CN113220987A - Knowledge crowdsourcing platform construction method based on incentive recommendation - Google Patents
Knowledge crowdsourcing platform construction method based on incentive recommendation Download PDFInfo
- Publication number
- CN113220987A CN113220987A CN202110452290.3A CN202110452290A CN113220987A CN 113220987 A CN113220987 A CN 113220987A CN 202110452290 A CN202110452290 A CN 202110452290A CN 113220987 A CN113220987 A CN 113220987A
- Authority
- CN
- China
- Prior art keywords
- task completion
- user
- recommendation
- mode
- different
- 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.)
- Granted
Links
- 238000010276 construction Methods 0.000 title claims abstract description 11
- 238000002372 labelling Methods 0.000 claims abstract description 24
- 238000012360 testing method Methods 0.000 claims abstract description 17
- 230000005284 excitation Effects 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 5
- 238000007477 logistic regression Methods 0.000 claims description 3
- 230000000638 stimulation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012358 sourcing Methods 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a knowledge crowdsourcing platform construction method based on incentive recommendation, which comprises the following steps of: constructing a labeling task completion mode under different excitation modes; constructing and training a labeling task completion mode recommendation model LRC based on different incentives; according to the user great five character grid test value, calculating recommendation probabilities of different labeling task completion modes by using the model LRC, and recommending the different labeling task completion modes according to probability priority; and distributing and guiding the user to complete the task according to the labeling task completion mode selected by the user, acquiring a user task completion result, and calculating user rewards in the corresponding task completion mode. The invention meets the external and internal requirements of the user for completing the task by providing and recommending different task completion modes, reduces the money cost of the task publisher, and improves the quality of crowdsourcing task completion.
Description
Technical Field
The invention relates to the computer technology, in particular to the field of artificial hybrid intelligence, and particularly relates to a knowledge crowdsourcing platform construction method based on incentive recommendation.
Background
The current deep learning model requires mass data for training, and a large number of labeling tasks are generated. Unlike expert labeling, task labeling via crowd-sourcing is a lower cost option. However, the pursuit of autonomy, individuality and diversity by user groups with different characters varies, with different task participation incentives, and an optimal task rewarding scheme should meet the motivational needs of different users.
According to the analysis of the existing crowdsourcing platform, the current mainstream mode of stimulating users to participate and complete tasks is money stimulation, the diversified stimulation requirements of users with different characters can not be met, and the problems that the participation enthusiasm of the users is low, the task completion quality is good and uneven, and task publishers need to bear high money cost exist.
Disclosure of Invention
The invention aims to provide a knowledge crowdsourcing platform construction method based on incentive recommendation.
The specific technical scheme for realizing the purpose of the invention is as follows:
a knowledge crowdsourcing platform construction method based on incentive recommendation is characterized by comprising the following specific steps:
step 1: constructing a labeling task completion mode under different excitation modes, which specifically comprises the following steps: a personal challenge mode taking self achievement as main money as auxiliary incentive, a friend PK mode taking social competition as main money as auxiliary incentive and a common mode taking money as incentive;
step 2: constructing and training a labeling task completion mode recommendation model LRC based on different incentives;
and step 3: according to the user great five character grid test value, calculating recommendation probabilities of different labeling task completion modes by using the model LRC, and recommending the different labeling task completion modes according to probability priority;
and 4, step 4: and distributing and guiding the user to complete the task according to the labeling task completion mode selected by the user, acquiring a user task completion result, and calculating user rewards in the corresponding task completion mode.
The step 2 specifically comprises the following steps:
constructing an marked task completion mode recommendation model LRC based on different excitations by using a logistic regression method, collecting preference values of different users for different task completion modes and large quintet test values of the different users, constructing a data set by using the preference values and the large quintet test values, and training the model LRC by using the data set; the formula of the model LRC is:
wherein, YkThe recommendation probability of the kth task completion mode, x is the test value of the user large five characters, wiOr wkAnd b is an offset term, and N is the number of task completion modes.
The step 3 specifically comprises the following steps:
and (3) calculating the recommendation probability of different task completion modes by using the model LRC in the step 2 according to the user large five character test value, and forming a recommendation sequence according to the sequence of the probability from large to small in the different task completion modes and displaying the recommendation sequence in an interface.
Step 4, calculating the user reward in the corresponding task completion mode specifically comprises the following steps:
in the tasks distributed to the user in each task completion mode, part of the tasks have standard results and are used as a basis for measuring the quality of the tasks completed by the user. Each task completion mode provides a monetary reward when the minimum completion quality requirement is met, wherein the personal challenge mode gives the user a fixed basic monetary reward, and the winner of the friend PK mode and the monetary reward in the general mode are calculated according to the following formula:
Total=Base+Bonus (3)
wherein Base is basic wage, AuFor user task completion accuracy, R is money award total, AaveBonus is the amount of the reward, and Total is the monetary reward ultimately received by the user, for the average accuracy with which all users complete the task.
The invention has the beneficial effects that:
the method provides task completion modes with different incentive modes, can meet the requirements of users on different incentives, improves the participation enthusiasm of the users, and solves the problem of overhigh cost under single monetary incentive; according to the personality difference of the individual users, the recommendation sequence is adjusted, the users are further guided to select a mode suitable for the own incentive demands, and the quality of crowdsourcing task completion is improved.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following specific examples and the accompanying drawings. The procedures, conditions, experimental methods and the like for carrying out the present invention are general knowledge and common general knowledge in the art except for the contents specifically mentioned below, and the present invention is not particularly limited.
Examples
Referring to fig. 1, fig. 1 is a flowchart of a knowledge crowdsourcing platform construction method based on incentive recommendation according to an embodiment of the present invention, including the following steps:
step 1: constructing a labeling task completion mode under different excitation modes;
a personal challenge model with satisfaction of self-achievement as a primary incentive, a friend PK model with social competition as a primary incentive, and a general model with money as a primary incentive are constructed.
The personal challenge mode is that a single person independently completes a task, the task difficulty is gradually increased according to the user capacity, and the incentive requirement of the user for self achievement is mainly met; the friend PK mode is a mode that two persons compete to complete tasks, the task difficulty is simple, medium and difficult, and the social competition incentive requirements of users are mainly met; the common mode is that a single person independently completes a task, the task difficulty is random, and the monetary incentive requirements of users are mainly met.
Step 2: constructing and training a labeling task completion mode recommendation model LRC based on different incentives;
constructing an annotated task completion mode recommendation model LRC based on different excitations by using a logistic regression method, wherein the formula of the model LRC is as follows:
wherein, YkThe recommendation probability of the kth task completion mode, x is the test value of the user large five characters, wiOr wkAnd b is an offset term, and N is the number of task completion modes.
Collecting preference values of a user for different task completion modes and large five character test values of different users, numbering the different task completion modes from 0, selecting the task completion mode with the largest preference value from the preference values of the user for the different task completion modes, wherein the number of the task completion mode and the large five character test value of the user form a preference record of the user, the preference records of all the users form a data set, and the data set is used for training the LRC (line road controller) model.
And step 3: according to the user great five character grid test value, calculating the recommendation probability of different labeling task completion modes by using the LRC, and recommending the different labeling task completion modes according to probability priority;
and (3) testing the user large five-character lattice to obtain a 5-dimensional user character lattice test value vector, inputting the vector into the model LRC trained in the step (2) to obtain the recommendation probabilities of different task completion modes, arranging the different task completion modes according to the priority order from large to small according to the probabilities, and displaying the different task completion modes in an interface.
And 4, step 4: and allocating tasks according to the labeling task completion mode selected by the user, acquiring the task completion result of the user, and calculating the user reward in the corresponding task completion mode.
And distributing n tasks according to the labeling task completion mode selected by the user, wherein the m tasks have standard results, and if the accuracy rate of the user for completing the m tasks is more than 50%, giving monetary reward to the user. Wherein, the personal challenge mode gives the user a fixed basic money reward, and the winner of the friend PK mode and the money reward under the common mode are calculated according to the following formula:
Total=Base+Bonus (6)
wherein Base is basic wage, AuFor user task completion accuracy, R is money award total, AaveBonus is the amount of the reward, and Total is the monetary reward ultimately received by the user, for the average accuracy with which all users complete the task.
Claims (4)
1. A knowledge crowdsourcing platform construction method based on incentive recommendation is characterized by comprising the following specific steps:
step 1: constructing a labeling task completion mode under different excitation modes, which specifically comprises the following steps: a personal challenge mode taking self achievement as main money as auxiliary incentive, a friend PK mode taking social competition as main money as auxiliary incentive and a common mode taking money as incentive;
step 2: constructing and training a labeling task completion mode recommendation model LRC based on different incentives;
and step 3: according to the user great five character grid test value, calculating recommendation probabilities of different labeling task completion modes by using the model LRC, and recommending the different labeling task completion modes according to probability priority;
and 4, step 4: and distributing and guiding the user to complete the task according to the labeling task completion mode selected by the user, acquiring a user task completion result, and calculating user rewards in the corresponding task completion mode.
2. The knowledge crowdsourcing platform construction method based on incentive recommendation according to claim 1, wherein the step 2 is specifically:
constructing an marked task completion mode recommendation model LRC based on different excitations by using a logistic regression method, collecting preference values of different users for different task completion modes and large quintet test values of the different users, constructing a data set by using the preference values and the large quintet test values, and training the model LRC by using the data set; the formula of the model LRC is:
wherein, YkThe recommendation probability of the kth task completion mode, x is the test value of the user large five characters, wiOr wkAnd b is an offset term, and N is the number of task completion modes.
3. The knowledge crowdsourcing platform construction method based on incentive recommendation according to claim 1, wherein the step 3 is specifically:
and (3) calculating the recommendation probability of different task completion modes by using the model LRC in the step 2 according to the user large five character test value, and forming a recommendation sequence according to the sequence of the probability from large to small in the different task completion modes and displaying the recommendation sequence in an interface.
4. The knowledge crowdsourcing platform construction method based on incentive recommendation according to claim 1, wherein the calculating of the user reward in the corresponding task completion mode in step 4 is specifically:
in the tasks distributed to the user in each task completion mode, part of the tasks have standard results and are used as a basis for measuring the quality of the tasks completed by the user; each task completion mode provides a monetary reward when the minimum completion quality requirement is met, wherein the personal challenge mode gives the user a fixed basic monetary reward, and the winner of the friend PK mode and the monetary reward in the general mode are calculated according to the following formula:
Total=Base+Bonus (3)
wherein Base is basic wage, AuFor user task completion accuracy, R is money award total, AaveBonus is the amount of the reward, and Total is the monetary reward ultimately received by the user, for the average accuracy with which all users complete the task.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110452290.3A CN113220987B (en) | 2021-04-26 | 2021-04-26 | Knowledge crowdsourcing platform construction method based on incentive recommendation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110452290.3A CN113220987B (en) | 2021-04-26 | 2021-04-26 | Knowledge crowdsourcing platform construction method based on incentive recommendation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113220987A true CN113220987A (en) | 2021-08-06 |
CN113220987B CN113220987B (en) | 2022-07-26 |
Family
ID=77088993
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110452290.3A Active CN113220987B (en) | 2021-04-26 | 2021-04-26 | Knowledge crowdsourcing platform construction method based on incentive recommendation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113220987B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110245980A (en) * | 2019-05-29 | 2019-09-17 | 阿里巴巴集团控股有限公司 | The method and apparatus for determining target user's exiting form based on neural network model |
CN110647678A (en) * | 2019-09-02 | 2020-01-03 | 杭州数理大数据技术有限公司 | Recommendation method based on user character label |
US10839446B1 (en) * | 2019-08-20 | 2020-11-17 | Capital One Services, Llc | Systems and methods for recommending personalized rewards based on customer profiles and customer preferences |
-
2021
- 2021-04-26 CN CN202110452290.3A patent/CN113220987B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110245980A (en) * | 2019-05-29 | 2019-09-17 | 阿里巴巴集团控股有限公司 | The method and apparatus for determining target user's exiting form based on neural network model |
US10839446B1 (en) * | 2019-08-20 | 2020-11-17 | Capital One Services, Llc | Systems and methods for recommending personalized rewards based on customer profiles and customer preferences |
CN110647678A (en) * | 2019-09-02 | 2020-01-03 | 杭州数理大数据技术有限公司 | Recommendation method based on user character label |
Non-Patent Citations (1)
Title |
---|
刘媛妮等: "移动群智感知激励机制研究综述", 《重庆邮电大学学报(自然科学版)》 * |
Also Published As
Publication number | Publication date |
---|---|
CN113220987B (en) | 2022-07-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ponce et al. | Tailored gamification and serious game framework based on fuzzy logic for saving energy in connected thermostats | |
Kanfer | Motivation theory and industrial and organizational psychology | |
US7308418B2 (en) | Determining design preferences of a group | |
Jansen et al. | Development of decision making: Sequential versus integrative rules | |
US7730002B2 (en) | Method for iterative design of products | |
CN109009171A (en) | Attention assessment method, system and computer readable storage medium | |
Hwang | A test-sheet-generating algorithm for multiple assessment requirements | |
US11380423B2 (en) | Computing system implementing a health service for correlating health knowledge and activity data with predictive health outcomes | |
CN113808698B (en) | Computerized social adaptation training method and system | |
CA2967057C (en) | System and method for providing a health determination service based on user knowledge and activity | |
CN115565640A (en) | Personalized neural regulation and control cognitive training system based on game behavior evaluation | |
CN109636484B (en) | Motivation method of crowd sensing system based on behavior economics preference theory | |
CN107301611A (en) | A kind of autism child intelligence Teaching Evaluation System | |
CN111861372B (en) | Method and system for testing word mastering degree | |
Nabizadeh et al. | How do students behave in a gamified course?—A ten-year study | |
CN113220987B (en) | Knowledge crowdsourcing platform construction method based on incentive recommendation | |
Hou et al. | Optimal two-stage elimination contests for crowdsourcing | |
Wilson et al. | The role of working hours, work environment and physical leisure activity on the need for recovery following a day's work among UK white-water raft guides: a within-subjects multilevel approach | |
CN116663652A (en) | Collaborative learning intervention method and system under reinforcement learning support | |
CN114943278B (en) | Continuous online group incentive method and device based on reinforcement learning and storage medium | |
CN108108993A (en) | Ideal money optimization method based on deep neural network | |
Goplerud et al. | Weathering the cuts: A Delphi survey on surviving cutbacks in community mental health | |
Reeves et al. | Dynamic Business Strategy: Competing in a Fast-changing, Uncertain Context | |
CN103021211A (en) | Support function recommending apparatus, a support function recommending method, and a support function recommending system | |
Groth | Comparison as Reflective and Affective Practice: Orientations Towards the Middle in Recreational Road Cycling |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |