CN105930352A - Crowdsourcing task oriented exploratory search method - Google Patents
Crowdsourcing task oriented exploratory search method Download PDFInfo
- Publication number
- CN105930352A CN105930352A CN201610216307.4A CN201610216307A CN105930352A CN 105930352 A CN105930352 A CN 105930352A CN 201610216307 A CN201610216307 A CN 201610216307A CN 105930352 A CN105930352 A CN 105930352A
- Authority
- CN
- China
- Prior art keywords
- task
- service provider
- search
- demand
- user
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a crowdsourcing task oriented exploratory search method. The method includes the following steps: performing pretreatment on statements of a service provider search task, and then serving the pretreated statements as input data of the invention; extracting information of tasks which are completed by the service provider; extracting characteristic words of a historical completed task through a LDA model-based subject word extraction method, and performing pretreatment; performing similarity calculation on pretreated search statements and a task to be completed through a cosine function, and showing the task to the provider; recording information that the service provider searches the task; prompting a user to perform sequencing according to the number of tender people; prompting the user to perform sequencing according to the release time of the tasks; and closing the exploratory search after the service provider finds the task meeting the demands of the service provider. The method overcomes the defects that the conventional task search function is weak, and is limited by individual interest; the method help the user search a task fitting the service provider from a service provider search task perspective, and integrates search habits of the service provides.
Description
Technical field
The present invention relates to search technique field, particularly to a kind of exploratory searching method towards mass-rent task.
Background technology
In the mass-rent platform of current main flow, there is the task of substantial amounts of different demand to wait and be done, and every day also can
Issue new task.Service provider needs from these substantial amounts of tasks, finds and is suitable for oneself task, when needing to waste a lot
Between browse mission requirements.Mass-rent platform mainly provides the method that facet is searched at present.Facet search refers to by the genus to task
Property (such as classification trade mode, money reward, required time etc.) constantly screening, method of filtered search result.Searched for by facet, will
The narrowed number of task is in certain scope, and then service provider the most therefrom selects task.But in actual task is searched for, due to
Technical ability that service provider is good at and the difference of preference, the result of search mission is probably identical.The working time of service provider is original
The most limited, the most also to spend the so many time looking for task, then the time doing task will tail off, the efficiency of service provider
The lowest, obtained remuneration can not maximize, and the efficiency of platform also will not be the highest.
Before the present invention makes, the task search function of existing mass-rent platform is the most weak, it is common that workman is by clear
Looking at task list to search task interested, respective platform provides the function by keyword search task, such as Mturk,
Workman can input task key word interested to search for inter-related task, but does not accounts for the individual preference of different workman,
The result obtained can not fully meet the individual demand of workman.Owing to everyone has oneself different preference, it is therefore desirable to
Carry out personalized search.If service provider is search mission when, automatic fitration is fallen that service provider is bad at and loses interest in
Task, it is possible to save the substantial amounts of time.So far, also do not account for service provider's historical information, be not bound with service yet
The search custom of business oneself.The exploratory search technique of the present invention is producing in this context.When user goes to search
The when of rope task, result is that user is good at mostly, facilitates user to find wanting of task.And search in user's exploration
When, obtain the custom of user's search.Make in the displaying of Search Results, the place that prominent user pays close attention to most.
Summary of the invention
The purpose of the present invention is that and overcomes drawbacks described above, develops a kind of exploratory searching method towards mass-rent task.
The technical scheme is that
A kind of exploratory searching method towards mass-rent task, it is mainly characterized by following steps:
(1) statement of service provider's search mission is expanded vocabulary quantity, Chinese point through gerund synonym, near synonym inquiry
As the input data of the present invention after word pretreatment；
(2) extract the mission bit stream that service provider is complete, analyze technology, the demand obtaining service provider and being good at, and will clothes
Business business's history completes task as document, uses key words extraction method based on LDA model to extract history and completes task
Feature Words, and task to be done is carried out as desired pretreatment；
(3) with cosine function, pretreated search statement and task to be done are carried out Similarity Measure, preferentially will
Technology that service provider is good at and the task of demand and the high task of historic task feature present to service provider before coming；
(4) record service provider checks the information of task, the search feature of Analysis Service business, i.e. checking of task when service provider
Money reward bigger time, prompting user be ranked up according to the order of money reward size；Number of bidders when the task that service provider is checked
Time less, prompting user according to number of bidders number be ranked up；When service provider browses the task of up-to-date issue, prompting is used
Family is ranked up according to the issuing time of task；
(5) if the search through step (4), service provider have found meeting oneself and requiring of task, then this exploration
Search terminates；If service provider is again without finding meeting oneself and requiring of task, prompting service provider describes the need of oneself in detail
Ask, the demand of completing before service provider of task is recommended service provider's reference simultaneously；And repeat step (3) to step (4), directly
Meeting oneself and requiring of task is found to service provider.
Advantages of the present invention and effect are the angle from service provider's search mission, help user to search out applicable service provider
Task.The method can not only allow service provider quickly find oneself task applicable, and combines searching of service provider self
Rope is accustomed to, and searching out of task also more conforms to service provider.
Mainly there is a following advantage:
(1) task of mass-rent platform is searched at present, have employed facet search.After using the method, even if decreasing search
The number of result, but owing to the radix of task is too big, service provider needs for taking much time and searches number.We
Method combines service provider's history and completes task situation, it is contemplated that human factors, and the display of result is more conformed to service
The demand of business.
(2) this method uses exploratory search technique, can accurately obtain custom and the feature of searchers so that search knot
The number of fruit reduces, and improves the work efficiency of service provider.
Accompanying drawing explanation
Fig. 1 schematic flow sheet of the present invention.
Fig. 2 present invention is according to the exploratory search routine schematic diagram of service provider's feature.
Fig. 3 present invention relates to the interface schematic diagram of service provider's search mission.
Detailed description of the invention
The technical thought of the present invention is:
Before service provider is searched for, extract service provider's history and complete task, obtain field and technology that service provider is good at.When with
The when of the removal search task of family, result is that user is presented in the sequence of field and the technology being good at service provider, and user can be accurate
The major requirement of true task of learning, facilitates user to find wanting of task.And user's exploratory search when, obtain
The custom of user's search.Make in the displaying of Search Results, the place that prominent user most frequently pays close attention to.
The present invention is specifically described below.
As shown in Figure 1:
Step (1). service provider's search statement is expanded vocabulary quantity, Chinese point through gerund synonym, near synonym inquiry
As the input data of the present invention after the pretreatment such as word, such as service provider inputs searches element statement: " JSP Website development ", pretreatment
Afterwards " JSP Website development website making enterprise web site ecommerce " task to be done is carried out according to the demand of following form
Classification pretreatment；
Step (2). extract task that service provider history completes (technology that this required by task is wanted and demand), obtain service provider
The technology be good at, demand.Such as this service provider once completed the website of certain dental hospital, the specific requirement of this website: " use
Java completes the Core Feature of website: company information displaying, doctor's introduction, company introduction, news dynamically, contact method, reservation
Take number, on-line consulting (QQ), various project column ", can obtain, from this task, technology (java), the demand that service provider is good at
(company information, company introduction, news are dynamic).The history of service provider is completed mission bit stream and makes document, then use based on
The key phrases extraction method of LDA model, obtains the historic task feature (company information, company introduction, news are dynamic) of service provider
Step (3). with cosine function, pretreated search statement and task to be done are carried out Similarity Measure, will
Result presents to user.Before preferential technology service provider being good at and the task of demand and the high task of historic task feature come
(sorted by the size of priority) in face.The Search Results of the most different service providers is different, comes task above all
Task is completed similar so that the scope of service provider's search greatly reduces with service provider history；
In upper table, the computational methods of cosine function are:
Step (4). the information of task is checked by record service provider, search feature (the exploration search routine of Analysis Service business
As shown in Figure 2) (when the money reward of the task that service provider is checked is bigger, prompting user is ranked up according to the order of money reward size；
When the number of bidders of the task that service provider is checked is less, prompting user according to number of bidders number be ranked up；Work as service
When business browses the task of up-to-date issue, prompting user is ranked up according to the issuing time of task) (prompting circle of exploration search
Face such as Fig. 3) such Search Results service provider especially is ready the result seen, the content of search is accurate, meets again searching of oneself
Suo Tedian so that search time is greatly reduced；
Step (5) if. through the search of step (4), service provider have found meet oneself require task, then this visit
Search terminates without hesitation；If service provider is again without finding meeting oneself and requiring of task, prompting service provider describes oneself in detail
Demand, the demand of completing before service provider of task is recommended service provider's reference simultaneously, so so that search statement
More accurate, the scope simultaneously scanned for is the least, it is thus achieved that the time of oneself wanting of task is the fastest；And repeat step (3)-> step
(4), until service provider finds meeting oneself and requiring of task.
Claims (1)
1. the exploratory searching method towards mass-rent task, it is characterised in that following steps:
(1) by the statement of service provider's search mission through gerund synonym, that vocabulary quantity, Chinese word segmentation are expanded near synonym inquiry is pre-
As the input data of the present invention after process；
(2) extract the mission bit stream that service provider is complete, analyze technology, the demand obtaining service provider and being good at, and by service provider
History completes task as document, uses key words extraction method based on LDA model to extract history and completes the feature of task
Word, and task to be done is carried out as desired pretreatment；
(3) with cosine function, pretreated search statement and task to be done are carried out Similarity Measure, preferentially will service
Technology that business is good at and the task of demand and the high task of historic task feature present to service provider before coming；
(4) record service provider checks the information of task, the search feature of Analysis Service business, the reward of i.e. checking of task when service provider
When gold is bigger, prompting user is ranked up according to the order of money reward size；The number of bidders of checking when service provider of task is less
Time, prompting user according to number of bidders number be ranked up；When service provider browses the task of up-to-date issue, prompting user presses
Issuing time according to task is ranked up；
(5) if the search through step (4), service provider have found meeting oneself and requiring of task, then this exploration search
Terminate；If service provider is again without finding meeting oneself and requiring of task, prompting service provider describes the demand of oneself in detail, with
Time the demand of completing before service provider of task is recommended service provider's reference；And repeat step (3) to step (4), until clothes
Business business finds meeting oneself and requiring of task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610216307.4A CN105930352A (en) | 2016-04-05 | 2016-04-05 | Crowdsourcing task oriented exploratory search method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610216307.4A CN105930352A (en) | 2016-04-05 | 2016-04-05 | Crowdsourcing task oriented exploratory search method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105930352A true CN105930352A (en) | 2016-09-07 |
Family
ID=56840216
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610216307.4A Pending CN105930352A (en) | 2016-04-05 | 2016-04-05 | Crowdsourcing task oriented exploratory search method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105930352A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109710736A (en) * | 2018-12-19 | 2019-05-03 | 浙江大学 | A kind of active crowdsourcing task creating method towards searching order |
| CN110443502A (en) * | 2019-08-06 | 2019-11-12 | 合肥工业大学 | Crowdsourcing task recommendation method and system based on worker's capability comparison |
| CN111666207A (en) * | 2020-05-18 | 2020-09-15 | 中国科学院软件研究所 | Crowdsourcing test task selection method and electronic device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136352A (en) * | 2013-02-27 | 2013-06-05 | 华中师范大学 | Full-text retrieval system based on two-level semantic analysis |
| US20140188838A1 (en) * | 2012-12-28 | 2014-07-03 | Eduard Mikhailovich Strugov | Information search engine, processing and rating system |
| CN105117398A (en) * | 2015-06-25 | 2015-12-02 | 扬州大学 | Software development problem automatic answering method based on crowdsourcing |
-
2016
- 2016-04-05 CN CN201610216307.4A patent/CN105930352A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140188838A1 (en) * | 2012-12-28 | 2014-07-03 | Eduard Mikhailovich Strugov | Information search engine, processing and rating system |
| CN103136352A (en) * | 2013-02-27 | 2013-06-05 | 华中师范大学 | Full-text retrieval system based on two-level semantic analysis |
| CN105117398A (en) * | 2015-06-25 | 2015-12-02 | 扬州大学 | Software development problem automatic answering method based on crowdsourcing |
Non-Patent Citations (2)
| Title |
|---|
| EMAN ALDHAHRI等: "Towards an Effective Crowd sourcing Recommendation System：A Survey of the State-of-the-Art", 《IEEE》 * |
| MAN-CHING YUEN等: "Task Matching in Crowdsourcing", 《IEEE INTERNATIONAL CONFERENCES ON INTERNET OF THINGS, AND CYBER, PHYSICAL AND SOCIAL COMPUTING》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109710736A (en) * | 2018-12-19 | 2019-05-03 | 浙江大学 | A kind of active crowdsourcing task creating method towards searching order |
| CN110443502A (en) * | 2019-08-06 | 2019-11-12 | 合肥工业大学 | Crowdsourcing task recommendation method and system based on worker's capability comparison |
| CN111666207A (en) * | 2020-05-18 | 2020-09-15 | 中国科学院软件研究所 | Crowdsourcing test task selection method and electronic device |
| CN111666207B (en) * | 2020-05-18 | 2021-06-15 | 中国科学院软件研究所 | Crowdsourcing test task selection method and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108829858B (en) | Data query method and device and computer readable storage medium | |
| CN101398834B (en) | Processing method and device for input information and input method system | |
| US20110153654A1 (en) | Natural language-based tour destination recommendation apparatus and method | |
| US8874604B2 (en) | Method and system for searching an electronic map | |
| Ogrodniczuk | Polish parliamentary corpus | |
| WO2018000557A1 (en) | Search results display method and apparatus | |
| TWI594137B (en) | System and method for online handwriting recognition in web queries and computer-readable medium comprising instructions implementing the same | |
| KR102237274B1 (en) | The system and algorithm to provide with the matching information among users by using keyword's significance | |
| US9075898B1 (en) | Generating and ranking incremental search suggestions for personal content | |
| EP2425353A1 (en) | Method and apparatus for identifying synonyms and using synonyms to search | |
| US8359306B2 (en) | Intelligent automatic recognition toolbar search method and system | |
| CN101639857A (en) | Method, device and system for establishing knowledge questioning and answering sharing platform | |
| CN112182148B (en) | Standard aided writing method based on full text retrieval | |
| CN104503998A (en) | Type identifying method and device aiming at query sentence of user | |
| CN108021715B (en) | Heterogeneous label fusion system based on semantic structure feature analysis | |
| CN106599215A (en) | Question generation method and question generation system based on deep learning | |
| US9256805B2 (en) | Method and system of identifying an entity from a digital image of a physical text | |
| CN105930352A (en) | Crowdsourcing task oriented exploratory search method | |
| CN108153754B (en) | Data processing method and device | |
| US10430473B2 (en) | Deep mining of network resource references | |
| KR20070082075A (en) | Method and apparatus for serving search result using template based on query and contents clustering | |
| CN101923560A (en) | Sorting technique of deep web database only providing simple query interface | |
| CN104899310A (en) | Information ranking method, and method and device for generating information ranking model | |
| CN102194000A (en) | Method, device and terminal for processing information | |
| CN113435213B (en) | Method and device for returning answers to user questions and knowledge base |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160907 |