CN108255826A - A kind of common question and answer are to FAQ generation methods, apparatus and system - Google Patents
A kind of common question and answer are to FAQ generation methods, apparatus and system Download PDFInfo
- Publication number
- CN108255826A CN108255826A CN201611234261.5A CN201611234261A CN108255826A CN 108255826 A CN108255826 A CN 108255826A CN 201611234261 A CN201611234261 A CN 201611234261A CN 108255826 A CN108255826 A CN 108255826A
- Authority
- CN
- China
- Prior art keywords
- filling
- faq
- template
- keyword
- semantic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
Abstract
The present invention provides a kind of common question and answer to FAQ generation methods, apparatus and system, belongs to intelligent answer technical field, the method includes：Receive the filling content of the semantic template of current operation；According to the semantic template of the current operation and the filling content generation FAQ.FAQ generation methods provided by the invention, apparatus and system, utilize the FAQ of semantic template control generation standard, without being modified processing to off-gauge FAQ or user's question sentence, the FAQ of the standard can be preferably utilized analysis layer in analysis layer, and system is made to return to the matching degree higher of answer.
Description
Technical field
The present invention relates to intelligent answer technical field, more particularly to a kind of common question and answer to FAQ generation methods, device and
System.
Background technology
With increasing rapidly for Internet service, market is increasingly surging for the demand of intelligent answer, and each large enterprises are numerous and confused
Release the intelligent answer product of oneself.Such as the intelligent answer product Siri of Apple Inc..
Intelligent answer is a field of artificial intelligence.Intelligent answer is accurately positioned user's in the form of question-response
It puts question to, by being interacted with user, provides personalized information service to the user.
Intelligent Answer System generally comprises data Layer, analysis layer, access layer and application layer.Wherein, the corpus of data Layer
Basic data is very important, vital influence is played on the accuracy for finally returning that answer.Expect the important set in library
It is right therewith by a standard question sentence and one into part FAQ (Frequently Asked Questions, i.e., common question and answer to)
The standard response sentence composition answered.In the industry FAQ is established usually using the mode of manual maintenance typing.Manual maintenance typing is because of typing people
Member's language or the height of specialized capability are irregular, cause the FAQ of typing standardization, accuracy, can matching degree, completeization
On be unable to control, and then utilization of the impact analysis layer to it.
In recent years, in order to improve successful match rate, using the FAQ to having built up, established in a manner of manual entry with
The corresponding sentence pattern template of standard question sentence.One standard question sentence can correspond to one or more sentence pattern templates, but a sentence pattern template
At most it is only capable of a corresponding standard question sentence.The question sentence of user with sentence pattern template is matched, then passes through sentence pattern template and FAQ
Mapping, matches FAQ.Though this method can improve successful match rate to a certain extent, since the clause of user's question sentence is thousand changes
Ten thousand changes, matched mode directly is carried out to the question sentence of user in this way, template number can be caused very more.Therefore, this FAQ lifes
Into mode, user's question sentence can only be covered to a small extent, and there is significant limitation.
Invention content
The embodiment of the present invention the technical problem to be solved is that provide a kind of common question and answer to FAQ generation methods, device and
System in intelligent Answer System, to provide the mode of a set of generation standardization FAQ, makes it preferably sharp in analysis layer
With so as to improve the quality for returning to answer.
The present invention solves above-mentioned technical problem by the following technical programs：
One aspect of the present invention provides a kind of common question and answer to FAQ generation methods, including：
Receive the filling content of the semantic template of current operation；
According to the semantic template of the current operation and the filling content generation FAQ.
Further, the method further includes：REST interface generative semantics templates are transmitted according to declarative state.
Further, the filling content includes filling keyword and filling answer, and the filling keyword includes preposition
Word, subject term and postposition.
Further, it is further included after the filling content for receiving current operation semantic template：
Replacing for the filling keyword is extracted in keywords database according to the corresponding part of speech of the filling keyword and meaning
Change word；
The semantic template and the filling content generation FAQ according to the current operation includes：
According to the semantic template of the current operation and substitute generation FAQ.
Further, the semantic template includes：Conceptual illustration template, attribute form template, behavior template and row
For ground point template；
The corresponding part of speech of keyword of the conceptual illustration template is noun, and the keyword that the attribute forms template corresponds to
Part of speech for noun, the corresponding part of speech of key item of the behavior template is verb, the behavior point template key
The corresponding part of speech of item is adjective.
Further, it is described to be included according to the current operation semantic template with the filling content generation FAQ：
When receiving preservation instruction, FAQ will be generated in the filling content storage to corresponding semantic template；
Alternatively, the filling content is obtained every preset time, by the filling fills to corresponding semantic modules
Middle generation FAQ.
Another aspect of the present invention also provides a kind of common question and answer to FAQ generating means, including：
Receiving module, for receiving the filling content of the semantic template of current operation；
Generation module, for the semantic template according to the current operation and the filling content generation FAQ.
Further, the filling content includes filling keyword and filling answer, and the filling keyword includes preposition
Word, subject term and postposition.
Described device further includes：
Extraction module, for filling out described in being extracted in keywords database according to the corresponding part of speech of the filling keyword and meaning
Fill the substitute of keyword.
Further, described device further includes：
Semantic template generation module, for transmitting REST interface generative semantics templates according to declarative state.
Another aspect of the present invention also provides a kind of common question and answer and system, any of the above-described FAQ generation dresses is generated to FAQ
It puts.
Compared with prior art, common question and answer provided in an embodiment of the present invention are to FAQ generation methods, apparatus and system, profit
With the FAQ of semantic template contral generation standard, analysis layer, should without being modified processing to off-gauge FAQ or user's question sentence
The FAQ of standard can be preferably utilized in analysis layer, and system is made to return to the matching degree higher of answer.
Description of the drawings
Fig. 1 shows a kind of common question and answer of the offer of the embodiment of the present invention one to FAQ generation method flow charts；
Fig. 2 shows the schematic diagrames that the embodiment of the present invention one shows semantic template；
Fig. 3 shows that the embodiment of the present invention one shows the schematic diagram two of the corresponding filling content of the semantic template of user's selection；
Fig. 4 shows that the embodiment of the present invention one shows the schematic diagram two of the corresponding filling content of the semantic template of user's selection；
Fig. 5 shows a kind of common question and answer provided by Embodiment 2 of the present invention to FAQ generation method flow charts；
Fig. 6 shows that the embodiment of the present invention two shows the schematic diagram of semantic template；
Fig. 7 shows that the embodiment of the present invention two shows the schematic diagram of substitute；
Fig. 8 shows a kind of common question and answer of the offer of the embodiment of the present invention three to FAQ generating means structure diagrams.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, the technical solution in the present invention is carried out below
Clear, complete description, it is clear that described embodiment is part of the embodiment of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are not under the premise of creative work is made, the institute that is obtained
There is other embodiment, shall fall within the protection scope of the present invention.
Embodiment one
The embodiment of the present invention one provides a kind of common question and answer to FAQ generation methods, as shown in Figure 1, this method includes：
S101, receive current operation semantic template filling content；
Wherein, semantic template is the template for the question sentence for representing a certain type.In the present embodiment, the meeting in system initialization
4 semantic templates are automatically generated, are respectively：Conceptual illustration template --- [CONCEPT] is that, attribute form template ---
[CONCEPT] has which [ATTRIBUTE], behavior template --- [CONCEPT] how [ACTION], behavior place mould
Plate --- [CONCEPT] where [ACTION] most [ADJECTIVE].
In this step, interactive interface is accessed by network address according to user's operation, such as user, then shows all semantic moulds
Plate, each semantic template are equipped with sub-project number.Such as having+number shown in Fig. 2 on the right side of each semantic template, user clicks some language
After the sub-project number of adopted template, the corresponding filling content of the semantic template can be entered, FAQ is generated by filling content.
S102, the semantic template according to the current operation and the filling content generation FAQ；
Wherein, filling content includes filling keyword and filling answer.
In this step, display according to fig. 2 is as a result, user can select the corresponding sub-project number of semantic template.Such as with
Click the conceptual illustration in Fig. 2 in family --- [CONCEPT] be on the right side of what+number, then right side shows that conceptual illustration template is filled out
Content is filled, that is, fills keyword and filling answer, the results are shown in Figure 3.Wherein, filling keyword may further include master
Word, preposition and postposition are filled for user, can make the positioning of keyword more accurate in this way.User fills keyword and answers
User is respectively obtained after case and fills keyword and user's filling answer, generation filling content.It is filled out by filling keyword and user
A FQA can be generated by filling answer.
Further, it is also possible to once generate multiple FAQ.For example, on the basis of Fig. 3, user can continue to click attribute structure
Into the right side of template+number, as shown in figure 4, can show simultaneously multiple semantic templates to filling content, realize one secondary
Into multiple FAQ.
In this step, can FAQ be manually generated by user, it can also system automatically generated FAQ.
Wherein, the manual generating mode of user is：
When receiving preservation instruction, FAQ will be generated in the filling content storage to corresponding semantic template；
System automatically generated mode is：
The filling content is obtained every preset time, will be generated in the filling fills to corresponding semantic modules
FAQ。
Specifically, system can obtain filling content according to knowledge mapping.
Common question and answer provided in an embodiment of the present invention utilize semantic template control generation standard to FAQ generation methods
FAQ, for analysis layer without being modified processing to off-gauge FAQ or user's question sentence, the FAQ of the standard can be more preferable in analysis layer
Be utilized, make system return answer matching degree higher
Embodiment two
The embodiment of the present invention two provides a kind of common question and answer to FAQ generation methods, as shown in figure 5, this method includes：
S501, according to REST interface generative semantics templates.
REST (Representational State Transfer, declarative state are transmitted).In this step, Ke Yigen
Other other than 4 semantic templates automatically generated when needing and initialized by REST interface generation steps S101 according to user are semantic
Template.Such as can be needed to generate advantage template --- what advantage [CONCEPT] has according to user.
Specifically, if desired increase other templates except the semantic template that four initialization generate, such as inquiry advantage
Template can operate in accordance with the following steps.
The rest interface generative semantics templates provided by semantic template generation module, [CONCEPT] has any advantage.
Step 2, foreground interactive interface is accessed, it is found that below original already present 4 semantic templates, had more one just
Just newly-increased semantic template；
Step 3, click advantage --- [CONCEPT] have on the right side of what advantage+number；
Step, 4, preposition, CONCEPT keywords, postposition, answer are inserted in the input frame of response respectively；
Step 5, the save button in the lower right corner is clicked, completes to increase standard FAQ newly.
S502, display semantic template；
This step increases the semantic template of display S501 generations on the basis of S101, and display mode is identical with S101, no longer
It repeats, display result such as Fig. 6.
S503, the corresponding filling content of semantic template of display user's selection；
This step is identical with S102, repeats no more.
S504, replacing for the filling keyword is extracted in keywords database according to the corresponding part of speech of filling keyword and meaning
Change word.
The display of content is filled by S503, user can fill keyword in key word item.According to institute in this step
It states the corresponding part of speech of filling keyword and meaning extracts the substitute that user fills keyword in keywords database.Wherein, it is described
The corresponding part of speech of filling keyword of conceptual illustration template is noun, and the attribute forms the corresponding word of filling keyword of template
Property be noun, the corresponding part of speech of filling keyword of the behavior template is verb, the behavior point template filling
The corresponding part of speech of keyword is adjective.Such as：User inputs QQ in conceptual illustration template CONCEPT, then according to noun part-of-speech
Substitute QQ stores, the QQ coin, QQ member etc. of QQ is extracted in keywords database with the meaning of QQ.
Substitute is found through the above way, can show that multiple words fill selection for user by the form of combobox.
Such as Fig. 7, user input QQ in keyword is filled, combobox shows QQ stores, the substitutes such as QQ coin, QQ member automatically, side
Just user chooses.
In one embodiment, once multiple FAQ can be increased newly for same subject term.The subject term of each semantic template,
That is [CONCEPT] of template beginning and his preposition and postposition, when replacing the word, do not need to user in multiple FAQ
In be consistent manually, system can automatic synchronization.It can operate in accordance with the following steps.
Step 1, foreground interactive interface is accessed；
Step 2, click conceptual illustration --- [CONCEPT] be on the right side of what+number, then click behavior ---
[CONCEPT] how on the right side of [ACTION]+number；
Step 3, in the CONCEPT input frames of conceptual illustration, arbitrary keyword, such as expenses standard are inputted；
Step 4, any position in interface in addition to conceptual illustration CONCEPT input frames is clicked, it can be seen that behavior side
The subject term of the FAQ of formula is replaced by expenses standard；Equally, if preposition and postposition have input and
Effect.
S505, the semantic template selected according to the user and user fill content generation FAQ.
This step is identical with S102, repeats no more.
Embodiment three
The embodiment of the present invention three provides a kind of common question and answer to FAQ generating means 8, as shown in figure 8, the device 8 includes：
Receiving module 81, for receiving the filling content of the semantic template of current operation；
Generation module 82, for the semantic template according to the current operation and the filling content generation FAQ.
Further, the filling content includes filling keyword and filling answer, and the filling keyword includes preposition
Word, subject term and postposition.
Described device further includes：
Extraction module 83, described in being extracted in keywords database according to the corresponding part of speech of the filling keyword and meaning
Fill the substitute of keyword.
Semantic template generation module 80, for according to REST interface generative semantics templates.
Example IV
The embodiment of the present invention four provide a kind of common question and answer to FAQ generate system, any one described in including embodiment three
Kind FAQ generating means.
It will be understood by those skilled in the art that it can realize these structure charts and/or block diagram with computer program instructions
And/or the combination of each frame and these structure charts and/or the block diagram in block diagram and/or flow chart in flow chart.It can incite somebody to action
These computer program instructions are supplied to the processor of programmable data processing method to generate machine, so as to pass through programmable number
It is performed to instruct according to the processor of processing method and creates the frame for being used to implement structure chart and/or block diagram and/or flow chart or more
The method specified in a frame.
It should be noted that：Each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference from other examples, and just to refer each other for identical similar part between each embodiment.
Although the preferred embodiment of the embodiment of the present invention has been described, those skilled in the art once know base
This creative concept can then make these embodiments other change and modification.So appended claims are intended to be construed to
Including preferred embodiment and fall into all change and modification of range of embodiment of the invention.
Detailed Jie has been carried out to FAQ generation methods, apparatus and system to a kind of common question and answer provided by the present invention above
It continues, specific case used herein is expounded the principle of the present invention and embodiment, and the explanation of above example is only
It is the method and its core concept for being used to help understand the present invention；Meanwhile for those of ordinary skill in the art, according to this hair
Bright thought, there will be changes in specific embodiments and applications, in conclusion the content of the present specification should not manage
It solves as limitation of the present invention.
Claims (10)
1. a kind of common question and answer are to FAQ generation methods, which is characterized in that including：
Receive the filling content of the semantic template of current operation；
According to the semantic template of the current operation and the filling content generation FAQ.
2. according to the method described in claim 1, it is characterized in that, the method further includes：It is transmitted according to declarative state
REST interfaces generate the semantic template.
3. according to the method described in claim 1, it is characterized in that, the filling content includes filling keyword and filling is answered
Case, the filling keyword include preposition, subject term and postposition.
4. the according to the method described in claim 3, it is characterized in that, filling content of the semantic template for receiving current operation
It further includes later：
The substitute of the filling keyword is extracted in keywords database according to the corresponding part of speech of the filling keyword and meaning；
The semantic template and the filling content generation FAQ according to the current operation includes：
According to the semantic template of the current operation and substitute generation FAQ.
5. according to the method described in claim 4, it is characterized in that, the semantic template includes：Conceptual illustration template, attribute structure
Into template, behavior template and behavior ground point template；
The corresponding part of speech of keyword of the conceptual illustration template is noun, and the attribute forms the corresponding word of keyword of template
Property be noun, the corresponding part of speech of keyword of the behavior template is verb, the behavior point template keyword pair
The part of speech answered is adjective.
6. according to any methods of claim 1-5, which is characterized in that it is described according to the current operation semantic template and
The filling content generation FAQ includes：
When receiving preservation instruction, FAQ will be generated in the filling content storage to corresponding semantic template；
Alternatively, the filling content is obtained every preset time, it will be raw in the filling fills to corresponding semantic modules
Into FAQ.
7. a kind of common question and answer are to FAQ generating means, which is characterized in that including：
Receiving module, for receiving the filling content of the semantic template of current operation；
Generation module, for the semantic template according to the current operation and the filling content generation FAQ.
8. device according to claim 7, which is characterized in that the filling content includes filling keyword and filling is answered
Case, the filling keyword include preposition, subject term and postposition；
Described device further includes：
Extraction module closes for extracting the filling in keywords database according to the corresponding part of speech of the filling keyword and meaning
The substitute of keyword.
9. device according to claim 7 or 8, which is characterized in that described device further includes：
Semantic template generation module, for transmitting REST interface generative semantics templates according to declarative state.
10. a kind of common question and answer generate system to FAQ, which is characterized in that including any FAQ generations of claim 7-9
Device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611234261.5A CN108255826A (en) | 2016-12-28 | 2016-12-28 | A kind of common question and answer are to FAQ generation methods, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611234261.5A CN108255826A (en) | 2016-12-28 | 2016-12-28 | A kind of common question and answer are to FAQ generation methods, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108255826A true CN108255826A (en) | 2018-07-06 |
Family
ID=62719398
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611234261.5A Pending CN108255826A (en) | 2016-12-28 | 2016-12-28 | A kind of common question and answer are to FAQ generation methods, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108255826A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109710747A (en) * | 2019-01-16 | 2019-05-03 | 北京猎户星空科技有限公司 | Information processing method, device and electronic equipment |
| WO2020010834A1 (en) * | 2018-07-13 | 2020-01-16 | 众安信息技术服务有限公司 | Faq question and answer library generalization method, apparatus, and device |
-
2016
- 2016-12-28 CN CN201611234261.5A patent/CN108255826A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020010834A1 (en) * | 2018-07-13 | 2020-01-16 | 众安信息技术服务有限公司 | Faq question and answer library generalization method, apparatus, and device |
| CN109710747A (en) * | 2019-01-16 | 2019-05-03 | 北京猎户星空科技有限公司 | Information processing method, device and electronic equipment |
| CN109710747B (en) * | 2019-01-16 | 2021-04-06 | 北京猎户星空科技有限公司 | Information processing method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6726800B2 (en) | Method and apparatus for human-machine interaction based on artificial intelligence | |
| US10891322B2 (en) | Automatic conversation creator for news | |
| US20160275075A1 (en) | Natural Expression Processing Method, Processing and Response Method, Device, and System | |
| TW201921267A (en) | Method and system for generating a conversational agent by automatic paraphrase generation based on machine translation | |
| US11775254B2 (en) | Analyzing graphical user interfaces to facilitate automatic interaction | |
| US11264006B2 (en) | Voice synthesis method, device and apparatus, as well as non-volatile storage medium | |
| CN109923558A (en) | Mixture of expert neural network | |
| CN106202059A (en) | Machine translation method and machine translation apparatus | |
| AU2015409645A1 (en) | Universal translation | |
| US11158349B2 (en) | Methods and systems of automatically generating video content from scripts/text | |
| CN107291680A (en) | A kind of system and implementation method that automatically generate composition based on template | |
| CN107766320A (en) | A kind of Chinese pronoun resolution method for establishing model and device | |
| US10496751B2 (en) | Avoiding sentiment model overfitting in a machine language model | |
| CN109214008A (en) | A kind of sentiment analysis method and system based on keyword extraction | |
| CN109272003A (en) | A kind of method and apparatus for eliminating unknown error in deep learning model | |
| Park et al. | Systematic review on chatbot techniques and applications | |
| US20190108773A1 (en) | Language fluency system | |
| JP6988924B2 (en) | Question group extraction method, question group extraction device and question group extraction program | |
| CN108255826A (en) | A kind of common question and answer are to FAQ generation methods, apparatus and system | |
| CN109190116A (en) | Semantic analytic method, system, electronic equipment and storage medium | |
| US20230317058A1 (en) | Spoken language processing method and apparatus, and storage medium | |
| US20230016962A1 (en) | Multilingual natural language understanding model platform | |
| CN110046346B (en) | Corpus intention monitoring method and device and terminal equipment | |
| CN113515932A (en) | Method, device, equipment and storage medium for processing question and answer information | |
| Xu et al. | Towards zero-shot persona dialogue generation with in-context learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180706 |