CN103699955B - Business model analysis method and device based on self-defined classifying rules - Google Patents
Business model analysis method and device based on self-defined classifying rules Download PDFInfo
- Publication number
- CN103699955B CN103699955B CN201310589864.7A CN201310589864A CN103699955B CN 103699955 B CN103699955 B CN 103699955B CN 201310589864 A CN201310589864 A CN 201310589864A CN 103699955 B CN103699955 B CN 103699955B
- Authority
- CN
- China
- Prior art keywords
- business
- recognition
- type
- business model
- text
- 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.)
- Active
Links
Abstract
The invention discloses a kind of business model analysis method and device based on self-defined classifying rules, method therein includes：Obtain identification text；Business Classification and Identification is carried out to the identification text using the business model being successfully established；Wherein, the process of setting up of the business model includes：For an industry, collect the business objective of the sector；Determine that each business objective distinguishes corresponding association keyword；Using the association keyword for each business objective is respectively provided with recognition rule, with the first type of the business model for setting up each business objective；Just type is trained business model using multiple recognition training texts to each business objective, to improve the recognition rule of each business objective, so as to be successfully established the business model of each business objective.The technical scheme that the present invention is provided can be automatically obtained business classification.
Description
Technical field
The present invention relates to business Classification Management technology, more particularly to a kind of business model based on self-defined classifying rules
Analysis method and device.
Background technology
In speech analysis techniques field, it is very important that business classification is carried out to call.
Under normal circumstances, in a call the management system of the heart recorded with record information in, often containing to call carry out
The classification description of business classification, can classify using with the classification description in record information to talk business.However, existing
There is problems with record information：
1st, it is imperfect with record information；It is i.e. not all that business classification is carried out to call with all including in record information
Classification is described, and the management system of some call centers is even wholly without with record information.
2nd, it is not rigorous with record information；Description i.e. to call category is often by attending a banquet what manual operations was recorded, depositing
Situations such as being perfunctory to deal with and subjectivity malfunctions.
3rd, the business for self-defined classification (classification such as in ad hoc survey) cannot generally be recorded in advance in systems.
The above mentioned problem existed with record information can cause business to be classified can not accurately and smoothly be carried out, or even cannot be carried out
Business is classified, and this is very unfavorable for the management system of call center.
In view of the problem that the classification of existing business is present, the present inventor is actively subject to research and innovation, to found one kind
New business model analysis method and device based on self-defined classifying rules, can overcome asking for existing business classification presence
Topic, makes it have more practicality.By continuous research and design, and by studying sample and improvement repeatedly, create finally really
The present invention having practical value.
The content of the invention
It is a primary object of the present invention to, the problem for overcoming the classification of existing business to exist, and provide it is a kind of it is new based on
The business model analysis method and device of self-defined classifying rules, problem to be solved are automatically business to be carried out
Classification is processed.
The purpose of the present invention and solve its technical problem and can be realized using following technical scheme.
According to a kind of business model analysis method based on self-defined classifying rules proposed by the present invention, including：Obtain and know
Other text；
Business Classification and Identification is carried out to the identification text using the business model of each business objective being successfully established；
Wherein, the process of setting up of the business model includes：
For an industry, collect the business objective of the sector；
Determine that each business objective distinguishes corresponding association keyword；
Using the association keyword for each business objective is respectively provided with recognition rule, to set up the business of each business objective
The first type of model；
Just type is trained business model using multiple recognition training texts to each business objective, to improve each industry
The recognition rule of target of being engaged in, so as to be successfully established the business model of each business objective；
It is described using multiple recognition training texts to the business model of each business objective just type be trained including：
Step 1, using single business model just type to multiple recognition training texts be identified checking, with to the multiple
Recognition training text is screened；
Step 2, according to the modified result of the screening single business model just type；
Step 3, repeat the above steps 1-2, so that one by one to all business models, just type is modified；
Step 4, using current revised all business models, just type is identified checking to multiple recognition training texts,
To obtain two kinds of recognition training texts, i.e., can not be by the recognition training of current revised all business models just type identification checking
Text and meet the recognition training text of the first types of multiple current revised business models simultaneously, and according to both identification instructions
Practice text to continue to correct corresponding service model just type；
Step 5,4 are repeated the above steps, until the current two kinds of quantity of recognition training text for obtaining and previous acquisition
The quantity difference of two kinds of recognition training texts is identical, so as to be successfully established the business model of each business objective.
According to a kind of business model analytical equipment based on self-defined classifying rules proposed by the present invention, including：Obtain mould
Block, identification module and set up module；
The acquisition module, for obtaining identification text；
The identification module, enters for the business model using each business objective being successfully established to the identification text
Industry business Classification and Identification；
Wherein, the business model is set up by the module of setting up, and the module of setting up includes：
Collect submodule, for for an industry, collecting the business objective of the sector；
Determination sub-module, for determining that each business objective distinguishes corresponding association keyword；
Set submodule, for using it is described association keyword be each business objective be respectively provided with recognition rule, with foundation
The first type of the business model of each business objective；
Training submodule, for the business model using multiple recognition training texts to each business objective, just type is carried out
Training, to improve the recognition rule of each business objective, so as to be successfully established the business model of each business objective；
The training submodule includes：
First module, for just type to be identified checking to multiple recognition training texts using single business model, with right
The multiple recognition training text is screened；
Second unit, for the first type of the single business model of the modified result according to the screening；
Unit the 3rd, for repeating to call the first module and second unit, with one by one at the beginning of all business models
Type is modified；
Unit the 4th, for just type to be known to multiple recognition training texts using current revised all business models
Do not verify, to obtain two kinds of recognition training texts, i.e., can not be by the just type identification checking of current revised all business models
Recognition training text and meet the recognition training text of the first types of multiple current revised business models simultaneously, and according to this two
Recognition training text is planted to continue to correct corresponding service model just type；
Unit the 5th, for repeating to call Unit the 4th, until the current two kinds of numbers of recognition training text for obtaining
Amount is identical with the quantity difference of two kinds of recognition training texts of previous acquisition, so as to be successfully established the business mould of each business objective
Type.
By above-mentioned technical proposal, business model analysis method and device based on self-defined classifying rules of the invention are extremely
There is following advantages and beneficial effect less：The present invention sets up business model by for each business objective, and utilization is successfully established
Business model business Classification and Identification is carried out to identification text, can automatic to business such as voice calls and fast and accurately enter
Industry business classification treatment；So as to improve the automaticity of service management.
Described above is only the general introduction of technical solution of the present invention, in order to better understand technological means of the invention,
And can be practiced according to the content of specification, and in order to allow the above and other objects, features and advantages of the invention can
Become apparent, below especially exemplified by preferred embodiment, and coordinate Figure of description, describe in detail as follows.
Brief description of the drawings
Fig. 1 is the flow chart of the business model analysis method based on self-defined classifying rules of the invention；
Fig. 2 is the schematic diagram of the business model analytical equipment based on self-defined classifying rules of the invention.
Specific embodiment
Further to illustrate the present invention to reach technological means and effect that predetermined goal of the invention is taken, below in conjunction with
Accompanying drawing and preferred embodiment, to according to business model analysis method and device based on self-defined classifying rules proposed by the present invention
Specific embodiment, structure, feature and its effect, describe in detail as after.
Embodiment one, the business model analysis method based on self-defined classifying rules.The flow of the method is as shown in Figure 1.
In Fig. 1, S100, acquisition identification text.
Specifically, the identification text can be the identification text converted by voice call, such as to currently carrying out
Voice call or the voice call of recording carry out speech recognition, so as to obtain the voice call pair according to voice identification result
The identification text answered.The present embodiment can obtain identification text using existing speech recognition technology, herein no longer to voice
Identification process is described in detail.
S110, business Classification and Identification is carried out to the above-mentioned identification text for getting using the business model being successfully established,
To determine the business objective belonging to the identification text.
Specifically, above-mentioned the identification text for getting and the business model being successfully established can be carried out matching operation,
And the business objective according to belonging to matching result determines the identification text, it is achieved thereby that the business classification to recognizing text.
One specific example, it is determined that the keyword in identification text, and according to the sequencing judgement knowledge of business model
Keyword in other text meets the degree of the recognition rule of business model, and the present embodiment can be in keyword and certain business model
Matching degree when reaching predetermined extent, it is determined that current business model matches with identification text, and determine the identification text
The corresponding business objective of business model for belonging to current, no longer carries out follow-up deterministic process, it is achieved thereby that to identification text
This business classification.In addition, in the above example, it is also possible to which judge identification text and all business models respectively meets journey
Degree, and therefrom choose the best business model of matching degree, the corresponding business objective of business model that this is selected is used as knowledge
Business objective belonging to other text.Certainly, the present invention can also be on the basis of the business model being successfully established, using other
Mode carries out business Classification and Identification to identification text, no longer describes in detail one by one herein.
Business model of the invention to set up process described in detail below：
S200, for an industry, collect the business objective of the sector.
Specifically, the business objective that different industries are included would also vary from, a specific example of business objective is,
For this industry of China Mobile, its business objective can generally include：10086；Mend card, change card and standby card；Open an account, mistake
Family and cancellation；Call reminding；GPRS；CRBT etc..
For a specific industry, all business objectives in the sector can be in the form of business objective list
Pool together.
S210, determine that each business objective distinguishes corresponding association keyword.
Specifically, each business objective in an industry can be to that should have corresponding association keyword, and different business
The corresponding association keyword of target generally should be incomplete same.
One specific example of the corresponding association keyword of business objective is, with the possible phase of " GPRS " this business objective
The association keyword of pass includes：GPRS, flow, surfing Internet with cell phone and WAP etc..
S220, using above-mentioned association keyword for each business objective is respectively provided with recognition rule, to set up each business objective
Business model just type.The recognition rule is properly termed as the recognition rule of the first type of business model.
Specifically, the recognition rule in the present embodiment mainly includes two parts content, that is, associate keyword and association is closed
Logical relation between keyword.
Above-mentioned logical relation can include：With or, at least one of non-and adjacent logical relation；Wherein：
"AND" logical relation is binary logic relation, it is possible to use rule symbol " ＆ " is represented；"AND" logical relation it is excellent
First level could be arranged to normal priority；"AND" logical relation refers to " must be while occurring ", i.e., two parts must deposit simultaneously
Meet "AND" logical relation just will be considered that, for example, " I wants to handle broadband access network " meets " online ＆ broadbands " rule, and " I thinks
Handle surfing Internet with cell phone " " online ＆ broadbands " rule is not met then；
"or" logical relation is binary logic relation, it is possible to use rule symbol " | " is represented；"or" logical relation it is excellent
First level could be arranged to normal priority；"or" logical relation refers to " as long as occurring in which any one "；
" non-" logical relation is monadic logic relation, it is possible to use rule symbol "！" represent；" non-" logical relation it is excellent
First level could be arranged to high priority；" non-" logical relation refers to that the keyword in " can not occur ", the i.e. logical relation is necessary
Do not exist；
" adjoining " logical relation is binary logic relation, it is possible to use rule symbol " # " is represented；" adjoining " logical relation
Priority could be arranged to normal priority；" adjoining " logical relation refers to " occurring in succession ", and two parts in relation must
Can just must be considered as meeting " adjoining " relation, a tool of closer distance in the presence of and with tandem and closer distance simultaneously
The example of body is that closer distance can be defined previously as within two partial distances, 5 Chinese characters (containing 5 Chinese characters), and such as " I wants to do
One card " meets " doing # cards " rule.
When recognition rule is set, recognition rule should be made to meet following regulations：
A, the symbol of characterization logic relation are half-angle character；
B, a recognition rule must have the DBC case for characterizing the DBC case for starting and characterizing end, for example,
One recognition rule is terminated with DBC case " (" start, and with DBC case ") "；
C, in a recognition rule, other logical relations in addition to NOT sum " adjacent logical relation " can be continuous
Use, for example, a recognition rule can be " (walk in the Divine Land of Global Link | M-ZONE |) "；
D, the complexity in view of calculating, in a recognition rule, other logics in addition to adjacent logical relation
Relation can be with mutually nested, and adjacent logical relation can only be nested；For example, one has the mutually nested recognition rule can be with
Be " (Global Link | (dynamic area) | walk in the Divine Land) ", and one have the adjacent logical relation that is nested can be " (Global Link |
(dynamic # areas) | walk in the Divine Land) ", similar to recognition rule as " ((open-minded | open) # (GPRS | surfing Internet with cell phone)) ", it is based on
The consideration of computation complexity, the present embodiment can be supported；
E, in a recognition rule, in a recognition rule, other logical relations in addition to NOT logic relation
Priority answer and explicit represented using the symbol for characterizing priority；If for example, recognition rule for " (brand mutually turns |
(turning # Global Links)) ", then " turning # Global Links " has the priority higher than " brand mutually turns "；But, if the recognition rule is write
Into " (brand mutually turns | and turn # Global Links) ", then the priority of each logical relation that is sequentially sequentially ranked may result in and not meet industry
Business logic；
F, bracket should occur in pairs, i.e., the number of times that " (" and ") " occurs should be identical, otherwise may be considered syntax error；
There should be the symbol of characterization logic relation between G, adjacent association keyword before and after two, i.e. any two is adjacent
Keyword between there must be logical relation symbol；
H, can before and after logical relation symbol some spaces of setting, to strengthen the readability of recognition rule；
I, the symbol without actual logic meaning can be inserted in recognition rule, to strengthen the readability of recognition rule, such as
Can be inserted in recognition rule " (" and ") " without actual logic meaning.
It is provided with after recognition rule for each business objective, just type is also just successfully established the business model of each business objective
Get up.
S230, using multiple recognition training texts, to the business model of each business objective, just type is trained, each to improve
The recognition rule of business objective, so as to form the business model of each business objective.The process of training is specific as follows：
Training step 1, first with single business model, just type is identified checking, and root to multiple recognition training texts
Multiple recognition training texts are screened according to identification the result；I.e. using multiple recognition training texts (as being no less than 10
The recording identification text of ten thousand) to certain business model, just type is trained, such that it is able to all of recognition training text is distinguished
To meet the recognition training text of the first type of the business model and not meeting the recognition training text of the first type of the business model.To instruction
Practice step 2.
Training step 2, according to above-mentioned screening obtain result come correct the single business model just type.To training step 3.
If specifically, certain recognition training text should comply with the business model just type, and passing through above-mentioned training step 1
Afterwards, the recognition training text does not meet the business model just type, then should supplement the industry according to the recognition training text
The recognition rule of the first type of business model；If certain recognition training text not should comply with the business model just type, and pass through above-mentioned
After training step 1, the recognition training text meets the business model just type, then should adjust this according to the recognition training text
The recognition rule of the first type of business model.
Training step 3, the above-mentioned training step 1-2 of repetition, so that one by one to all business models, just type is modified, that is, repair
The recognition rule of the first type of just all business models.To training step 4.
Training step 4, using current revised all business models, just type (is such as no less than to multiple recognition training texts
The recording identification text of 100000) checking is identified, to obtain two kinds of recognition training text, i.e. a, can not be worked as
Just type recognizes the recognition training text of checking to preceding revised all business models, b, while meeting multiple current revised
The recognition training text of the first type of business model；And continued to correct corresponding business mould according to the recognition training text of both types
The first type of type.
If specifically, certain recognition training text should comply with current a certain business model just type, however, in this step
In rapid training, the recognition training text do not meet any business model just type, then should according to the recognition training text come
The recognition rule of the first type of corresponding business model is supplemented, so that it can meet a certain business model just type；If certain is recognized
Training text should not simultaneously meet multiple business models just type, and in the training of this step, the recognition training text is simultaneously
Multiple business models just type is met, then the identification rule of the first type of corresponding business model should have been adjusted according to the recognition training text
Then, so that it only meets a certain business model just type.
Above-mentioned training step 4 is called in training step 5, repetition, until the current two kinds of quantity of recognition training text for obtaining
With the quantity all same (both do not increased or do not reduced) of two kinds of recognition training texts of previous acquisition, so as to by each business mesh
Business model of the first type of current business model as each business objective is marked, the recognition rule of the first type of current business model is
The recognition rule of business model.
The recognition rule of the business model obtained by above-mentioned training should be corresponding comprising that can fully describe the business model
Business objective presence certain/keyword sequence of some logical relations, that is to say, that the recognition rule description of business model
The keyword and its logical relation being likely to occur in actual call other for certain specific transactions target class.
Business objective classification can refer to be expected that by one that the rule of corresponding business model can recognize that to lead to
Words belonging to attribute, the attribute can be Business Name, action type and customized thematic attribute etc. (as complain early warning or
Preferential activity in person campus etc.).
The specific example for stating a business objective classification and corresponding recognition rule is as follows：
Information service class->GPRS:((GPRS | flow | surfing Internets with cell phone | WAP)).
Embodiment two, the business model analytical equipment based on self-defined classifying rules, the structure of the device are as shown in Figure 2.
In Fig. 2, the device mainly includes：Acquisition module 1, identification module 2 and set up module 3.
Acquisition module 1 is mainly used in obtaining identification text.The identification text can be the knowledge converted by voice call
The voice of the current ongoing voice call of speech recognition submodule 11 pairs or recording in other text, such as acquisition module 1
Call carries out speech recognition, thus speech recognition submodule 11 can to obtain the voice call according to voice identification result corresponding
Identification text.Speech recognition submodule 11 can obtain identification text using existing speech recognition technology, no longer detailed herein
Describe in detail bright.
Identification module 2 is mainly used in the business model using each business objective being successfully established to above-mentioned acquisition module 1
The identification text for getting carries out business Classification and Identification.Identification module 2 can by the above-mentioned identification text for getting with succeeded
The business model of foundation carries out matching operation, and the business objective according to belonging to matching result determines the identification text, so that real
The business classification to recognizing text is showed.
One specific example, identification module 2 determines the keyword in identification text, and suitable according to the priority of business model
Sequence judge identification text in keyword meet business model recognition rule degree, identification module 2 can keyword with
When the matching degree of certain business model reaches predetermined extent, it is determined that current business model matches with identification text, and determine
The identification text belongs to the current corresponding business objective of business model, and identification module 2 no longer carries out follow-up deterministic process,
It is achieved thereby that the business classification to the identification text.In addition, in the above example, identification module 2 can also judge to know respectively
The matching degree of other text and all business models, and the best business model of matching degree is therefrom chosen, this is selected
The corresponding business objective of business model is used as the business objective belonging to identification text.Certainly, identification module 2 can also succeed
On the basis of the business model of foundation, business Classification and Identification is carried out to identification text using other modes, it is no longer detailed one by one herein
Describe in detail bright.
Module 3 is set up to be mainly used in setting up business model.The business model being successfully established by setting up module 3 can be stored in
Set up in module 3, it is also possible to be stored in identification module 2, can also be stored in independently of identification module 2 and set up module 3
In memory module.
Setting up module 3 mainly includes：Collect submodule 31, determination sub-module 32, submodule 33 is set and training submodule
Block 34.
Collect submodule 31 to be mainly used in, for an industry, collecting the business objective of the sector.Collecting submodule 31 can be with
All business objectives in the sector are pooled together in the form of business objective list.
Determination sub-module 32 is mainly used in determining that each business objective distinguishes corresponding association keyword.Determination sub-module 32 is
The association keyword that each business objective in the sector is set generally should be incomplete same.
Submodule 33 is set to be mainly used in using above-mentioned association keyword for each business objective is respectively provided with recognition rule, with
Set up the first type of business model of each business objective.Content, Yi Jishi that the recognition rule of the setting of submodule 33 mainly includes are set
The regulation that rule should not meet etc. the specific description as in above-mentioned embodiment one, is not repeated.Submodule is set
Recognition rule set by 33 can be specially：Information service class->GPRS:((GPRS | flow | surfing Internets with cell phone | WAP)).
After submodule 33 being set and is provided with recognition rule for each business objective, the first type of business model of each business objective
Also just it is successfully established.
Training submodule 34 is mainly used in entering the first type of business model of each business objective using multiple recognition training texts
Row training, to improve the recognition rule of each business objective, so as to be successfully established the business model of each business objective.
Above-mentioned training submodule 34 mainly includes：First module, second unit, Unit the 3rd, Unit the 4th and the 5th
Unit.
First module is mainly used in being identified multiple recognition training texts checking using the first type of single business model, with
Multiple recognition training texts are screened.
Second unit is mainly used according to the selection result amendment of the first module first type of the single business model.
Unit the 3rd is mainly used in repeating to call first module and second unit to perform corresponding operation, with one by one to institute
Just type is modified business model.
Unit the 4th is mainly used in entering multiple recognition training texts using the first type of current revised all business models
Row identification checking, to obtain two kinds of recognition training texts, i.e., can not be tested by the just type identification of current revised all business models
The recognition training text of card and meet the recognition training text of the first types of multiple current revised business models simultaneously, and according to
Both recognition training texts continue to correct corresponding service model just type.
Unit the 5th is mainly used in repeating to call Unit the 4th, until the current two kinds of quantity of recognition training text for obtaining
Quantity difference with two kinds of recognition training texts of previous acquisition is identical, so as to be successfully established the business model of each business objective.
Concrete operations performed by first module to Unit the 5th refer to the training step 1-5's in above-described embodiment one
Description, no longer describes in detail herein.
Each embodiment in this specification is described by the way of progressive, identical similar portion between each embodiment
Divide mutually referring to what each embodiment was stressed is the difference with other embodiment.Especially for device reality
Apply for example, because it is substantially similar to embodiment of the method, so describing fairly simple, related part is referring to embodiment of the method
Part explanation.Device embodiment described above is only schematical, wherein described illustrate as separating component
Module and unit can be or may not be physically separate, the part shown as unit can be or also may be used
Not being physical location, you can with positioned at a place, or can also be distributed on multiple NEs.Can be according to reality
The need for select some or all of module therein and realize the purpose of this embodiment scheme.Those of ordinary skill in the art exist
In the case of not paying creative work, you can to understand and implement.
The above is only presently preferred embodiments of the present invention, and any formal limitation is not made to the present invention, though
So the present invention is disclosed as above with preferred embodiment, but be not limited to technology of the invention, any to be familiar with this specialty
Technical staff without departing from the scope of the present invention, when using the technology contents of the disclosure above make it is a little change or
The Equivalent embodiments of equivalent variations are modified to, as long as being the content without departing from technical solution of the present invention, according to technology of the invention
Any simple modification, equivalent variations and modification that essence is made to above example, still fall within the model of technical solution of the present invention
In enclosing.
Claims (8)
1. a kind of business model analysis method based on self-defined classifying rules, it is characterised in that methods described includes：
Obtain identification text；
Business Classification and Identification is carried out to the identification text using the business model of each business objective being successfully established；
Wherein, the process of setting up of the business model includes：
For an industry, collect the business objective of the sector；
Determine that each business objective distinguishes corresponding association keyword；
Using the association keyword for each business objective is respectively provided with recognition rule, to set up the business model of each business objective
First type；
Just type is trained business model using multiple recognition training texts to each business objective, to improve each business mesh
Target recognition rule, so as to be successfully established the business model of each business objective；
It is described using multiple recognition training texts to the business model of each business objective just type be trained including：
Step 1, using single business model just type to multiple recognition training texts be identified checking, with to the multiple identification
Training text is screened；
Step 2, according to the modified result of the screening single business model just type；
Step 3, repeat the above steps 1-2, so that one by one to all business models, just type is modified；
Step 4, using current revised all business models, just type is identified checking to multiple recognition training texts, to obtain
Obtain two kinds of recognition training texts, i.e., can not be by the recognition training text of current revised all business models just type identification checking
And meet the recognition training text of the first types of multiple current revised business models simultaneously, and according to both recognition trainings text
The first type of this continuation amendment corresponding service model；
Step 5,4 are repeated the above steps, until two kinds of the current two kinds of quantity of recognition training text for obtaining and previous acquisition
The quantity difference of recognition training text is identical, so as to be successfully established the business model of each business objective.
2. analysis method as claimed in claim 1, it is characterised in that the recognition rule includes：Association keyword and pass
Join the logical relation between keyword, and the logical relation includes：With or, in non-and adjacent logical relation at least one
It is individual；Wherein, the adjacent logical relation refers to that the simultaneous two character pitch distances associated between keyword meet pre-
Fixed condition and the two association keywords there is predetermined tandem.
3. analysis method as claimed in claim 2, it is characterised in that the recognition rule should meet following conditions：
A, the symbol of characterization logic relation are half-angle character；
B, a recognition rule must have the DBC case for characterizing the DBC case for starting and characterizing end；
C, in a recognition rule, except non-sum adjoining logical relation in addition to other logical relations can continuously use；
D, in a recognition rule, other logical relations in addition to adjacent logical relation with mutually nested, and can be abutted
Logical relation can only be nested；
E, in a recognition rule, the priority of other logical relations in addition to NOT logic relation answers explicit use table
The symbol of priority is levied to represent；
F, bracket should occur in pairs；
There should be the symbol of characterization logic relation between G, adjacent association keyword before and after two.
4. the analysis method as described in claim 1 or 2 or 3, it is characterised in that the acquisition identification text includes：To voice
Call carries out speech recognition, to obtain the corresponding identification text of the voice call.
5. a kind of business model analytical equipment based on self-defined classifying rules, it is characterised in that described device includes：Obtain mould
Block, identification module and set up module；
The acquisition module, for obtaining identification text；
The identification module, industry is entered for the business model using each business objective being successfully established to the identification text
Business Classification and Identification；
Wherein, the business model is set up by the module of setting up, and the module of setting up includes：
Collect submodule, for for an industry, collecting the business objective of the sector；
Determination sub-module, for determining that each business objective distinguishes corresponding association keyword；
Submodule is set, for being respectively provided with recognition rule for each business objective using the association keyword, to set up each industry
The first type of the business model of target of being engaged in；
Training submodule, for the business model using multiple recognition training texts to each business objective, just type is instructed
Practice, to improve the recognition rule of each business objective, so as to be successfully established the business model of each business objective；
The training submodule includes：
First module, for just type to be identified checking to multiple recognition training texts using single business model, with to described
Multiple recognition training texts are screened；
Second unit, for the first type of the single business model of the modified result according to the screening；
Unit the 3rd, for repeating to call the first module and second unit, so that one by one just type enters to all business models
Row amendment；
Unit the 4th, for just type to be identified testing to multiple recognition training texts using current revised all business models
Card, to obtain two kinds of recognition training texts, i.e., can not be by the identification of current revised all business models just type identification checking
Training text and meet the recognition training text of the first types of multiple current revised business models simultaneously, and according to both knowledges
Other training text continues to correct corresponding service model just type；
Unit the 5th, for repeating to call Unit the 4th, until the current two kinds of quantity of recognition training text for obtaining with
The quantity difference of two kinds of recognition training texts of previous acquisition is identical, so as to be successfully established the business model of each business objective.
6. analytical equipment as claimed in claim 5, it is characterised in that the recognition rule includes：Association keyword and pass
Join the logical relation between keyword, and the logical relation includes：With or, in non-and adjacent logical relation at least one
It is individual；Wherein, the adjacent logical relation refers to that the simultaneous two character pitch distances associated between keyword meet pre-
Fixed condition and the two association keywords there is predetermined tandem.
7. analytical equipment as claimed in claim 6, it is characterised in that the condition that the recognition rule should meet includes：
A, the symbol of characterization logic relation are half-angle character；
B, a recognition rule must have the DBC case for characterizing the DBC case for starting and characterizing end；
C, in a recognition rule, except non-sum adjoining logical relation in addition to other logical relations can continuously use；
D, in a recognition rule, other logical relations in addition to adjacent logical relation with mutually nested, and can be abutted
Logical relation can only be nested；
E, in a recognition rule, the priority of other logical relations in addition to NOT logic relation answers explicit use table
The symbol of priority is levied to represent；
F, bracket should occur in pairs；
There should be the symbol of characterization logic relation between G, adjacent association keyword before and after two.
8. the analytical equipment as described in claim 5 or 6 or 7, it is characterised in that the acquisition module includes：
Speech recognition submodule, for carrying out speech recognition to voice call, to obtain the corresponding identification text of the voice call.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310589864.7A CN103699955B (en) | 2013-09-06 | 2013-11-20 | Business model analysis method and device based on self-defined classifying rules |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310401172.5 | 2013-09-06 | ||
| CN2013104011725 | 2013-09-06 | ||
| CN201310401172 | 2013-09-06 | ||
| CN201310589864.7A CN103699955B (en) | 2013-09-06 | 2013-11-20 | Business model analysis method and device based on self-defined classifying rules |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103699955A CN103699955A (en) | 2014-04-02 |
| CN103699955B true CN103699955B (en) | 2017-06-13 |
Family
ID=50361477
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310589864.7A Active CN103699955B (en) | 2013-09-06 | 2013-11-20 | Business model analysis method and device based on self-defined classifying rules |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103699955B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104064182A (en) * | 2014-06-24 | 2014-09-24 | 中国人民财产保险股份有限公司 | A voice recognition system and method based on classification rules |
| CN107680588B (en) * | 2017-05-10 | 2020-10-20 | 平安科技（深圳）有限公司 | Intelligent voice navigation method, device and storage medium |
| CN107391638A (en) * | 2017-07-10 | 2017-11-24 | 北京神州泰岳软件股份有限公司 | The new ideas of rule-associated model find method and device |
| CN107424601B (en) * | 2017-09-11 | 2023-08-08 | 深圳怡化电脑股份有限公司 | Information interaction system, method and device based on voice recognition |
| CN109508370B (en) * | 2018-09-28 | 2022-07-08 | 北京百度网讯科技有限公司 | Comment extraction method, comment extraction device and storage medium |
| CN109408637B (en) * | 2018-10-15 | 2021-12-07 | 苏州慧筑信息科技有限公司 | Method and system for automatically analyzing engineering list |
| CN111385273B (en) * | 2018-12-29 | 2022-07-01 | 中国移动通信集团北京有限公司 | Internet of things business process identification method and device, electronic equipment and medium |
| CN117440091A (en) * | 2023-09-22 | 2024-01-23 | 杭州东岸网络信息服务有限公司 | Voice data processing method and call control method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693244A (en) * | 2011-03-23 | 2012-09-26 | 日电（中国）有限公司 | Method and device for identifying information in non-structured text |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101447185B (en) * | 2008-12-08 | 2012-08-08 | 深圳市北科瑞声科技有限公司 | Audio frequency rapid classification method based on content |
| CN102402717A (en) * | 2010-09-13 | 2012-04-04 | 日电（中国）有限公司 | Data analysis facility and method |
| US8374865B1 (en) * | 2012-04-26 | 2013-02-12 | Google Inc. | Sampling training data for an automatic speech recognition system based on a benchmark classification distribution |
| CN102760436B (en) * | 2012-08-09 | 2014-06-11 | 河南省烟草公司开封市公司 | Voice lexicon screening method |
| CN102880649B (en) * | 2012-08-27 | 2016-03-02 | 北京搜狗信息服务有限公司 | A kind of customized information disposal route and system |
-
2013
- 2013-11-20 CN CN201310589864.7A patent/CN103699955B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693244A (en) * | 2011-03-23 | 2012-09-26 | 日电（中国）有限公司 | Method and device for identifying information in non-structured text |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103699955A (en) | 2014-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103699955B (en) | Business model analysis method and device based on self-defined classifying rules | |
| CN107329967B (en) | Question answering system and method based on deep learning | |
| CN103207855B (en) | For the fine granularity sentiment analysis system and method for product review information | |
| CN112214610A (en) | Entity relation joint extraction method based on span and knowledge enhancement | |
| CN106777232A (en) | Question and answer abstracting method, device and terminal | |
| CN105843875A (en) | Smart robot-oriented question and answer data processing method and apparatus | |
| CN109032591B (en) | Crowdsourcing software developer recommendation method based on meta-learning | |
| CN107103005A (en) | The collection method and device of question and answer language material | |
| CN106484916A (en) | A kind of enrollment answering method and device | |
| CN105912629A (en) | Intelligent question and answer method and device | |
| CN113468296B (en) | Model self-iteration type intelligent customer service quality inspection system and method capable of configuring business logic | |
| CN108388553A (en) | Talk with method, electronic equipment and the conversational system towards kitchen of disambiguation | |
| CN108009287A (en) | A kind of answer data creation method and relevant apparatus based on conversational system | |
| CN109800309A (en) | Classroom Discourse genre classification methods and device | |
| CN105183808A (en) | Problem classification method and apparatus | |
| CN108664237B (en) | It is a kind of based on heuristic and neural network non-API member's recommended method | |
| CN104064182A (en) | A voice recognition system and method based on classification rules | |
| CN108831229A (en) | A kind of Chinese automatic grading method | |
| CN111190973A (en) | Method, device, equipment and storage medium for classifying statement forms | |
| CN110516056A (en) | Interactive autonomous learning method, autonomous learning systems and storage medium | |
| CN110909132B (en) | Police service learning content analysis classifying method based on semantic analysis | |
| CN113962216A (en) | Text processing method and device, electronic equipment and readable storage medium | |
| CN114023355B (en) | Agent outbound quality inspection method and system based on artificial intelligence | |
| CN112015861A (en) | Intelligent test paper algorithm based on user historical behavior analysis | |
| CN114637849B (en) | Legal relation cognition method and system based on artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Wangjiang Road high tech Development Zone Hefei city Anhui province 230088 No. 666 Applicant after: Iflytek Co., Ltd. Address before: Wangjiang Road high tech Development Zone Hefei city Anhui province 230088 No. 666 Applicant before: Anhui USTC iFLYTEK Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200520 Address after: 200335 room 1966, floor 1, building 8, No. 33, Guangshun Road, Changning District, Shanghai Patentee after: IFLYTEK (Shanghai) Technology Co., Ltd Address before: Wangjiang Road high tech Development Zone Hefei city Anhui province 230088 No. 666 Patentee before: IFLYTEK Co.,Ltd. |
|
| TR01 | Transfer of patent right |