CN108536682A - A kind of identification categorizing system applied to service trade trade matching - Google Patents
A kind of identification categorizing system applied to service trade trade matching Download PDFInfo
- Publication number
- CN108536682A CN108536682A CN201810344637.0A CN201810344637A CN108536682A CN 108536682 A CN108536682 A CN 108536682A CN 201810344637 A CN201810344637 A CN 201810344637A CN 108536682 A CN108536682 A CN 108536682A
- Authority
- CN
- China
- Prior art keywords
- party
- request
- keyword
- industry
- technical ability
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/253—Grammatical analysis; Style critique
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of identification categorizing systems applied to service trade trade matching, wherein demand party subsystem is used to obtain the demand text information of party in request's input, and demand text information is sent to and brings party subsystem together；Service party subsystem is used to obtain the service text information of service side's input, and service text information is sent to and brings party subsystem together；One group for the treatment of progress for brining party subsystem together is used for real-time reception demand text information, and parses the industry and technical ability for obtaining party in request, then the industry and technical ability that parsing obtains are fed back to demand party subsystem；Another group for the treatment of progress services text information for real-time reception, and parses the industry and technical ability for obtaining service side, then the industry and technical ability that parsing obtains are fed back to service party subsystem.The identification categorizing system fast and effeciently can carry out industry to party in request and service side and technical ability is classified, and realizes and mutually recommend, and substantially increases the working efficiency of the side of brining together.
Description
Technical field
It is especially a kind of applied to service trade trade matching the present invention relates to a kind of analysis system applied to service trade
Identify categorizing system.
Background technology
With the deep propulsion that supply side is reformed, the successful match rate of service trade trade matching is improved to promote resource distribution
With the efficiency utilized, it has also become focal issue of people's attention.Traditionally, one section of text is submitted by party in request (or project initiator)
Demand is included into suitable trade classification, and indicates the skill requirement of required service side by description of the word material as demand；Pinch
Conjunction side's (or platform) audits the material of party in request, after making necessary modification, is pushed to suitable potential service side, receives and push away
The people sent registers project if interested in the project, and whether subsequent need side and service side link up again further cooperates；For
Service side is also to seek suitable party in request by identical method.
That there are problems is as follows for above conventional procedures：1, for vast small elementary item mesh, party in request or service side are very
It is unwilling to require efforts when more and the demand of oneself or service is included into suitable trade classification, has for skill requirement or service
When, also summarizes unclear, can not form standardized format；2, for the side of brining together, due to the above problems, must not
Special human resources are not configured and are engaged in the thing that industry is sorted out and technical ability is sorted out, and only carry out this step, it could targetedly
Facilitate the cooperation of both sides in ground.
Invention content
Goal of the invention：Industry and technical ability identification can targetedly be realized by providing one kind, to facilitate party in request's kimonos
The system that business side cooperates.
Technical solution：Identification categorizing system of the present invention applied to service trade trade matching, including demand prescription
System brings party subsystem and service party subsystem together；
Demand party subsystem, the demand text information for obtaining party in request's input, and demand text information is sent to
Bring party subsystem together；
Party subsystem is serviced, the service text information for obtaining service side's input, and service text information is sent to
Bring party subsystem together；
Bring party subsystem, including two groups for the treatment of progress together；One group for the treatment of progress is used for real-time reception demand text information, and
Parsing obtains the industry and technical ability of party in request, then the industry and technical ability that parsing obtains are fed back to demand party subsystem；Another group
Treatment progress services text information for real-time reception, and parses the industry and technical ability for obtaining service side, then parsing is obtained
Industry and technical ability feed back to service party subsystem.
Further, demand party subsystem includes party in request's text conversion module, party in request's recommending module and party in request
Recognition feedback module；Service party subsystem includes service side's text conversion module, service side's recommending module and service side's identification
Feedback module；
Party in request's text conversion module, the input information for receiving party in request, and input information is converted into demand text
Word information, then demand text information is sent to and brings party subsystem together；
Service side's text conversion module, the input information for receiving service side, and input information is converted into service text
Word information, then service text information is sent to and brings party subsystem together；
Party in request's Recognition feedback module, the identification of industry and technical ability for receiving the party in request for brining party subsystem transmission together
As a result, and provide the identification match index of each group recognition result for party in request, and each recognition result is both provided with selection option,
If party in request is not selected, default choice whole recognition result provides pushing away for service side after selecting to confirm for party in request
Recommend request；
Service side's Recognition feedback module, the identification of industry and technical ability for receiving the service side for brining party subsystem transmission together
As a result, and provide the identification match index of each group recognition result for service side, and each recognition result is both provided with selection option,
If service side does not select, default choice whole recognition result provides pushing away for party in request after selecting to confirm for service side
Recommend request；
Party in request's recommending module, for after party in request receives recommendation request, recommending the industry with party in request to party in request
With the corresponding service side of technical ability, and provide the recommendation index of each service side, if party in request recommendation request is not carried out it is any
Response, then give tacit consent to party in request and receive recommendation request；
Service side's recommending module, for after service side receives recommendation request, recommending the industry with service side to service side
With the corresponding party in request of technical ability, and provide the recommendation index of each party in request, if service side recommendation request is not carried out it is any
Response, then default service side receives recommendation request.
Further, bring together party subsystem include syntactic analysis module, Lexical Analysis Module, syntax normal form library, participle library,
Keyword retrieval module, technical ability dictionary, industry dictionary, keyword memory module, industry and technical ability mapping block, industry and technical ability
Identification module and identification correction verification module；
Syntactic analysis module, including two-way treatment progress；Treatment progress for reading party in request's text conversion in real time all the way
The demand text information that module is sent, and demand text information is carried out at subordinate sentence according to the syntax normal form in syntax normal form library
Reason, obtains each demand input subordinate sentence, and text information puts in order and sequentially numbered to demand input subordinate sentence as desired；
Another way treatment progress is used for the service text information that real-time reading service side's text conversion module is sent, and according to syntax normal form
Syntax normal form in library carries out subordinate sentence processing to service text information, obtains each import of services subordinate sentence, and according to service word
Information arrangement sequence sequentially numbers import of services subordinate sentence；
Lexical Analysis Module, including two-way treatment progress；Treatment progress is used to be successively read each need according to number all the way
Input subordinate sentence is sought, and word segmentation processing is carried out to each demand input subordinate sentence according to participle library, each demand is obtained and segments, and according to
Sequencing segments each demand and numbers；Another way treatment progress is used to be successively read each import of services point according to number
Sentence, and word segmentation processing is carried out to each import of services subordinate sentence according to participle library, each service participle is obtained, and according to sequencing
Each service is segmented and is numbered；
Keyword retrieval module, including two-way treatment progress；Treatment progress is each for being successively read according to number all the way
Demand segments, and each demand participle is retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, by technical ability word
In library party in request's skill that corresponding technical ability keyword is stored in keyword memory module is segmented with the demand currently retrieved
In energy keyword queue, the industry keyword corresponding with the demand participle currently retrieved in industry dictionary is stored in key
In party in request's industry keyword queue in word memory module；Another way treatment progress is used to be successively read each clothes according to number
Business participle, each service participle is retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, by technical ability dictionary
In technical ability keyword corresponding with the service participle currently retrieved be stored in service side's technical ability in keyword memory module
In keyword queue, the industry keyword corresponding with the service participle currently retrieved in industry dictionary is stored in keyword
In service side's industry keyword queue in memory module；
Industry and technical ability mapping block, including two-way treatment progress；Treatment progress for reading party in request's skill in real time all the way
Energy keyword queue and party in request's industry keyword queue map each of party in request according to the mapping table of technical ability and industry
A technical ability keyword and industry-by-industry keyword filter out the party in request's all mappings pair that disclosure satisfy that mapping table；It is another
Road treatment progress is used for real-time reading service side's technical ability keyword queue and service side's industry keyword queue, according to technical ability and row
The mapping table of industry, each technical ability keyword and the industry-by-industry keyword of mapping services side, filters out and disclosure satisfy that mapping
The service side of relation table all mappings pair；Mapping an industry keyword and a technical ability keyword to being made of；
Industry and technical ability identification module, including two-way treatment progress；Treatment progress is all reflected for receiving party in request all the way
It penetrates pair, and merges the mapping of mutually keyword of the same trade to for one group of party in request's recognition result, exporting each group after merging treatment needs
The side's of asking recognition result；Another way treatment progress merges reflecting for phase keyword of the same trade for receiving service side's all mappings pair
It penetrates to for one group of service side's recognition result, exporting each group service side's recognition result after merging treatment；The industry when merging
Keyword remains unchanged, and technical ability keyword is added and duplicate removal；
Identify correction verification module, including two-way treatment progress；Treatment progress for reading the identification of each group party in request in real time all the way
As a result, and be compared with demand text information, when in a certain group of party in request's recognition result industry keyword and technical ability it is crucial
Word is satisfied by nearly adopted matching rule in demand text information, that is, meets industry keyword and technical ability keyword and believe in demand word
There are near synonym, synonym or same words in breath, then using the sector keyword and technical ability keyword as the industry of party in request and
One group of recognition result of technical ability is sent to party in request's Recognition feedback module；Treatment progress for reading each group service side in real time all the way
Recognition result, and be compared with service text information, when the industry keyword and technical ability in a certain group of service side's recognition result
Keyword is satisfied by nearly adopted matching rule in servicing text information, that is, meets industry keyword and technical ability keyword in service text
There are near synonym, synonym or same words in word information, then using the sector keyword and technical ability keyword as the row of service side
One group of recognition result of industry and technical ability is sent to service side's Recognition feedback module；Near synonym and synonymous word judgment are respectively by near synonym
Library and thesaurus retrieval obtain.
Further, near synonym, synonym and same word quantity meter of the match index by the nearly adopted matching rule of satisfaction are identified
It calculates and obtains, specific formula for calculation is：
T=am+bn+cl
In formula, T is identification match index, and a is that near synonym match weight, and b is that synonym matches weight, and c is same word
With weight, m is near synonym quantity, and n is synonym quantity, and l is same word quantity, a<b<c.
Further, a=0.6, b=0.9, c=1.
Further, when being matched using nearly adopted matching rule, near synonym, synonym and the same word of successful match
It is rejected from recognition result and text information, is no longer participate in the subsequent match of this group of recognition result, recognition result includes party in request
Recognition result and service side's recognition result, text information include demand text information and service text information.
Further, search rule requires participle to be retrieved in technical ability dictionary and industry dictionary with same word, synonymous
The form of word and near synonym is retrieved successively, if there are same word, synonym or near synonym, stops retrieving, and will wait for
The participle of retrieval and the keyword in dictionary exclude, not repeated retrieval.
Further, recommend index by party in request's recognition result industry keyword and technical ability keyword and service side know
The correspondence degree of correlation of industry keyword and technical ability keyword in other result determines, wherein pushing away in party in request's recommending module
Recommending formula of index is：
Q1=F1 [e1h1+e2h2+e3h3-e4h4]+G1 [j1k1+j2k2+j3k3-j4k4]
In formula, F1 is that the industry of party in request recommends weight, G1 to recommend weight, F1 for the technical ability of party in request>G1, e1, e2, e3
It is respectively that the industry keyword in party in request's recognition result and the industry keyword in service side's recognition result are judged as closely with e4
The degree of correlation weight of adopted word, synonym, same word and other words of party in request, e4<e1<e2<E3, j1, j2, j3 and j4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in service side's recognition result in party in request's recognition result
The degree of correlation weight of word, same word and other words of party in request, j4<j1<j2<J3, h1, h2, h3 and h4 are respectively demand
There are near synonym, synonym, same for industry keyword of the industry keyword in service side's recognition result in square recognition result
The quantity of word and other words of party in request, k1, k2, k3 and k4 are respectively that the technical ability keyword in party in request's recognition result is taking
There are the quantity of near synonym, synonym, same word and other words of party in request for technical ability keyword in business side's recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively；
Recommendation formula of index in service side's recommending module is：
Q2=F2 [u1w1+u2w2+u3w3-u4w4]+G2 [x1y1+x2y2+x3y3-x4y4]
In formula, F2 is that the industry of service side recommends weight, G2 to recommend weight, F2 for the technical ability of service side>G2, u1, u2, u3
It is respectively that the industry keyword in service side's recognition result and the industry keyword in party in request's recognition result are judged as closely with u4
The degree of correlation weight of adopted word, synonym, same word and service side other words, u4<u1<u2<U3, x1, x2, x3 and x4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in party in request recognition result in service side's recognition result
The degree of correlation weight of word, same word and service side other words, x4<x1<x2<X3, w1, w2, w3 and w4 are respectively to service
There are near synonym, synonym, same for industry keyword of the industry keyword in party in request's recognition result in square recognition result
The quantity of other words of word and service side, y1, y2, y3 and y4, which are respectively the technical ability keyword in service side's recognition result, to be needed
There are the quantity of near synonym, synonym, same word and service side other words for technical ability keyword in the side's of asking recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively.
Further, other words of party in request refer to industry keyword in party in request's recognition result in service side's recognition result
In industry keyword in there is no near synonym, the pairing relationship of synonym and same word and party in request's recognition result
Technical ability key of the technical ability keyword in service side's recognition result in the pairings of near synonym, synonym and same word is not present
Relationship；Other words of service side refer to that industry of the industry keyword in party in request's recognition result in service side's recognition result is crucial
There is no the technical ability keywords near synonym, the pairing relationship of synonym and same word and service side's recognition result in word
There is no the pairing relationships of near synonym, synonym and same word in technical ability key in party in request's recognition result；In the presence of with
The near synonym of relationship, synonym and same word are rejected from party in request's recognition result and service side's recognition result, do not repeated
Pairing.
Further, F1=F2=0.6, G1=G2=0.4；E4=u4=0.1, e1=u1=0.3, e2=u2=0.6,
E3=u3=1；J4=x4=0.05, j1=x1=0.3, j2=x2=0.6, j3=x3=1.
Compared with prior art, the present invention advantage is：(1) using demand party subsystem, bring together party subsystem with
And the service mutual cooperative cooperating of party subsystem, can be respectively the identification classification of party in request and service side's offer industry and technical ability,
And according to the matching relationship of industry and technical ability, realize the mutual recommendation of party in request and service side；(2) identification match index side is utilized
Just party in request or service side understand the matching result of automatic identification, to fast and easy select it is opposite with itself industry and technical ability
The classification answered, contributes to fast self-help to classify；(3) it is mutually matched by recommending index that party in request or service side can be facilitated to understand
Degree effectively increases to quickly select cooperation object and brings efficiency together.
Description of the drawings
Fig. 1 is the working-flow figure of the present invention.
Specific implementation mode
Technical solution of the present invention is described in detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to
The embodiment.
Embodiment 1：
As shown in Figure 1, the identification categorizing system disclosed by the invention applied to service trade trade matching includes：Demand prescription
System brings party subsystem and service party subsystem together；
Demand party subsystem, the demand text information for obtaining party in request's input, and demand text information is sent to
Bring party subsystem together；
Party subsystem is serviced, the service text information for obtaining service side's input, and service text information is sent to
Bring party subsystem together；
Bring party subsystem, including two groups for the treatment of progress together；One group for the treatment of progress is used for real-time reception demand text information, and
Parsing obtains the industry and technical ability of party in request, then the industry and technical ability that parsing obtains are fed back to demand party subsystem；Another group
Treatment progress services text information for real-time reception, and parses the industry and technical ability for obtaining service side, then parsing is obtained
Industry and technical ability feed back to service party subsystem.
Further, demand party subsystem includes party in request's text conversion module, party in request's recommending module and party in request
Recognition feedback module；Service party subsystem includes service side's text conversion module, service side's recommending module and service side's identification
Feedback module；
Party in request's text conversion module, the input information for receiving party in request, and input information is converted into demand text
Word information, then demand text information is sent to and brings party subsystem together；
Service side's text conversion module, the input information for receiving service side, and input information is converted into service text
Word information, then service text information is sent to and brings party subsystem together；
Party in request's Recognition feedback module, the identification of industry and technical ability for receiving the party in request for brining party subsystem transmission together
As a result, and provide the identification match index of each group recognition result for party in request, and each recognition result is both provided with selection option,
If party in request is not selected, default choice whole recognition result provides pushing away for service side after selecting to confirm for party in request
Recommend request；
Service side's Recognition feedback module, the identification of industry and technical ability for receiving the service side for brining party subsystem transmission together
As a result, and provide the identification match index of each group recognition result for service side, and each recognition result is both provided with selection option,
If service side does not select, default choice whole recognition result provides pushing away for party in request after selecting to confirm for service side
Recommend request；
Party in request's recommending module, for after party in request receives recommendation request, recommending the industry with party in request to party in request
With the corresponding service side of technical ability, and provide the recommendation index of each service side, if party in request recommendation request is not carried out it is any
Response, then give tacit consent to party in request and receive recommendation request；
Service side's recommending module, for after service side receives recommendation request, recommending the industry with service side to service side
With the corresponding party in request of technical ability, and provide the recommendation index of each party in request, if service side recommendation request is not carried out it is any
Response, then default service side receives recommendation request.
Further, bring together party subsystem include syntactic analysis module, Lexical Analysis Module, syntax normal form library, participle library,
Keyword retrieval module, technical ability dictionary, industry dictionary, keyword memory module, industry and technical ability mapping block, industry and technical ability
Identification module and identification correction verification module；
Syntactic analysis module, including two-way treatment progress；Treatment progress for reading party in request's text conversion in real time all the way
The demand text information that module is sent, and demand text information is carried out at subordinate sentence according to the syntax normal form in syntax normal form library
Reason, obtains each demand input subordinate sentence, and text information puts in order and sequentially numbered to demand input subordinate sentence as desired；
Another way treatment progress is used for the service text information that real-time reading service side's text conversion module is sent, and according to syntax normal form
Syntax normal form in library carries out subordinate sentence processing to service text information, obtains each import of services subordinate sentence, and according to service word
Information arrangement sequence sequentially numbers import of services subordinate sentence；
Lexical Analysis Module, including two-way treatment progress；Treatment progress is used to be successively read each need according to number all the way
Input subordinate sentence is sought, and word segmentation processing is carried out to each demand input subordinate sentence according to participle library, each demand is obtained and segments, and according to
Sequencing segments each demand and numbers；Another way treatment progress is used to be successively read each import of services point according to number
Sentence, and word segmentation processing is carried out to each import of services subordinate sentence according to participle library, each service participle is obtained, and according to sequencing
Each service is segmented and is numbered；
Keyword retrieval module, including two-way treatment progress；Treatment progress is each for being successively read according to number all the way
Demand segments, and each demand participle is retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, by technical ability word
In library party in request's skill that corresponding technical ability keyword is stored in keyword memory module is segmented with the demand currently retrieved
In energy keyword queue, the industry keyword corresponding with the demand participle currently retrieved in industry dictionary is stored in key
In party in request's industry keyword queue in word memory module；Another way treatment progress is used to be successively read each clothes according to number
Business participle, each service participle is retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, by technical ability dictionary
In technical ability keyword corresponding with the service participle currently retrieved be stored in service side's technical ability in keyword memory module
In keyword queue, the industry keyword corresponding with the service participle currently retrieved in industry dictionary is stored in keyword
In service side's industry keyword queue in memory module；
Industry and technical ability mapping block, including two-way treatment progress；Treatment progress for reading party in request's skill in real time all the way
Energy keyword queue and party in request's industry keyword queue map each of party in request according to the mapping table of technical ability and industry
A technical ability keyword and industry-by-industry keyword filter out the party in request's all mappings pair that disclosure satisfy that mapping table；It is another
Road treatment progress is used for real-time reading service side's technical ability keyword queue and service side's industry keyword queue, according to technical ability and row
The mapping table of industry, each technical ability keyword and the industry-by-industry keyword of mapping services side, filters out and disclosure satisfy that mapping
The service side of relation table all mappings pair；Mapping an industry keyword and a technical ability keyword to being made of；
Industry and technical ability identification module, including two-way treatment progress；Treatment progress is all reflected for receiving party in request all the way
It penetrates pair, and merges the mapping of mutually keyword of the same trade to for one group of party in request's recognition result, exporting each group after merging treatment needs
The side's of asking recognition result；Another way treatment progress merges reflecting for phase keyword of the same trade for receiving service side's all mappings pair
It penetrates to for one group of service side's recognition result, exporting each group service side's recognition result after merging treatment；The industry when merging
Keyword remains unchanged, and technical ability keyword is added and duplicate removal；
Identify correction verification module, including two-way treatment progress；Treatment progress for reading the identification of each group party in request in real time all the way
As a result, and be compared with demand text information, when in a certain group of party in request's recognition result industry keyword and technical ability it is crucial
Word is satisfied by nearly adopted matching rule in demand text information, that is, meets industry keyword and technical ability keyword and believe in demand word
There are near synonym, synonym or same words in breath, then using the sector keyword and technical ability keyword as the industry of party in request and
One group of recognition result of technical ability is sent to party in request's Recognition feedback module；Another way treatment progress for reading each group service in real time
Square recognition result, and be compared with service text information, when the industry keyword and skill in a certain group of service side's recognition result
Energy keyword is satisfied by nearly adopted matching rule in servicing text information, that is, meets industry keyword and technical ability keyword is servicing
There are near synonym, synonym or same words in text information, then using the sector keyword and technical ability keyword as service side's
One group of recognition result of industry and technical ability is sent to service side's Recognition feedback module；Near synonym and synonymous word judgment are respectively by nearly justice
Dictionary and thesaurus retrieval obtain.
Further, near synonym, synonym and same word quantity meter of the match index by the nearly adopted matching rule of satisfaction are identified
It calculates and obtains, specific formula for calculation is：
T=am+bn+cl
In formula, T is identification match index, and a is that near synonym match weight, and b is that synonym matches weight, and c is same word
With weight, m is near synonym quantity, and n is synonym quantity, and l is same word quantity, a<b<c.
Further, a=0.6, b=0.9, c=1.
Further, when being matched using nearly adopted matching rule, near synonym, synonym and the same word of successful match
It is rejected from recognition result and text information, is no longer participate in the subsequent match of this group of recognition result, recognition result includes party in request
Recognition result and service side's recognition result, text information include demand text information and service text information.
Further, search rule requires participle to be retrieved in technical ability dictionary and industry dictionary with same word, synonymous
The form of word and near synonym is retrieved successively, if there are same word, synonym or near synonym, stops retrieving, and will wait for
The participle of retrieval and the keyword in dictionary exclude, not repeated retrieval.
Further, recommend index by party in request's recognition result industry keyword and technical ability keyword and service side know
The correspondence degree of correlation of industry keyword and technical ability keyword in other result determines, wherein pushing away in party in request's recommending module
Recommending formula of index is：
Q1=F1 [e1h1+e2h2+e3h3-e4h4]+G1 [j1k1+j2k2+j3k3-j4k4]
In formula, F1 is that the industry of party in request recommends weight, G1 to recommend weight, F1 for the technical ability of party in request>G1, e1, e2, e3
It is respectively that the industry keyword in party in request's recognition result and the industry keyword in service side's recognition result are judged as closely with e4
The degree of correlation weight of adopted word, synonym, same word and other words of party in request, e4<e1<e2<E3, j1, j2, j3 and j4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in service side's recognition result in party in request's recognition result
The degree of correlation weight of word, same word and other words of party in request, j4<j1<j2<J3, h1, h2, h3 and h4 are respectively demand
There are near synonym, synonym, same for industry keyword of the industry keyword in service side's recognition result in square recognition result
The quantity of word and other words of party in request, k1, k2, k3 and k4 are respectively that the technical ability keyword in party in request's recognition result is taking
There are the quantity of near synonym, synonym, same word and other words of party in request for technical ability keyword in business side's recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively；
Recommendation formula of index in service side's recommending module is：
Q2=F2 [u1w1+u2w2+u3w3-u4w4]+G2 [x1y1+x2y2+x3y3-x4y4]
In formula, F2 is that the industry of service side recommends weight, G2 to recommend weight, F2 for the technical ability of service side>G2, u1, u2, u3
It is respectively that the industry keyword in service side's recognition result and the industry keyword in party in request's recognition result are judged as closely with u4
The degree of correlation weight of adopted word, synonym, same word and service side other words, u4<u1<u2<U3, x1, x2, x3 and x4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in party in request recognition result in service side's recognition result
The degree of correlation weight of word, same word and service side other words, x4<x1<x2<X3, w1, w2, w3 and w4 are respectively to service
There are near synonym, synonym, same for industry keyword of the industry keyword in party in request's recognition result in square recognition result
The quantity of other words of word and service side, y1, y2, y3 and y4, which are respectively the technical ability keyword in service side's recognition result, to be needed
There are the quantity of near synonym, synonym, same word and service side other words for technical ability keyword in the side's of asking recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively.
Further, other words of party in request refer to industry keyword in party in request's recognition result in service side's recognition result
In industry keyword in there is no near synonym, the pairing relationship of synonym and same word and party in request's recognition result
Technical ability key of the technical ability keyword in service side's recognition result in the pairings of near synonym, synonym and same word is not present
Relationship；Other words of service side refer to that industry of the industry keyword in party in request's recognition result in service side's recognition result is crucial
There is no the technical ability keywords near synonym, the pairing relationship of synonym and same word and service side's recognition result in word
There is no the pairing relationships of near synonym, synonym and same word in technical ability key in party in request's recognition result；In the presence of with
The near synonym of relationship, synonym and same word are rejected from party in request's recognition result and service side's recognition result, do not repeated
Pairing.
Further, F1=F2=0.6, G1=G2=0.4；E4=u4=0.1, e1=u1=0.3, e2=u2=0.6,
E3=u3=1；J4=x4=0.05, j1=x1=0.3, j2=x2=0.6, j3=x3=1.
Identification categorizing system disclosed by the invention applied to service trade trade matching at runtime, party in request and service side
Party subsystem is brought together by demand party subsystem and service party subsystem access respectively；
When party in request in use, first choice is received the input information of party in request by party in request's text conversion module, and will input
Information is converted to demand text information, then demand text information is sent to and brings party subsystem together；Again by party in request's Recognition feedback
Module receives the recognition result of the industry and technical ability for the party in request for brining party subsystem transmission together, and provides each group identification for party in request
As a result identification match index, and each recognition result is both provided with selection option, if party in request is not selected, gives tacit consent to choosing
Whole recognition results are selected, the recommendation request of service side is provided after selecting to confirm for party in request；Finally by party in request's recommending module
After party in request receives recommendation request, recommend service side corresponding with the industry of party in request and technical ability to party in request, and provide
The recommendation index of each service side gives tacit consent to party in request and receives to recommend to ask if party in request does not carry out any response to recommendation request
It asks；Wherein, identification match index is calculated by the near synonym, synonym and the same word quantity that meet nearly adopted matching rule and is obtained, tool
Body calculation formula is：
T=am+bn+cl
In formula, T is identification match index, and a is that near synonym match weight, and b is that synonym matches weight, and c is same word
With weight, m is near synonym quantity, and n is synonym quantity, and l is same word quantity, a<b<c；Preferably, a=0.6, b=0.9, c
=1.
As service side in use, first choice is received the input information of service side by service side's text conversion module, and will input
Information is converted to service text information, then service text information is sent to and brings party subsystem together；Again by service side's Recognition feedback
Module receives the recognition result of the industry and technical ability for the service side for brining party subsystem transmission together, and provides each group identification for service side
As a result identification match index, and each recognition result is both provided with selection option, if service side does not select, gives tacit consent to choosing
Whole recognition results are selected, the recommendation request of party in request is provided after selecting to confirm for service side；Finally by service side's recommending module
After service side receives recommendation request, recommend party in request corresponding with the industry of service side and technical ability to service side, and provide
The recommendation index of each party in request, if service side does not carry out any response to recommendation request, default service side receives to recommend to ask
It asks；Wherein, recommend index by the industry keyword and technical ability keyword and service side's recognition result in party in request's recognition result
Industry keyword and technical ability keyword correspondence degree of correlation determine, wherein the recommendation index meter in party in request's recommending module
Calculating formula is：
Q1=F1 [e1h1+e2h2+e3h3-e4h4]+G1 [j1k1+j2k2+j3k3-j4k4]
In formula, F1 is that the industry of party in request recommends weight, G1 to recommend weight, F1 for the technical ability of party in request>G1, e1, e2, e3
It is respectively that the industry keyword in party in request's recognition result and the industry keyword in service side's recognition result are judged as closely with e4
The degree of correlation weight of adopted word, synonym, same word and other words of party in request, e4<e1<e2<E3, j1, j2, j3 and j4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in service side's recognition result in party in request's recognition result
The degree of correlation weight of word, same word and other words of party in request, j4<j1<j2<J3, h1, h2, h3 and h4 are respectively demand
There are near synonym, synonym, same for industry keyword of the industry keyword in service side's recognition result in square recognition result
The quantity of word and other words of party in request, k1, k2, k3 and k4 are respectively that the technical ability keyword in party in request's recognition result is taking
There are the quantity of near synonym, synonym, same word and other words of party in request for technical ability keyword in business side's recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively；
Recommendation formula of index in service side's recommending module is：
Q2=F2 [u1w1+u2w2+u3w3-u4w4]+G2 [x1y1+x2y2+x3y3-x4y4]
In formula, F2 is that the industry of service side recommends weight, G2 to recommend weight, F2 for the technical ability of service side>G2, u1, u2, u3
It is respectively that the industry keyword in service side's recognition result and the industry keyword in party in request's recognition result are judged as closely with u4
The degree of correlation weight of adopted word, synonym, same word and service side other words, u4<u1<u2<U3, x1, x2, x3 and x4 difference
It is judged as near synonym, synonymous for technical ability keyword and the technical ability keyword in party in request recognition result in service side's recognition result
The degree of correlation weight of word, same word and service side other words, x4<x1<x2<X3, w1, w2, w3 and w4 are respectively to service
There are near synonym, synonym, same for industry keyword of the industry keyword in party in request's recognition result in square recognition result
The quantity of other words of word and service side, y1, y2, y3 and y4, which are respectively the technical ability keyword in service side's recognition result, to be needed
There are the quantity of near synonym, synonym, same word and service side other words for technical ability keyword in the side's of asking recognition result；Nearly justice
Word and synonymous word judgment are obtained by near synonym library and thesaurus retrieval respectively；Wherein, other words of party in request refer to that party in request knows
There is no near synonym, synonyms and same in industry keyword of the industry keyword in service side's recognition result in other result
In the pairing relationship of one word and technical ability key of the technical ability keyword in service side's recognition result in party in request's recognition result
There is no the pairing relationships of near synonym, synonym and same word；Other words of service side refer to the row in service side's recognition result
There is no the pairings of near synonym, synonym and same word to close in industry keyword of the industry keyword in party in request's recognition result
There is no nearly justice in technical ability key of the technical ability keyword in party in request's recognition result in system and service side's recognition result
The pairing relationship of word, synonym and same word；There are the near synonym of pairing relationship, synonym and same words to identify from party in request
As a result it and in service side's recognition result rejects, is no longer participate in pairing；Preferably, F1=F2=0.6, G1=G2=0.4；E4=
U4=0.1, e1=u1=0.3, e2=u2=0.6, e3=u3=1；J4=x4=0.05, j1=x1=0.3, j2=x2=
0.6, j3=x3=1.
Party subsystem is brought together for the text information that real-time processing requirement party subsystem or service party subsystem are sent, is gone forward side by side
Row identification classification, obtains corresponding industry and technical ability feeds back to demand party subsystem or service party subsystem, the specific steps are：
Step 1, the treatment progress all the way of syntactic analysis module reads the demand that party in request's text conversion module is sent in real time
Text information, and subordinate sentence processing is carried out to demand text information according to the syntax normal form in syntax normal form library, obtain each demand
Input subordinate sentence, and as desired text information put in order to demand input subordinate sentence sequentially numbered；Another way treatment progress
The service text information that real-time reading service side text conversion module is sent, and according to the syntax normal form in syntax normal form library to clothes
Business text information carries out subordinate sentence processing, obtains each import of services subordinate sentence, and put in order to service according to service text information
Input subordinate sentence is sequentially numbered；
Step 2, the treatment progress all the way of Lexical Analysis Module is successively read each demand according to number and inputs subordinate sentence, and root
Word segmentation processing is carried out to each demand input subordinate sentence according to participle library, obtains each demand participle, and according to sequencing to each
Demand participle number；Another way treatment progress is successively read each import of services subordinate sentence according to number, and according to participle library to each
A import of services subordinate sentence carries out word segmentation processing, obtains each service participle, and segment and number to each service according to sequencing；
Step 3, the treatment progress all the way of keyword retrieval module is successively read each demand according to number and segments, Mei Gexu
It asks participle to be retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, will be examined with current in technical ability dictionary
The demand of rope segments in party in request's technical ability keyword queue that corresponding technical ability keyword is stored in keyword memory module,
Corresponding industry keyword will be segmented in industry dictionary with the demand currently retrieved to be stored in keyword memory module
In party in request's industry keyword queue；Another way treatment progress is successively read each service according to number and segments, each service point
Word is retrieved in technical ability dictionary and industry dictionary respectively, according to search rule, by technical ability dictionary with currently retrieve
Service segments corresponding technical ability keyword and is stored in service side's technical ability keyword queue in keyword memory module, will go
Industry keyword corresponding with the service participle currently retrieved in industry dictionary is stored in the service in keyword memory module
In Fang Hangye keyword queues；
Step 4, the treatment progress all the way of industry and technical ability mapping block read in real time party in request's technical ability keyword queue and
Party in request's industry keyword queue, according to the mapping table of technical ability and industry, map each technical ability keyword of party in request with
Industry-by-industry keyword filters out the party in request's all mappings pair that disclosure satisfy that mapping table；Another way treatment progress is real-time
The technical ability keyword queue of reading service side and service side's industry keyword queue are reflected according to the mapping table of technical ability and industry
The each technical ability keyword and industry-by-industry keyword for penetrating service side filter out and disclosure satisfy that the service side of mapping table is whole
Mapping pair；Mapping an industry keyword and a technical ability keyword to being made of；
Step 5, the treatment progress all the way of industry and technical ability identification module receives party in request's all mappings pair, and merges identical
The mapping of industry keyword is to for one group of party in request's recognition result, exporting each group party in request recognition result after merging treatment；Separately
Treatment progress receives service side's all mappings pair all the way, and merges the mapping of mutually keyword of the same trade to being identified for one group of service side
As a result, each group service side's recognition result after output merging treatment；When merging, industry keyword remains unchanged, and technical ability is closed
Keyword is added and duplicate removal；
Step 6, identify the treatment progress all the way of correction verification module for reading each group party in request recognition result in real time, and with need
Ask text information to be compared, when in a certain group of party in request's recognition result industry keyword and technical ability keyword in demand word
It is satisfied by nearly adopted matching rule in information, that is, meets industry keyword and technical ability keyword and there is nearly justice in demand text information
Word, synonym or same word, then using the sector keyword and technical ability keyword as one group of knowledge of the industry of party in request and technical ability
Other result is sent to party in request's Recognition feedback module；Another way treatment progress reads each group service side's recognition result in real time, and with
Service text information be compared, when in a certain group of service side's recognition result industry keyword and technical ability keyword service text
It is satisfied by nearly adopted matching rule in word information, that is, meets industry keyword and technical ability keyword and exists closely in servicing text information
Adopted word, synonym or same word, then using the sector keyword and technical ability keyword as one group of the industry of service side and technical ability
Recognition result is sent to service side's Recognition feedback module；Near synonym and synonymous word judgment are examined by near synonym library and thesaurus respectively
Rope obtains；Wherein, when being matched using nearly adopted matching rule, near synonym, synonym and the same word of successful match are from knowledge
It is rejected in other result and text information, is no longer participate in the subsequent match of this group of recognition result, recognition result includes party in request's identification
As a result with service side's recognition result, text information includes demand text information and service text information.
As described above, although the present invention has been indicated and described with reference to specific preferred embodiment, must not explain
For the limitation to invention itself.It without prejudice to the spirit and scope of the invention as defined in the appended claims, can be right
Various changes can be made in the form and details for it.
Claims (10)
1. a kind of identification categorizing system applied to service trade trade matching, which is characterized in that including demand party subsystem, bring together
Party subsystem and service party subsystem；
Demand party subsystem, the demand text information for obtaining party in request's input, and demand text information is sent to and is brought together
Party subsystem；
Party subsystem is serviced, the service text information for obtaining service side's input, and service text information is sent to and is brought together
Party subsystem；
Bring party subsystem, including two groups for the treatment of progress together；One group for the treatment of progress is used for real-time reception demand text information, and parses
The industry and technical ability of party in request are obtained, then the industry and technical ability that parsing obtains are fed back into demand party subsystem；Another group of processing
Process services text information for real-time reception, and parses the industry and technical ability for obtaining service side, then the industry that parsing is obtained
Service party subsystem is fed back to technical ability.
2. the identification categorizing system according to claim 1 applied to service trade trade matching, which is characterized in that party in request
Subsystem includes party in request's text conversion module, party in request's recommending module and party in request's Recognition feedback module；Service side's subsystem
System includes service side's text conversion module, service side's recommending module and service side's Recognition feedback module；
Party in request's text conversion module, the input information for receiving party in request, and input information is converted into demand word letter
Breath, then demand text information is sent to and brings party subsystem together；
Service side's text conversion module, the input information for receiving service side, and input information is converted into service word letter
Breath, then service text information is sent to and brings party subsystem together；
Party in request's Recognition feedback module, the identification knot of industry and technical ability for receiving the party in request for brining party subsystem transmission together
Fruit, and be the identification match index that party in request provides each group recognition result, and each recognition result is both provided with selection option, if
Party in request is not selected, then default choice whole recognition result, provides the recommendation of service side for party in request after selecting to confirm
Request；
Service side's Recognition feedback module, the identification knot of industry and technical ability for receiving the service side for brining party subsystem transmission together
Fruit, and be the identification match index that service side provides each group recognition result, and each recognition result is both provided with selection option, if
Service side does not select, then default choice whole recognition result, provides the recommendation of party in request for service side after selecting to confirm
Request；
Party in request's recommending module, for after party in request receives recommendation request, recommending the industry and skill with party in request to party in request
The corresponding service side of energy, and the recommendation index of each service side is provided, if party in request does not carry out any response to recommendation request,
Then give tacit consent to party in request and receives recommendation request；
Service side's recommending module, for after service side receives recommendation request, recommending the industry and skill with service side to service side
The corresponding party in request of energy, and the recommendation index of each party in request is provided, if service side does not carry out any response to recommendation request,
Then default service side receives recommendation request.
3. the identification categorizing system according to claim 2 applied to service trade trade matching, which is characterized in that the side of brining together
Subsystem includes syntactic analysis module, Lexical Analysis Module, syntax normal form library, participle library, keyword retrieval module, technical ability word
Library, industry dictionary, keyword memory module, industry and technical ability mapping block, industry and technical ability identification module and identification verification
Module；
Syntactic analysis module, including two-way treatment progress；Treatment progress for reading party in request's text conversion module in real time all the way
The demand text information of transmission, and subordinate sentence processing is carried out to demand text information according to the syntax normal form in syntax normal form library, it obtains
Each demand inputs subordinate sentence, and text information puts in order and is sequentially numbered to demand input subordinate sentence as desired；It is another
Road treatment progress is used for the service text information that real-time reading service side's text conversion module is sent, and according in syntax normal form library
Syntax normal form subordinate sentence processing is carried out to service text information, obtain each import of services subordinate sentence, and according to service text information
It puts in order and import of services subordinate sentence is sequentially numbered；
Lexical Analysis Module, including two-way treatment progress；Treatment progress is used to be successively read each demand according to number defeated all the way
Enter subordinate sentence, and word segmentation processing is carried out to each demand input subordinate sentence according to participle library, obtains each demand participle, and according to successively
Sequence segments each demand and numbers；Another way treatment progress is used to be successively read each import of services subordinate sentence according to number, and
Word segmentation processing is carried out to each import of services subordinate sentence according to participle library, obtains each service participle, and according to sequencing to each
A service participle number；
Keyword retrieval module, including two-way treatment progress；Treatment progress is used to be successively read each demand according to number all the way
Participle, each demand participle are retrieved in technical ability dictionary and industry dictionary respectively, will be in technical ability dictionary according to search rule
Segment party in request's technical ability that corresponding technical ability keyword is stored in keyword memory module with the demand currently retrieved and close
In keyword queue, the industry keyword corresponding with the demand participle currently retrieved in industry dictionary is stored in keyword and is deposited
In Chu Mo party in request's industry keyword queues in the block；Another way treatment progress is used to be successively read each service point according to number
Word, each service participle are retrieved in technical ability dictionary and industry dictionary respectively, will be in technical ability dictionary according to search rule
It is crucial that technical ability keyword corresponding with the service participle currently retrieved is stored in service side's technical ability in keyword memory module
In word queue, the industry keyword corresponding with the service participle currently retrieved in industry dictionary is stored in keyword storage
In mould service side's industry keyword queue in the block；
Industry and technical ability mapping block, including two-way treatment progress；Treatment progress is closed for reading party in request's technical ability in real time all the way
Keyword queue and party in request's industry keyword queue map each skill of party in request according to the mapping table of technical ability and industry
Energy keyword and industry-by-industry keyword filter out the party in request's all mappings pair that disclosure satisfy that mapping table；At another way
Reason process is used for real-time reading service side's technical ability keyword queue and service side's industry keyword queue, according to technical ability and industry
Mapping table, each technical ability keyword and the industry-by-industry keyword of mapping services side, filters out and disclosure satisfy that mapping relations
The service side of table all mappings pair；Mapping an industry keyword and a technical ability keyword to being made of；
Industry and technical ability identification module, including two-way treatment progress；Treatment progress is used to receive party in request's all mappings pair all the way,
And merge the mapping of mutually keyword of the same trade to for one group of party in request's recognition result, exporting each group party in request after merging treatment and knowing
Other result；Another way treatment progress for receiving service side's all mappings pair, and merge the mapping of mutually keyword of the same trade to for
One group of service side's recognition result exports each group service side's recognition result after merging treatment；The industry keyword when merging
It remains unchanged, technical ability keyword is added and duplicate removal；
Identify correction verification module, including two-way treatment progress；Treatment progress reads each group party in request recognition result for real-time all the way,
And be compared with demand text information, when in a certain group of party in request's recognition result industry keyword and technical ability keyword need
It asks and is satisfied by nearly adopted matching rule in text information, that is, meet industry keyword and technical ability keyword is deposited in demand text information
In near synonym, synonym or same word, then using the sector keyword and technical ability keyword as the industry of party in request and technical ability
One group of recognition result is sent to party in request's Recognition feedback module；Another way treatment progress for reading each group service side identification in real time
As a result, and be compared with service text information, when in a certain group of service side's recognition result industry keyword and technical ability it is crucial
Word is satisfied by nearly adopted matching rule in servicing text information, that is, meets industry keyword and technical ability keyword in service word letter
There are near synonym, synonym or same words in breath, then using the sector keyword and technical ability keyword as the industry of service side and
One group of recognition result of technical ability is sent to service side's Recognition feedback module；Near synonym and synonymous word judgment respectively by near synonym library and
Thesaurus retrieval obtains.
4. the identification categorizing system according to claim 3 applied to service trade trade matching, which is characterized in that identification
It is calculated and is obtained by the near synonym, synonym and the same word quantity that meet nearly adopted matching rule with index, specific formula for calculation is：
T=am+bn+cl
In formula, T is identification match index, and a is that near synonym match weight, and b is that synonym matches weight, and c is same word matching power
Weight, m are near synonym quantity, and n is synonym quantity, and l is same word quantity, a<b<c.
5. the identification categorizing system according to claim 4 applied to service trade trade matching, which is characterized in that a=
0.6, b=0.9, c=1.
6. the identification categorizing system according to claim 4 applied to service trade trade matching, which is characterized in that utilizing
When nearly justice matching rule is matched, near synonym, synonym and the same word of successful match are from recognition result and text information
It rejects, is no longer participate in the subsequent match of this group of recognition result, recognition result includes that party in request's recognition result and service side identify knot
Fruit, text information include demand text information and service text information.
7. the identification categorizing system according to claim 3 applied to service trade trade matching, which is characterized in that retrieval rule
Then require participle to be retrieved in technical ability dictionary and industry dictionary in the form of same word, synonym and near synonym successively into
Row retrieval, if there are same word, synonym or near synonym, stops retrieving, and by the pass in participle and dictionary to be retrieved
Keyword excludes, not repeated retrieval.
8. the identification categorizing system according to claim 2 applied to service trade trade matching, which is characterized in that recommendation refers to
Number by the industry keywords in party in request's recognition result and the industry keyword in technical ability keyword and service side's recognition result and
The correspondence degree of correlation of technical ability keyword determines, wherein the recommendation formula of index in party in request's recommending module is：
Q1=F1 [e1h1+e2h2+e3h3-e4h4]+G1 [j1k1+j2k2+j3k3-j4k4]
In formula, F1 is that the industry of party in request recommends weight, G1 to recommend weight, F1 for the technical ability of party in request>G1, e1, e2, e3 and e4
Industry keyword respectively in party in request's recognition result and the industry keyword in service side's recognition result be judged as near synonym,
The degree of correlation weight of synonym, same word and other words of party in request, e4<e1<e2<E3, j1, j2, j3 and j4 are respectively to need
Technical ability keyword in the side's of asking recognition result and the technical ability keyword in service side's recognition result be judged as near synonym, synonym,
The degree of correlation weight of same word and other words of party in request, j4<j1<j2<J3, h1, h2, h3 and h4 are respectively that party in request knows
Industry keyword of the industry keyword in service side's recognition result in other result there are near synonym, synonym, same word with
And the quantity of other words of party in request, k1, k2, k3 and k4 are respectively technical ability keyword in party in request's recognition result in service side
There are the quantity of near synonym, synonym, same word and other words of party in request for technical ability keyword in recognition result；Near synonym and
Synonymous word judgment is obtained by near synonym library and thesaurus retrieval respectively；
Recommendation formula of index in service side's recommending module is：
Q2=F2 [u1w1+u2w2+u3w3-u4w4]+G2 [x1y1+x2y2+x3y3-x4y4]
In formula, F2 is that the industry of service side recommends weight, G2 to recommend weight, F2 for the technical ability of service side>G2, u1, u2, u3 and u4
Industry keyword respectively in service side's recognition result and the industry keyword in party in request's recognition result be judged as near synonym,
The degree of correlation weight of synonym, same word and service side other words, u4<u1<u2<U3, x1, x2, x3 and x4 are respectively to take
Technical ability keyword in business side's recognition result and the technical ability keyword in party in request's recognition result be judged as near synonym, synonym,
The degree of correlation weight of other words of same word and service side, x4<x1<x2<X3, w1, w2, w3 and w4 are respectively that service side knows
Industry keyword of the industry keyword in party in request's recognition result in other result there are near synonym, synonym, same word with
And the quantity of service side other words, y1, y2, y3 and y4 are respectively technical ability keyword in service side's recognition result in party in request
There are the quantity of near synonym, synonym, same word and service side other words for technical ability keyword in recognition result；Near synonym and
Synonymous word judgment is obtained by near synonym library and thesaurus retrieval respectively.
9. the identification categorizing system according to claim 8 applied to service trade trade matching, which is characterized in that party in request
Other words refer to being not present in industry keyword of the industry keyword in service side's recognition result in party in request's recognition result
Technical ability keyword near synonym, the pairing relationship of synonym and same word and party in request's recognition result is known in service side
There is no the pairing relationships of near synonym, synonym and same word in technical ability key in other result；Other words of service side refer to
There is no near synonym, synonymous in industry keyword of the industry keyword in party in request's recognition result in service side's recognition result
The pairing relationship and skill of the technical ability keyword in party in request's recognition result in service side's recognition result of word and same word
There is no the pairing relationships of near synonym, synonym and same word during energy is crucial；There are the near synonym of pairing relationship, synonym and
Same word is rejected from party in request's recognition result and service side's recognition result, does not repeat to match.
10. the identification categorizing system according to claim 8 applied to service trade trade matching, which is characterized in that F1=
F2=0.6, G1=G2=0.4；E4=u4=0.1, e1=u1=0.3, e2=u2=0.6, e3=u3=1；J4=x4=
0.05, j1=x1=0.3, j2=x2=0.6, j3=x3=1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810344637.0A CN108536682B (en) | 2018-04-17 | 2018-04-17 | Recognition and classification system applied to matching transaction in service industry |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810344637.0A CN108536682B (en) | 2018-04-17 | 2018-04-17 | Recognition and classification system applied to matching transaction in service industry |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108536682A true CN108536682A (en) | 2018-09-14 |
| CN108536682B CN108536682B (en) | 2021-09-17 |
Family
ID=63481127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810344637.0A Active CN108536682B (en) | 2018-04-17 | 2018-04-17 | Recognition and classification system applied to matching transaction in service industry |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108536682B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544329A (en) * | 2018-10-09 | 2019-03-29 | 阿里巴巴集团控股有限公司 | Bring matching process, device and system together |
| CN109801144A (en) * | 2019-01-30 | 2019-05-24 | 杭州大橙知一科技有限公司 | A kind of method of commerce and equipment for brining scene together based on foreign trade |
| CN109816321A (en) * | 2018-11-27 | 2019-05-28 | 深圳市汇邦企业服务有限公司 | A kind of service management, device, equipment and computer readable storage medium |
| CN109859045A (en) * | 2019-01-30 | 2019-06-07 | 杭州大橙知一科技有限公司 | Bring the information processing method and equipment of scene together based on foreign trade |
| CN110659874A (en) * | 2019-09-09 | 2020-01-07 | 山东泰山智库经济发展有限公司 | Human resource capitalization transaction management system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070156392A1 (en) * | 2005-12-30 | 2007-07-05 | International Business Machines Corporation | Method and system for automatically building natural language understanding models |
| CN102609870A (en) * | 2012-02-03 | 2012-07-25 | 陈晓亮 | System and method for matchmaking and transaction of electronic commerce |
| CN103794214A (en) * | 2014-03-07 | 2014-05-14 | 联想(北京)有限公司 | Information processing method, device and electronic equipment |
| CN106663230A (en) * | 2014-03-14 | 2017-05-10 | 庞德·萨利尔 | Career analytics platform |
| CN107247806A (en) * | 2017-07-04 | 2017-10-13 | 山东浪潮云服务信息科技有限公司 | A kind of patent big data analysis and enterprise's application platform |
| CN107391547A (en) * | 2017-06-01 | 2017-11-24 | 武汉威特管理咨询有限公司 | A kind of manpower object data matching process and system |
| CN107506389A (en) * | 2017-07-27 | 2017-12-22 | 北京德塔精要信息技术有限公司 | A kind of method and apparatus for extracting position skill requirement |
-
2018
- 2018-04-17 CN CN201810344637.0A patent/CN108536682B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070156392A1 (en) * | 2005-12-30 | 2007-07-05 | International Business Machines Corporation | Method and system for automatically building natural language understanding models |
| CN102609870A (en) * | 2012-02-03 | 2012-07-25 | 陈晓亮 | System and method for matchmaking and transaction of electronic commerce |
| CN103794214A (en) * | 2014-03-07 | 2014-05-14 | 联想(北京)有限公司 | Information processing method, device and electronic equipment |
| CN106663230A (en) * | 2014-03-14 | 2017-05-10 | 庞德·萨利尔 | Career analytics platform |
| CN107391547A (en) * | 2017-06-01 | 2017-11-24 | 武汉威特管理咨询有限公司 | A kind of manpower object data matching process and system |
| CN107247806A (en) * | 2017-07-04 | 2017-10-13 | 山东浪潮云服务信息科技有限公司 | A kind of patent big data analysis and enterprise's application platform |
| CN107506389A (en) * | 2017-07-27 | 2017-12-22 | 北京德塔精要信息技术有限公司 | A kind of method and apparatus for extracting position skill requirement |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544329A (en) * | 2018-10-09 | 2019-03-29 | 阿里巴巴集团控股有限公司 | Bring matching process, device and system together |
| CN109544329B (en) * | 2018-10-09 | 2023-09-05 | 创新先进技术有限公司 | Method, device and system for matching |
| CN109816321A (en) * | 2018-11-27 | 2019-05-28 | 深圳市汇邦企业服务有限公司 | A kind of service management, device, equipment and computer readable storage medium |
| CN109801144A (en) * | 2019-01-30 | 2019-05-24 | 杭州大橙知一科技有限公司 | A kind of method of commerce and equipment for brining scene together based on foreign trade |
| CN109859045A (en) * | 2019-01-30 | 2019-06-07 | 杭州大橙知一科技有限公司 | Bring the information processing method and equipment of scene together based on foreign trade |
| CN109801144B (en) * | 2019-01-30 | 2021-03-12 | 米奥兰特(浙江)网络科技有限公司 | Trade method and equipment based on foreign trade matching scene |
| CN110659874A (en) * | 2019-09-09 | 2020-01-07 | 山东泰山智库经济发展有限公司 | Human resource capitalization transaction management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108536682B (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108536682A (en) | A kind of identification categorizing system applied to service trade trade matching | |
| CN105095223B (en) | File classification method and server | |
| CN104063314B (en) | A kind of automated test data generation device and method | |
| CN107729336A (en) | Data processing method, equipment and system | |
| CN110377900A (en) | Checking method, device, computer equipment and the storage medium of Web content publication | |
| CN105787025A (en) | Network platform public account classifying method and device | |
| CN113742488B (en) | Embedded knowledge graph completion method and device based on multitask learning | |
| CN103838857B (en) | Automatic service combination system and method based on semantics | |
| CN110110049A (en) | Service consultation method, device, system, service robot and storage medium | |
| CN108416375A (en) | Work order sorting technique and device | |
| CN111191051B (en) | Method and system for constructing emergency knowledge map based on Chinese word segmentation technology | |
| CN109977175B (en) | Data configuration query method and device | |
| CN115130711A (en) | Data processing method and device, computer and readable storage medium | |
| CN109960719A (en) | A kind of document handling method and relevant apparatus | |
| CN112364622A (en) | Dialog text analysis method, dialog text analysis device, electronic device and storage medium | |
| CN116306974A (en) | Model training method and device of question-answering system, electronic equipment and storage medium | |
| CN117172508B (en) | Automatic dispatch method and system based on city complaint worksheet recognition | |
| CN118245926A (en) | Processing method and device for analyzing legal document elements by using predictive model | |
| CN110209804B (en) | Target corpus determining method and device, storage medium and electronic device | |
| CN109902215A (en) | A kind of method and system of deals match | |
| CN107871055A (en) | A kind of data analysing method and device | |
| Goldberg et al. | CASTLE: crowd-assisted system for text labeling and extraction | |
| CN115952770A (en) | Data standardization processing method and device, electronic equipment and storage medium | |
| CN116361428A (en) | Question-answer recall method, device and storage medium | |
| CN113282686B (en) | Association rule determining method and device for unbalanced sample |
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 |