-dimensional vector is [ x ]1,x2,…xm,…xM]，xmIs a binary value, the crowdsourcing task needs one of the development tools, and the corresponding number value x is corresponding to the development toolm1, otherwise 0; n is the number of the technical field, and the N-dimensional vector is [ y1,y2,…yn,…yN]，ynIs a binary value, the crowdsourcing task needs one of the technical field knowledge and the corresponding number value yn1, otherwise 0; the task difficulty is divided into 4 grades, and 4-dimensional vectors [ z ] are used1,z2,z3,z4]The crowdsourcing task is represented as one level, the corresponding number value is 1, and the number values of the other three items are 0.
Due to the adoption of the technical scheme, the invention has the following technical advantages: by comprehensively evaluating the capability of each user, tasks within the capability range are accurately matched among a large number of crowd-sourced tasks, so that the completion quality of the crowd-sourced tasks is effectively improved, and the user experience is improved.
Drawings
FIG. 1 is a flow chart of a method incorporating embodiments of the present invention;
FIG. 2 is a task difficulty matching flow chart of the present invention.
Detailed Description
The present invention will be described in detail with reference to the following embodiments and specific examples.
The technical scheme of the embodiment of the invention comprises the following steps:
step 1, dynamically analyzing the user capability characteristics according to personal data of a registered user of a platform and crowdsourcing task data which is participated and completed by the user in the past, and forming a capability model of the user, wherein the capability model of the user comprises two dimensions of 'skill' and 'contribution degree', and 3 primary indexes of 'development tool preference', 'technical field preference', 'task completion difficulty distribution' and the like are shared under the skill dimension; the 'contribution degree' dimension has 2 first-level indexes of 'user activity degree' and 'number of completed tasks'.
1. The user's skill ability is characterized by the following vector
UserCapability＝
[UserDevTools,UserTechDomains,UserTaskDiffiDIS]
UserCapability is an M + N +4 dimensional row vector, where:
UserDevTools＝[x1,x2,…xm,…xM]
the method is characterized in that the method is an M-dimensional row vector, xm is a binary numerical value and represents commonly used M development tools, such as visual studio, Matlab, Eclipse and the like, the commonly used tools are numbered, and when a user grasps a certain development tool, the corresponding number value x is the value of the numbermOtherwise, it is 0.
UserTechdomain＝[y1,y,…yn,…yN]
Is a row vector of dimension N, ynIs a binary numerical value representing N technical fields such as computer, communication, electronics, medicine and the like to which the current task of the platform belongs, numbers the technical fields, and when the user has professional knowledge in a certain field, the corresponding number value y is the number valuenOtherwise, it is 0.
Both M and N above can be increased with the increase of platform service range, and for adjustable parameters, the vectors DevTools and DevTools can be generated and continuously updated by reading user registration information and already participated task information.
The task difficulty is defined as:
Figure BDA0002257116440000061
wherein z is the reward amount of the task, a is a determined numerical value, and is the highest reward amount of the task in the technical field, the value range of the task difficulty is [1/e,1], the task difficulty is about 0.375 at the lowest, namely, the tasks which are recruited and released through crowdsourcing have certain difficulty. The difficulty of each task participated by the user is distributed on the [1/e,1] interval, in order to intuitively know the task difficulty distribution, the task difficulty grades are defined as [ easy, normal, hard, insane ], and 4 grades are totally defined, and the task difficulty of the user can be obtained
Figure BDA0002257116440000062
Wherein b, c and d are adjustable parameters, and are suggested to be 0.45, 0.60 and 0.85.
Counting the times of each task participated by the user falling into the corresponding difficulty level, and comparing the times with the total times to obtain the task completion difficulty distribution of the user, wherein the task completion difficulty distribution is in the form of UserTaskDiffiDIS [ gamma ]easy,Γnornal,Γhard,Γinsane]
Wherein, gamma iseasy+Γnornal+Γhard+Γinsane＝1。
2. The contribution of the user is characterized by the following model
User liveness statistics by technical field, NumTaskiFor all the tasks that the user i participates in the technical field, the user activity is that the participation rate of all tasks issued by the user in the current technical field is normalized relative to the average value of all the users, if the value is larger, the activity is more than 1, namely the activity is active, namely the activity is higher than the average level, and if the value is lower than 1, the activity is inactive.
Note: liveness is a relative value that is a comparison of users with each other.
The task completion number of the user is NumTask CompiThe task completion number of the user represents the capability of the user, the task difficulty of the user can be weighted by the task completion number, if the task completion number of the user in the technical field is higher than the average level, the acceptable task difficulty of the user is weighted and promoted, and if not, the weighting is 1.
The pseudo-weighting coefficients are:
Figure BDA0002257116440000071
and 2, establishing a task demand characteristic model for all the released crowdsourcing tasks on the platform, wherein 3 primary indexes of 'required development tools', 'subdivision technical field', 'task difficulty' and the like are shared.
TaskRequirement＝
[TaskDevTools,TaskTechDomains,TaskTaskDiffi]
Step 3, establishing a user capacity model database, storing all registered user capacity model data, and updating the user capacity model when the user capacity model meets the following conditions: (1) the user has updated the personal profile; (2) the user registers to participate in a crowdsourcing task; (3) the user has completed a crowdsourcing task.
Step 4, establishing a crowdsourcing task demand characteristic model database, storing demand characteristic model data of all the crowdsourcing tasks published by the platform, and updating the task demand characteristic model when the following conditions are met: (1) the demand publisher publishes a new task; (2) the requirement publisher modifies the requirement of the persistent task; (3) the task is completed.
And 5, matching the skill dimension of the capability model with the task requirement characteristic model, and weighting the contribution degree dimension.
Adopting an analytic hierarchy process:
matching preferences of a user's development tool, | UserDevTools |. TaskDevTools |. 1, then performing the next step, technical field matching, otherwise rejecting;
and (2) performing next step by using the technology field matching, | TaskTechDomains | > 1, matching the task difficulty, and otherwise, rejecting the task difficulty.
Task difficulty matching: according to the technical field, the difficulty of the current task is calculated, and the probability of the task falling into the TaskDfficity of the user and setting the interval is determined to be gammajLet us order
Figure BDA0002257116440000072
Calculating a task difficulty quasi-weighting coefficient for the probability that the task difficulty of the current user is more than or equal to the current task difficulty, and if the task difficulty is more than or equal to the current task difficulty, calculating a task difficulty quasi-weighting coefficient
Figure BDA0002257116440000073
And determines ω Γ of the corresponding region>And gamma is a threshold value of task difficulty distribution threshold, the next step is carried out, the task is matched, and otherwise, the task is eliminated.
Matching of the next task is performed.
And calculating the user activity of the user in the technical fields corresponding to all the matched tasks, and sequencing the tasks according to the activity. And if the user participates in the task for the first time, only matching the development tool with the technical field, and if the user information is incomplete, displaying the current hottest task by default.

Claims (5)

1. A crowdsourcing task matching method based on capability assessment comprises a crowdsourcing platform, wherein each crowdsourcing task in the crowdsourcing platform has an appreciation amount and can be obtained by user competition in the crowdsourcing platform, and the crowdsourcing platform is characterized in that a user capability model database and a crowdsourcing task demand characteristic model database are built in the crowdsourcing platform, the user capability model database stores user capability model data, the user capability model data comprises two dimensions of skill and contribution degree, the skill dimension comprises three indexes of development tool preference, technical field preference and task completion difficulty distribution, and the contribution degree dimension comprises two indexes of user liveness and task completion number; the crowdsourcing task demand characteristic model database stores crowdsourcing task demand characteristic model data, the crowdsourcing task demand characteristic model data comprises three indexes of a required development tool, a subdivision technical field and task difficulty, and the task matching working process comprises the following steps:
1) calling user capacity model data applying for participating in crowdsourcing tasks from a user capacity model database;
2) calling task demand characteristic model data of a crowdsourcing task to be matched from a crowdsourcing task demand characteristic model database;
3) matching a skill dimension in the called user capability model data with a requirement characteristic model of a crowdsourcing task, firstly matching development tool preference in the skill dimension with a required development tool in the task requirement characteristic model, entering a next step if matching requirements are met, and otherwise restarting the first step to call new task requirement characteristic model data;
4) matching technical field preference in the skill dimension in the user capability model data with subdivision technical fields in the task demand characteristic model, if the matching requirement is met, entering the next step, and otherwise, restarting the first step and calling new task demand characteristic model data;
5) judging according to technical field preference in skill dimension in user capability model data, completion task difficulty distribution and subdivision technical field and task difficulty in a task demand characteristic model, judging whether the task difficulty in the technical field is matched, if the matching requirement is met, marking the task matching and entering the matching of the next task, and if not, restarting the first step and calling new task demand characteristic model data;
6) judging whether the tasks in all the task databases are matched with the user capacity, if so, sequencing all the matched tasks from high to low according to the activity of each technical field in the user capacity model data, and otherwise, restarting the first step to call new task demand characteristic model data;
7) and recommending all the tasks matched with the user capability model after sequencing to the user.
2. The method of claim 1, wherein the skill dimension of the user capability model data is M + N + 4-dimensional vector, wherein M is the number of development tools and M-dimensional vector is [ x ]1，x2，...xm，...xM]，xmIs a binary value, the user grasps one of the development tools and the corresponding number value xm1, otherwise 0; n is the number of the technical field, and the N-dimensional vector is [ y1，y2，...yn，...yN]，ynIs a binary numerical value, and the user has one of the technical field knowledge and the corresponding serial number value yn1, otherwise 0; the task difficulty is divided into 4 grades, and 4-dimensional vectors [ z ] are used1，z2，z3，z4]The task that the user participates in is one of the levels, the corresponding number value is 1, and the number value isThe number values of the remaining three items are 0.
3. The method of claim 1, wherein the liveness in the contribution dimension of the user capability model data is normalized by the engagement rate of all tasks issued by the user in the current technical field relative to the mean of all users, and NumTask is used for matching the crowd-sourced tasks based on capability assessmentiFor all the tasks of the user i participating in the technical field, NumTasktotalFor all the tasks the user is involved in the technical field,
Figure FDA0002257116430000021
for the activity of the user in the technical field, I is the number of all participating users in the technical field, and the activity is calculated according to a formula
Figure FDA0002257116430000022
And (6) calculating.
4. The method of claim 1, wherein a pseudo-weighting factor is used in determining task difficulty matching in step 5, and the number of task completions of the user in the contribution dimension of the user capability model data is set to NumTask Compi，ωiFor the quasi-weighting coefficient, I is the number of all participating users in the technical field, and is calculated by formula
Figure FDA0002257116430000023
Calculating to obtain;
calculating an actual weighting coefficient of the task difficulty of the current user, wherein omega is the actual weighting coefficient and is expressed by a formula
Figure FDA0002257116430000024
Calculating to obtain;
let the task difficulty be UserTaskDfffidIS and the task difficulty be
Figure FDA0002257116430000025
Wherein z is the reward amount of the task, a is a determined numerical value and is the highest reward amount of the task in the technical field, and the value range of the task difficulty is [1/e,1 ]; the difficulty of each task participated by the user is distributed on the [1/e,1] interval, and the task difficulty level can be obtained by the following formula:
Figure FDA0002257116430000031
wherein b, c and d are determined parameters;
counting the times of each task participated by the user falling into the corresponding difficulty level, and comparing the times with the total times to obtain the task completion difficulty distribution of the user, namely
UserTaskDiffiDIS＝[Γ1，Γ2，Γ3，Γ4]
Wherein, gamma is1+Γ2+Γ3+Γ4＝1，Γ1Probability in the difficulty level easy interval, Γ2Probability in the normal interval of difficulty scale, Γ3Probability in the difficulty level hard interval, Γ4Probability in the difficulty class insane interval; calculating the difficulty of the current task, judging that the task falls into a certain difficulty level area of the user TaskDffcity, and setting the probability of the interval as gammajJ is any integer from 1 to 4, gamma is set as the probability that the task difficulty of the current user participating in the completion is more than or equal to the current task difficulty,
Figure FDA0002257116430000032
weighting gamma by using an actual difficulty weighting coefficient omega to obtain a weighted value omega gamma of the current user in the current task difficulty, then comparing the weighted value omega gamma with a preset difficulty threshold value y of the current task to be matched, if omega gamma is larger than y, matching the task difficulty, otherwise, mismatching the task difficulty.
5. The energy-based of claim 1The force evaluation crowdsourcing task matching method is characterized in that the skill dimension of feature model data required by the crowdsourcing task is an M + N + 4-dimensional vector, wherein M is the number of development tools, and the M-dimensional vector is [ x ]1，x2，...xm，...xM]，xmIs a binary value, the crowdsourcing task needs one of the development tools, and the corresponding number value x is corresponding to the development toolm1, otherwise 0; n is the number of the technical field, and the N-dimensional vector is [ y1，y2，...yn，...yN]，ynIs a binary value, the crowdsourcing task needs one of the technical field knowledge and the corresponding number value yn1, otherwise 0; the task difficulty is divided into 4 grades, and 4-dimensional vectors [ z ] are used1，z2，z3，z4]The crowdsourcing task is represented as one level, the corresponding number value is 1, and the number values of the other three items are 0.
CN201911058173.8A 2019-11-01 2019-11-01 Crowdsourcing task matching method based on capability evaluation Pending CN110852589A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911058173.8A CN110852589A (en) 2019-11-01 2019-11-01 Crowdsourcing task matching method based on capability evaluation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911058173.8A CN110852589A (en) 2019-11-01 2019-11-01 Crowdsourcing task matching method based on capability evaluation

Publications (1)

Publication Number Publication Date
CN110852589A true CN110852589A (en) 2020-02-28

Family

ID=69599191

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911058173.8A Pending CN110852589A (en) 2019-11-01 2019-11-01 Crowdsourcing task matching method based on capability evaluation

Country Status (1)

Country Link
CN (1) CN110852589A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111311129A (en) * 2020-03-31 2020-06-19 广东电网有限责任公司 Automatic work distribution method, system and computer equipment for power distribution network on-duty task allocation
CN111798156A (en) * 2020-07-16 2020-10-20 武汉空心科技有限公司 Task allocation workload evaluation system and method based on working platform
CN113435800A (en) * 2021-08-26 2021-09-24 平安科技（深圳）有限公司 Method and device for executing labeling task based on big data, electronic equipment and medium
CN113487161A (en) * 2021-06-30 2021-10-08 武汉空心科技有限公司 Engineer skill proficiency evaluation system and method based on working platform
CN115409667A (en) * 2022-07-22 2022-11-29 佛山市四方建业建筑工程有限公司 Municipal engineering construction information management method and device and computer equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150302340A1 (en) * 2014-04-18 2015-10-22 Xerox Corporation Methods and systems for recommending crowdsourcing tasks
CN106462818A (en) * 2014-06-09 2017-02-22 微软技术许可有限责任公司 Evaluating workers in crowdsourcing environment
CN107958317A (en) * 2016-10-17 2018-04-24 腾讯科技（深圳）有限公司 A kind of method and apparatus that crowdsourcing participant is chosen in crowdsourcing project
CN108876093A (en) * 2018-04-26 2018-11-23 浙江大学 A kind of many many wound design objective method for pushing created under platform

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150302340A1 (en) * 2014-04-18 2015-10-22 Xerox Corporation Methods and systems for recommending crowdsourcing tasks
CN106462818A (en) * 2014-06-09 2017-02-22 微软技术许可有限责任公司 Evaluating workers in crowdsourcing environment
CN107958317A (en) * 2016-10-17 2018-04-24 腾讯科技（深圳）有限公司 A kind of method and apparatus that crowdsourcing participant is chosen in crowdsourcing project
CN108876093A (en) * 2018-04-26 2018-11-23 浙江大学 A kind of many many wound design objective method for pushing created under platform

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
朱小宁: ""支持任务推送的众包系统的研究与实现"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *
李勇军: ""软件‘众包’任务分配方法"", 《计算机系统应用》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111311129A (en) * 2020-03-31 2020-06-19 广东电网有限责任公司 Automatic work distribution method, system and computer equipment for power distribution network on-duty task allocation
CN111798156A (en) * 2020-07-16 2020-10-20 武汉空心科技有限公司 Task allocation workload evaluation system and method based on working platform
CN113487161A (en) * 2021-06-30 2021-10-08 武汉空心科技有限公司 Engineer skill proficiency evaluation system and method based on working platform
CN113435800A (en) * 2021-08-26 2021-09-24 平安科技（深圳）有限公司 Method and device for executing labeling task based on big data, electronic equipment and medium
CN115409667A (en) * 2022-07-22 2022-11-29 佛山市四方建业建筑工程有限公司 Municipal engineering construction information management method and device and computer equipment

Similar Documents

Publication Publication Date Title
CN110852589A (en) Crowdsourcing task matching method based on capability evaluation
Wang et al. Forecast combinations: An over 50-year review
Chen A comparative analysis of score functions for multiple criteria decision making in intuitionistic fuzzy settings
Khalili-Damghani et al. A hybrid fuzzy multiple criteria group decision making approach for sustainable project selection
Srdjevic et al. Synthesis of individual best local priority vectors in AHP-group decision making
Tsao Fuzzy net present values for capital investments in an uncertain environment
Caron et al. Bayesian nonparametric Plackett–Luce models for the analysis of preferences for college degree programmes
CN109155153A (en) Adaptive motion sexuality prescription system
WO2015103964A1 (en) Method， apparatus， and device for determining target user
US20150032235A1 (en) Systems and methods for automated analysis of fitness data
US8170963B2 (en) Apparatus and method for processing information, recording medium and computer program
WO2020135642A1 (en) Model training method and apparatus employing generative adversarial network
CN112148986B (en) Top-N service re-recommendation method and system based on crowdsourcing
CN106789338B (en) Method for discovering key people in dynamic large-scale social network
CN111061959A (en) Developer characteristic-based crowd-sourcing software task recommendation method
CN109411093A (en) A kind of intelligent medical treatment big data analysis processing method based on cloud computing
Haruvy et al. Modeling and testing for heterogeneity in observed strategic behavior
CN106485069A (en) The method and system of rehabilitation information pushing
CN109544261A (en) A kind of intelligent perception motivational techniques based on diffusion and the quality of data
Basaran et al. A multi-criteria decision making to rank android based mobile applications for mathematics
CN110826900B (en) Crowd-sourcing-based crowd-sourcing contribution review method
JP2004220236A (en) Method and apparatus of data analysis, program, and recording medium with the program recorded thereon
CN112612207A (en) Multi-target game solving method and system under uncertain environment
Chang et al. A Stock-Movement Aware Approach for Discovering Investors' Personalized Preferences in Stock Markets
CN108549979B (en) Open-source software development team extension method based on precise embedded representation

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20200228

RJ01 Rejection of invention patent application after publication