CN109829037A - Method, system, server and the storage medium of intelligent automatic question answering - Google Patents
Method, system, server and the storage medium of intelligent automatic question answering Download PDFInfo
- Publication number
- CN109829037A CN109829037A CN201711175236.9A CN201711175236A CN109829037A CN 109829037 A CN109829037 A CN 109829037A CN 201711175236 A CN201711175236 A CN 201711175236A CN 109829037 A CN109829037 A CN 109829037A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- knowledge point
- request message
- original request
- similarity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000013507 mapping Methods 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 238000004590 computer program Methods 0.000 claims description 2
- 230000015654 memory Effects 0.000 description 11
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of method and system of intelligent automatic question answering, this method comprises: receiving the original request message of user, the original request message includes the first example and third example；Knowledge point corresponding with the original request message is searched in knowledge base by similarity calculation；When not finding corresponding knowledge point in knowledge base, according to the knowledge mapping pre-established, processing is made inferences, the reasoning processing includes: that the first example is inquired in the knowledge mapping pre-established；Associated path of first example in knowledge mapping is obtained, until the associated path includes third example；Answer is determined according to first example, third example and the associated path, and the associated path includes M attribute information, and M is greater than or equal to 2；Determining answer is sent to user.The present invention can greatly improve the accuracy rate of intelligent automatic question answering.
Description
Technical field
The present embodiments relate to human-computer interaction technology more particularly to a kind of method, the intelligence of intelligent automatic question answering are automatic
System, server and the storage medium of question and answer.
Background technique
Human-computer interaction is the science of the interactive relation between a research system and user, and system can be various
Machine, be also possible to the system and software of computerization.Intelligent automatically request-answering system is exactly to rely on human-computer interaction technology hair
A kind of artificial intelligence system that exhibition is got up, for example, intelligent customer service system, speech control system etc..
Knowledge base is the basis of intelligent automatically request-answering system, and in knowledge base, information is effectively organized to be retrieved
And utilization.Multiple knowledge points are stored in knowledge base, each knowledge point includes one or more problems and corresponding answer.When
When user inputs solicited message, the semantic similarity of problem, is greater than if there is similarity in computation requests information and knowledge base
The corresponding answer of problem of wherein similarity maximum is then returned to user by the problem of the first preset threshold.
Generally, what is stored in knowledge base is all relatively simple knowledge point, such as:
Whom the former husband of knowledge point 1:A? B
Whom the existing girl friend of knowledge point 2:B? C
Whom the first husband of knowledge point 3:C? D
Whom the eldest daughter of knowledge point 4:D? E
When for solicited message are as follows: whens what or C the relationship of the relationship of A and D, A and E be and E is what relationship etc.,
According to existing question and answer technology, corresponding answer can not be just got in knowledge base at this time.
Summary of the invention
In order to solve the above-mentioned technical problem, the embodiment of the present invention provides a kind of method of intelligent automatic question answering, intelligence automatically
System, server and the storage medium of question and answer, to find challenge on the basis of not changing knowledge base stored knowledge point
Answer improves the accuracy rate of question and answer.
The embodiment of the invention provides a kind of methods of intelligent automatic question answering, comprising:
The original request message of user is received, the original request message includes the first example and third example；
Knowledge point corresponding with the original request message is searched in knowledge base by similarity calculation；
When not finding corresponding knowledge point in knowledge base, according to the knowledge mapping pre-established, place is made inferences
Reason, the reasoning processing include:
The first example is inquired in the knowledge mapping pre-established；
Associated path of first example in knowledge mapping is obtained, until the associated path includes third example；
Answer is determined according to first example, third example and the associated path, and the associated path includes M
Attribute information, M are greater than or equal to 2；
Determining answer is sent to user.
Optionally, the knowledge mapping records the incidence relation between multiple examples in table form.
Optionally, it is that the original request message includes: the relationship of the first example and third example for what or the first example
It is any relationship with third example.
Optionally, knowledge point packet corresponding with the original request message is searched in knowledge base by similarity calculation
It includes: calculating the semantic similarity for the problem of storing in the original request message and knowledge base, the maximum being calculated is similar
Degree is compared with preset first preset threshold, when the similarity being calculated is greater than first preset threshold, by this
The corresponding knowledge point of similarity is as the knowledge point searched；Otherwise, corresponding knowledge point is not found in knowledge base.
The embodiment of the invention also provides a kind of systems of intelligent automatic question answering, comprising: user's request receiving module is used
In the original request message for receiving user, the original request message includes the first example and third example；
Knowledge point searching module, for being searched in knowledge base and the original request message pair by similarity calculation
The knowledge point answered；
Reasoning module, for when not finding corresponding knowledge point in knowledge base, according to the knowledge graph pre-established
Spectrum, makes inferences processing, the reasoning module includes:
Query By Example unit, for inquiring the first example in the knowledge mapping pre-established；
Associated path acquiring unit, for obtaining associated path of first example in knowledge mapping, until described
Associated path includes third example；
Answer determination unit, for determining answer according to first example, third example and the associated path, institute
Stating associated path includes M attribute information, and M is greater than or equal to 2；
Output module, for the answer determined to be sent to user.
Optionally, the knowledge mapping records the incidence relation between multiple examples in table form.
Optionally, it is that the original request message includes: the relationship of the first example and third example for what or the first example
It is any relationship with third example.
Optionally, the knowledge point searching module calculates the problem of storing in the original request message and knowledge base
The maximum similarity being calculated is compared, when the phase being calculated by semantic similarity with preset first preset threshold
When being greater than first preset threshold like degree, using the corresponding knowledge point of the similarity as the knowledge point searched；Otherwise, do not exist
Corresponding knowledge point is found in knowledge base.
The embodiment of the invention also provides a kind of servers, comprising:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of places
The method that reason device realizes above-mentioned intelligent automatic question answering.
The embodiment of the invention also provides a kind of computer storage mediums, are stored thereon with computer program, the program quilt
The method that processor realizes above-mentioned intelligent automatic question answering when executing.
The technical solution of the embodiment of the present invention, it is corresponding with original request message by that cannot be found in knowledge base
When knowledge point, make inferences processing to obtain the solicited message after reasoning, and is searched in knowledge base by similarity calculation and
The corresponding knowledge point of solicited message after reasoning constantly repeats above-mentioned when not finding corresponding knowledge point in knowledge base
The process of reasoning and calculating realizes until finding corresponding knowledge point in knowledge base and is not changing knowledge base storage
The answer that challenge is found on the basis of knowledge point improves question and answer under the premise of not changing knowledge base storage content
Accuracy rate saves memory space.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the method for intelligent automatic question answering that the embodiment of the present invention one provides；
Fig. 2 is a kind of structural schematic diagram of the system of intelligent automatic question answering provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of structural schematic diagram for server that the embodiment of the present invention three provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to
Convenient for describing, only some but not all contents related to the present invention are shown in the drawings.
The scheme of embodiment to facilitate the understanding of the present invention first does lower letter to the basic knowledge of intelligent automatically request-answering system
It is single to introduce:
One, knowledge point
Basic knowledge point most original and simplest form in knowledge base are exactly usually common FAQ, general form
It is that " ask-answer " is right.For example, " rate of CRBT " are exactly that clearly standard asks description for expression.Here " asking " should not be by narrowly
It is interpreted as " inquiring ", and should broadly understand one " input ", being somebody's turn to do " input " has corresponding " output ".For example, for being used for
For the semantics recognition of control system, the instruction of user, such as " opening radio " should also be understood to be one
" asking ", corresponding at this time " answering " can be the calling for executing the control program accordingly controlled.
User to machine when inputting, the most ideal situation is that asked using standard, then the intelligent semantic identifying system of machine
At once it will be appreciated that the meaning of user.However, user often not uses standard to ask, but some deformations that standard is asked
Form.For example, if being " changing a radio station " for the standard form of asking of the radio station switching of radio, then user may make
Order is " switching one radio station ", and what machine was also required to can to identify user's expression is the same meaning.
Therefore, for intelligent semantic identification, the extension that the standard that needs in knowledge base is asked is asked, which asks and mark
Standard asks that expression-form has slight difference, but expresses identical meaning.
It therefore, include multiple knowledge points in knowledge base, each knowledge point includes problem and answer, and problem includes that standard is asked
It is asked with multiple extensions.
The problems in knowledge point generally indicates that here is the symbol in semantic formula with the form of semantic formula:
1.1.1 the expression ([]) of part of speech
To distinguish the word in expression formula with part of speech, it is specified that part of speech must be present in square brackets " [] ", occur in square brackets
Part of speech be generally " narrow sense part of speech ", but " broad sense part of speech " can also be supported by configuring system parameter.Here is some letters
The example of single expression formula: [Fetion] [how] [open-minded], [introduction] [multimedia message] [business], [login] [method] of [Fetion],
[call reminding] [how] [charge].
Or the expression (|) of relationship 1.1.2
Part of speech in square brackets can occur repeatedly by "or" relationship, and the part of speech of these "or" relationships can calculate
It is individually calculated in a manner of " expansion " when similarity." expansion " is mainly according to the meaning of "or" by semantic formula exhibition
It is split into the process of multiple structures.As: [CRBT] [open-minded] [method | step] it is deployable at " [CRBT] [open-minded]
[step] " and " [method] of [CRBT] [open-minded] " two simple semantic formulas.The example of this kind of semantic formula is such as
Under: [how] [inquiry | know] [PUK code], [quit the subscription of | revocation | close | deactivate] [IP | 17951] [the preferential packet of national distance].
1.1.3 it is non-essential indicate (?)
Part of speech in square brackets can be added at the end of "? " expression, which may occur in which, to be occurred, i.e., non-essential pass
System, the part of speech of this inessential relationship similarly can individually be calculated in a manner of " expansion " when calculating similarity." exhibition
Open " it is mainly that will be launched into include and do not include containing non-essential part of speech (or " or combination " of part of speech) in semantic formula
The process of the simple semantic formula of two of this part of speech.
Such as: [introduction] [mobile video] [military column] [content] [what? ] deployable at " [introduction] [mobile video]
[military column] [content] " and " [introduction] [mobile video] [military column] [content] [what] " two simple semantic meaning representations
Formula.
Two, ontology
Ontology (Ontology) is the Formal Representation to certain set concept and its mutual relationship among specific area,
It is the abstract to a field.We obtain the knowledge in a certain field using ontology, the concept in the ontology describing field,
And the relationship between these concepts.
The composition of 2.1 ontologies
Ontology includes ontology class and example, and ontology class defines the structure of corresponding example, and example is to meet ontology class to determine
The individual subject of justice.
2.1.1 ontology class (Ontology Class)
In real world, often has and belong to the object of same class (such as: certain automobile only many automobiles in the world
In one), also have the different objects of many shared same characteristic features.The same characteristic features that can use these objects are them
Establish a prototype.Ontology class (Ontology Class) is exactly a kind of definition of this prototype, it is to common trait
Or all individuals of shared common trait is abstract.The feature of ontology class is known as generic attribute (Class Property), such as
" people " this ontology class has attributes such as " names ", " age " and " weight ".Generic attribute is by attribute-name and one group of semantic formula institute
It constitutes.In ontology class, we can describe semanteme expressed by attribute, the semanteme of generic attribute with one group of semantic formula
Expression formula is made of example symbol, part of speech, grammatical symbol, and example symbol is for representing the part of speech or semantic chunk of description individual.
2.1.2 example (Instance)
Example is the product after the instantiation of ontology class, represents and meets the specific individual subject that ontology class defines.
Example is made of instance name, example semantic, attribute and attribute value.Example semantic: example semantic is for describing example semantic
Semantic formula, which replaces in ontology generic attribute semantic formula during instantiating ontology class
" example symbol ", and then generate the semanteme of instance properties.
Instance properties: instance properties include general property and Custom Attributes.General property is the ontology for instantiating example
Attribute defined in class, such as " car Santana " this example is that this ontology class instantiates to obtain by " car ", then " sedan-chair
The attributes such as " wheelbase " " weight " and " discharge capacity " are general property defined in this ontology class of vehicle ".Custom Attributes be in order to
A more complete description is individual and the customized feature for being appropriate only for the individual describes except general property.
Attribute value: attribute value, that is, attribute specific descriptions, the matter or amount of the attribute for exact expression example.Such as " male
Property " is exactly the attribute value of " gender " attribute in " Zhang San " this example.
Embodiment one
Fig. 1 is a kind of flow chart of the method for intelligent automatic question answering that the embodiment of the present invention one provides, and the present embodiment can fit
For searching the corresponding knowledge point of user request information for including the case where two examples, this method can be asked automatically by intelligence
The system answered executes, which can be realized by software and/or hardware, can be generally integrated in the equipment such as server, the party
Method specifically comprises the following steps:
Step 210, the original request message of user is received, the original request message includes that the first example and third are real
Example.
The original request message of user includes two examples, if original request message is " relationship of first and second ", wherein
First is the first example, and second is third example；Alternatively, first is third example, second is the first example.
Specifically, the original request message may include: the relationship of the first example and third example be what or first
Example and third example are what relationships etc..
Step 220, knowledge point corresponding with the original request message is searched in knowledge base by similarity calculation.
The semantic similarity for calculating the problem of storing in the original request message and knowledge base, will be calculated most
Big similarity is compared with preset first preset threshold, to search knowledge point corresponding with original request message.
Specifically, knowledge point packet corresponding with the original request message is searched in knowledge base by similarity calculation
It includes: calculating the semantic similarity for the problem of storing in the original request message and knowledge base, the maximum being calculated is similar
Degree is compared with preset first preset threshold, when the similarity being calculated is greater than first preset threshold, by this
The corresponding knowledge point of similarity is as the knowledge point searched；Otherwise, corresponding knowledge point is not found in knowledge base.
Step 230, when not finding corresponding knowledge point in knowledge base, according to the knowledge mapping pre-established, into
Row reasoning processing.
When the similarity being calculated is respectively less than the first preset threshold, determine in knowledge base without directly storage with it is initial
At this moment the corresponding knowledge point of solicited message can make inferences processing according to the knowledge mapping pre-established.The knowledge graph
Spectrum includes the incidence relation between multiple examples and example in knowledge base.
The reasoning is handled
Step 231, the first example is inquired in the knowledge mapping pre-established.
Specifically, knowledge mapping can be in table form to record the incidence relation between multiple examples.Carry out
When reasoning is handled, the first example is first looked for, to get the associated path of the first example Yu third example.
Step 232, associated path of first example in knowledge mapping is obtained, until the associated path includes
Third example.
Wherein, associated path is the path of multiple examples in association knowledge map, including at least two examples, and
The knowledge point of at least two examples is respectively included, a knowledge point includes problem and answer, and wherein problem includes an example,
It include another example in answer, the knowledge point that an example is respectively included in problem and answer is exactly to divide in problem and answer
The associated path for the example for not including.For example, the corresponding knowledge point of example 1 includes knowledge point 1.1,1.2 etc., knowledge point 1.1
Example 2 is quoted in answer, and the corresponding knowledge point of example 2 includes knowledge point 2.1,2.2 etc., and example 3 is quoted in the answer of knowledge point 2.2,
Then the incidence relation between example 1 and example 2 is the associated path of example 1 and example 2, which includes example 1, knows
Know point 1.1 and example 2, the incidence relation between example 1, example 2 and example 3 be the associated path of example 1 and example 3,
It include example 1, knowledge point 1.1, example 2, knowledge point 2.2 and example 3 i.e. in the associated path.
Step 233, answer is determined according to first example, third example and the associated path.
Each knowledge point only stores an attribute information between two examples, the association road in the present embodiment knowledge base
Diameter includes M attribute information, and M is greater than or equal to 2, to compensate for the defect that existing knowledge library only stores simple knowledge point.
Answer corresponding with the original request message of user can be determined according to the first example, third example and associated path
The answer is returned to the user by case.
Text answers both can be directly returned to user by the present embodiment, and text answers can also be converted to voice lattice
User is issued after formula, is not limited the scope of the invention.
For example, the original request message of user is " Huawei P9 be what relationship with Mr. Ren Zhengfei ", initial request letter
Include the first example " Huawei P9 " and third example " Mr. Ren Zhengfei " in breath, calculates in the original request message and knowledge base
The problem of semantic similarity, the similarity being calculated is respectively less than the first preset threshold, shows to search in knowledge base
To knowledge point corresponding with original request message, at this moment enter the reasoning process of knowledge based map: in the knowledge pre-established
It is inquired in map the first example " Huawei P9 ", obtains the associated path of the first example " Huawei P9 " in knowledge mapping, obtain
The process of associated path are as follows: there is knowledge point 1.1 and 1.2 in the corresponding knowledge point of the first example " Huawei P9 ", wherein knowledge point 1.1
The problem of be " what the CPU model of Huawei P9 is ", answer is the second example of reference " sea think kylin 955 ", then searches second
The corresponding knowledge point of example " sea think kylin 955 ", finds knowledge point 2.1, wherein the problem of " to think the production of kylin 955 in sea
Whom quotient is ", answer is to quote the 4th example " HiSilicon Technologies Co., Ltd. ", in the corresponding knowledge point of the 4th example not
Other examples are quoted, therefore do not find the associated path including third example " Mr. Ren Zhengfei "；Wherein knowledge point 1.2
The problem of be " whom the manufacturer of Huawei P9 is ", answer be reference the 5th example " Huawei Tech Co., Ltd ", the 5th example
There is knowledge point 5.1 in " Huawei Tech Co., Ltd " corresponding knowledge point, wherein for " Huawei Technologies are limited the problem of knowledge point 5.1
Whom the president of company is ", answer is reference third example " Mr. Ren Zhengfei ", which includes that third example " is appointed just
Non- Mr. ", i.e., including third example " Mr. Ren Zhengfei " associated path include the first example, knowledge point 1.2, the 5th example,
Knowledge point 5.1 and third example；Therefore, determine that answer is that " Ren Zhengfei is first according to the first example, third example and associated path
Life is the president of the manufacturer of Huawei P9 ", which is returned into user.
It include the in the original request message for another example the original request message of user is " first and second be what relationship "
One example " first " and third example " second ", calculate the semantic similarity of the problems in the original request message and knowledge base, calculate
Obtained similarity is respectively less than the first preset threshold, shows to find in knowledge base corresponding with original request message
At this moment knowledge point enters the reasoning process of knowledge based map: inquiring the first example in the knowledge mapping pre-established
" first " obtains the associated path of the first example " first " in knowledge mapping, obtains the process of associated path are as follows: the first example
" first " corresponding knowledge point has 1.1, is wherein the problem of knowledge point 1.1 " whom the former husband of first is ", and corresponding answer is reference the
The problem of two examples " third ", then inquire the corresponding knowledge point of the second example " third " and have 2.1, knowledge point 2.1 is " third existing female
Whom friend is ", corresponding answer is the 4th example " fourth " of reference, then inquires the corresponding knowledge point of the 4th example " fourth " and have 4.1,
The problem of knowledge point 4.1 is " whom the daughter of fourth is ", and answer is reference third example " second ", is found including third example " second "
Associated path, i.e., including third example " second " associated path include the first example " second ", knowledge point 1.1, the second example
" third ", knowledge point 2.1, the 4th example " fourth ", knowledge point 4.1 and third example " second "；Therefore, real according to the first example, third
Example and associated path determine that answer is " daughter that second is the existing girl friend of the former husband of first ", which is returned to user.
The technical solution of the present embodiment includes the first example and third example by the original request message in user, and
When cannot be searched in knowledge base to knowledge point corresponding with original request message by similarity calculation, according to pre-establishing
Knowledge mapping, make inferences processing, the reasoning processing includes: that the first example is inquired in the knowledge mapping pre-established,
Associated path of first example in knowledge mapping is obtained, until the associated path includes third example, it is real according to first
Example, third example and associated path determine answer, realize and find packet on the basis of not changing the knowledge point of knowledge base storage
The corresponding answer of solicited message for including two examples improves the standard of question and answer under the premise of not changing knowledge base storage content
True rate, saves memory space.
Embodiment two
Fig. 2 is a kind of structural schematic diagram of the system of intelligent automatic question answering provided by Embodiment 2 of the present invention, the present embodiment
Be applicable to search and include the case where the corresponding knowledge point of the user request information of two examples, the system can by software with/
Or hardware realizes that can generally be integrated in the equipment such as server, which includes: user's request receiving module 410, knowledge
Point searching module 420, reasoning module 430 and output module (not shown).
Wherein, user's request receiving module 410, for receiving the original request message of user, the original request message
Including the first example and third example；
Knowledge point searching module 420, for being searched in knowledge base and the original request message by similarity calculation
Corresponding knowledge point；
Reasoning module 430, for being known according to what is pre-established when not finding corresponding knowledge point in knowledge base
Know map, makes inferences processing, reasoning module 430 includes:
Query By Example unit 431, for inquiring the first example in the knowledge mapping pre-established；
Associated path acquiring unit 432, for obtaining associated path of first example in knowledge mapping, until
The associated path includes third example；
Answer determination unit 433, for determining answer according to first example, the second example and the associated path,
The associated path includes M attribute information, and M is greater than or equal to 2.
Text answers both can be directly returned to user by output module in the present embodiment, text answers can also be turned
User is issued after being changed to phonetic matrix, is not limited the scope of the invention.
Specifically, the knowledge mapping can record the incidence relation between multiple examples in table form.
Specifically, the original request message may include: the relationship of the first example and third example be what or first
Example and third example are what relationships etc..
Specifically, the knowledge point searching module calculates the problem of storing in the original request message and knowledge base
The maximum similarity being calculated is compared, when the phase being calculated by semantic similarity with preset first preset threshold
When being greater than first preset threshold like degree, using the corresponding knowledge point of the similarity as the knowledge point searched；Otherwise, do not exist
Corresponding knowledge point is found in knowledge base.
The technical solution of the present embodiment receives the original request message of user by user's request receiving module, described first
Beginning solicited message includes the first example and third example；Knowledge point searching module is searched in knowledge base by similarity calculation
Knowledge point corresponding with original request message；When not finding corresponding knowledge point in knowledge base, reasoning module is according to pre-
The knowledge mapping first established makes inferences processing；The reasoning module includes: Query By Example unit, for pre-establishing
The first example is inquired in knowledge mapping；Associated path acquiring unit, for obtaining association road of first example in knowledge mapping
Diameter, until the associated path includes third example；Answer determination unit, for according to the first example, third example and association
Path determines answer, realizes and finds the request including two examples on the basis of not changing the knowledge point of knowledge base storage
The corresponding answer of information improves the accuracy rate of question and answer under the premise of not changing knowledge base storage content, saves storage
Space.
Embodiment three
Fig. 3 is a kind of structural schematic diagram for server that the embodiment of the present invention three provides, as shown in figure 3, the server packet
Include processor 510, storage device 520, input unit 530 and output device 540；The quantity of processor 510 can be in server
Be it is one or more, in Fig. 3 by taking a processor 510 as an example；Processor 510, storage device 520, input dress in server
Setting 530 can be connected with output device 540 by bus or other modes, in Fig. 3 for being connected by bus.
Storage device 520 is used as a kind of computer readable storage medium, and can be used for storing software program, computer can hold
Line program and module, such as the corresponding program instruction/module of the method for the intelligent automatic question answering in the embodiment of the present invention one.Place
Software program, instruction and the module that reason device 510 is stored in storage device 520 by operation, thereby executing terminal device
The method of above-mentioned intelligent automatic question answering is realized in various function application and data processing.
Storage device 520 can mainly include storing program area and storage data area, wherein storing program area can store behaviour
Application program needed for making system, at least one function；Storage data area, which can be stored, uses created data according to terminal
Deng.In addition, storage device 520 may include high-speed random access memory, it can also include nonvolatile memory, such as
At least one disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, storage dress
Setting 520 can further comprise the memory remotely located relative to processor 510, these remote memories can pass through network
It is connected to terminal device.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, mobile radio communication
And combinations thereof.
Input unit 530 can be used for receiving the number or character information of input, and generates and set with the user of terminal device
It sets and the related key signals of function control inputs.Output device 540 may include that display screen etc. shows equipment.
Example IV
The embodiment of the present invention four provides a kind of storage medium comprising computer executable instructions, and the computer can be held
A kind of method of the row instruction when being executed by computer processor for executing intelligent automatic question answering, this method specifically please refer to
Embodiment one, details are not described herein.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but in many cases, the former is more
Good embodiment.Based on this understanding, technical solution of the present invention substantially in other words contributes to the prior art
Part can be embodied in the form of software products, which can store in computer-readable storage medium
Floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (Random in matter, such as computer
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes method described in each embodiment of the present invention.
It is worth noting that, included each unit and module are only patrolled according to function in the embodiment of above-mentioned apparatus
It volume is divided, but is not limited to the above division, as long as corresponding functions can be realized；In addition, each function list
The specific name of member is also only for convenience of distinguishing each other, the protection scope being not intended to restrict the invention.
Note that the above is only a better embodiment of the present invention and the applied technical principle.Those skilled in the art can manage
Solution, the invention is not limited to the specific embodiments described herein, is able to carry out various apparent changes for a person skilled in the art
Change, readjust and substitutes without departing from protection scope of the present invention.Therefore, although by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, the case where not departing from present inventive concept
Under, it can also include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of method of intelligence automatic question answering characterized by comprising
The original request message of user is received, the original request message includes the first example and third example；
Knowledge point corresponding with the original request message is searched in knowledge base by similarity calculation；
When not finding corresponding knowledge point in knowledge base, according to the knowledge mapping pre-established, processing, institute are made inferences
Stating reasoning processing includes:
The first example is inquired in the knowledge mapping pre-established；
Associated path of first example in knowledge mapping is obtained, until the associated path includes third example；
Answer is determined according to first example, third example and the associated path, and the associated path includes M attribute letter
Breath, M are greater than or equal to 2；
Determining answer is sent to user.
2. the method according to claim 1, wherein the knowledge mapping records multiple examples in table form
Between incidence relation.
3. the method according to claim 1, wherein the original request message includes: the first example and third
What or the first example the relationship of example be and what relationship is third example be.
4. the method according to claim 1, wherein at the beginning of being searched in knowledge base by similarity calculation and be described
The corresponding knowledge point of beginning solicited message includes: that the calculating original request message is similar to semanteme the problem of storage in knowledge base
Degree, the maximum similarity being calculated is compared with preset first preset threshold, when the similarity being calculated is greater than
When first preset threshold, using the corresponding knowledge point of the similarity as the knowledge point searched；Otherwise, it is not looked into knowledge base
Find corresponding knowledge point.
5. a kind of system of intelligence automatic question answering characterized by comprising user's request receiving module, for receiving user's
Original request message, the original request message include the first example and third example；
Knowledge point searching module corresponding with the original request message is known for being searched in knowledge base by similarity calculation
Know point；
Reasoning module, for when not finding corresponding knowledge point in knowledge base, according to the knowledge mapping pre-established, into
Row reasoning processing, the reasoning module include:
Query By Example unit, for inquiring the first example in the knowledge mapping pre-established；
Associated path acquiring unit, for obtaining associated path of first example in knowledge mapping, until the association
Path includes third example；
Answer determination unit, for determining answer, the association according to first example, third example and the associated path
Path includes M attribute information, and M is greater than or equal to 2；
Output module, for the answer determined to be sent to user.
6. system according to claim 5, which is characterized in that the knowledge mapping records multiple examples in table form
Between incidence relation.
7. system according to claim 5, which is characterized in that the original request message includes: the first example and third
What or the first example the relationship of example be and what relationship is third example be.
8. system according to claim 5, which is characterized in that the knowledge point searching module calculates the initial request letter
Breath and semantic similarity the problem of storage in knowledge base, by the maximum similarity being calculated and preset first preset threshold
Be compared, when the similarity being calculated be greater than first preset threshold when, using the corresponding knowledge point of the similarity as
The knowledge point of lookup；Otherwise, corresponding knowledge point is not found in knowledge base.
9. a kind of server, which is characterized in that the server includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method of the intelligent automatic question answering as described in any in claim 1-4.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
The method of intelligent automatic question answering of the Shi Shixian as described in any in claim 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175236.9A CN109829037A (en) | 2017-11-22 | 2017-11-22 | Method, system, server and the storage medium of intelligent automatic question answering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175236.9A CN109829037A (en) | 2017-11-22 | 2017-11-22 | Method, system, server and the storage medium of intelligent automatic question answering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109829037A true CN109829037A (en) | 2019-05-31 |
Family
ID=66858163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711175236.9A Pending CN109829037A (en) | 2017-11-22 | 2017-11-22 | Method, system, server and the storage medium of intelligent automatic question answering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829037A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378789A (en) * | 2019-07-16 | 2019-10-25 | 招商银行股份有限公司 | Solution, system and the computer readable storage medium of disputed transaction |
| CN110737763A (en) * | 2019-10-18 | 2020-01-31 | 成都华律网络服务有限公司 | Chinese intelligent question-answering system and method integrating knowledge map and deep learning |
| CN111831812A (en) * | 2020-09-21 | 2020-10-27 | 北京智源人工智能研究院 | Reading comprehension data set automatic generation method and device based on knowledge graph |
| WO2021139184A1 (en) * | 2020-01-07 | 2021-07-15 | 平安科技（深圳）有限公司 | Data query and display method, device, computer system, and readable storage medium |
| CN113742474A (en) * | 2021-11-08 | 2021-12-03 | 北京博瑞彤芸科技股份有限公司 | Intelligent question and answer method and device based on knowledge graph |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608218A (en) * | 2015-12-31 | 2016-05-25 | 上海智臻智能网络科技股份有限公司 | Intelligent question answering knowledge base establishment method, establishment device and establishment system |
| WO2016112679A1 (en) * | 2015-01-14 | 2016-07-21 | 百度在线网络技术（北京）有限公司 | Method, system and storage medium for realizing intelligent answering of questions |
| CN105893535A (en) * | 2016-03-31 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Intelligent question and answer method, knowledge base optimizing method and device and intelligent knowledge base |
| CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
| WO2017092380A1 (en) * | 2015-12-03 | 2017-06-08 | 华为技术有限公司 | Method for human-computer dialogue, neural network system and user equipment |
| CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易（杭州）网络有限公司 | A kind of answer provides method and apparatus |
-
2017
- 2017-11-22 CN CN201711175236.9A patent/CN109829037A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016112679A1 (en) * | 2015-01-14 | 2016-07-21 | 百度在线网络技术（北京）有限公司 | Method, system and storage medium for realizing intelligent answering of questions |
| WO2017092380A1 (en) * | 2015-12-03 | 2017-06-08 | 华为技术有限公司 | Method for human-computer dialogue, neural network system and user equipment |
| CN105608218A (en) * | 2015-12-31 | 2016-05-25 | 上海智臻智能网络科技股份有限公司 | Intelligent question answering knowledge base establishment method, establishment device and establishment system |
| CN105893535A (en) * | 2016-03-31 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Intelligent question and answer method, knowledge base optimizing method and device and intelligent knowledge base |
| CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
| CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易（杭州）网络有限公司 | A kind of answer provides method and apparatus |
Non-Patent Citations (2)
| Title |
|---|
| 刘杰;樊孝忠;王涛;: "基于本体的受限领域问答系统研究", 广西师范大学学报(自然科学版), no. 01, 15 March 2009 (2009-03-15) * |
| 刘汉兴;林旭东;田绪红;: "基于本体的自动答疑系统的研究与实现", 计算机应用, no. 02, 1 February 2010 (2010-02-01) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378789A (en) * | 2019-07-16 | 2019-10-25 | 招商银行股份有限公司 | Solution, system and the computer readable storage medium of disputed transaction |
| CN110737763A (en) * | 2019-10-18 | 2020-01-31 | 成都华律网络服务有限公司 | Chinese intelligent question-answering system and method integrating knowledge map and deep learning |
| WO2021139184A1 (en) * | 2020-01-07 | 2021-07-15 | 平安科技（深圳）有限公司 | Data query and display method, device, computer system, and readable storage medium |
| CN111831812A (en) * | 2020-09-21 | 2020-10-27 | 北京智源人工智能研究院 | Reading comprehension data set automatic generation method and device based on knowledge graph |
| CN113742474A (en) * | 2021-11-08 | 2021-12-03 | 北京博瑞彤芸科技股份有限公司 | Intelligent question and answer method and device based on knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829037A (en) | Method, system, server and the storage medium of intelligent automatic question answering | |
| WO2018006469A1 (en) | Knowledge graph-based human-robot interaction method and system | |
| CN109308357B (en) | Method, device and equipment for obtaining answer information | |
| CN107885842A (en) | Method, apparatus, server and the storage medium of intelligent answer | |
| CN108664472B (en) | Natural language processing method, device and equipment | |
| DE202016008238U1 (en) | Non-deterministic task initiation by a personal assistant module | |
| CN108519998B (en) | Problem guiding method and device based on knowledge graph | |
| US9720982B2 (en) | Method and apparatus for natural language search for variables | |
| CN111191450A (en) | Corpus cleaning method, corpus entry device and computer-readable storage medium | |
| CN109948151A (en) | The method for constructing voice assistant | |
| CN112507139B (en) | Knowledge graph-based question and answer method, system, equipment and storage medium | |
| CN111460120A (en) | Conversation management method, device, equipment and storage medium | |
| JP2023506087A (en) | Voice Wakeup Method and Apparatus for Skills | |
| WO2018231472A1 (en) | Identifying relationships from communication content | |
| AU2022201193A1 (en) | System and method for designing artificial intelligence (ai) based hierarchical multi-conversation system | |
| CN113569017B (en) | Model processing method and device, electronic equipment and storage medium | |
| CN110019709B (en) | Automatic question and answer method for robot and automatic question and answer system for robot | |
| CN115470329A (en) | Dialog generation method and device, computer equipment and storage medium | |
| CN113868396A (en) | Task intelligent dialogue construction method and system based on knowledge graph | |
| CN110442703B (en) | Knowledge graph-based information recommendation method and device and computer equipment | |
| CN114363277A (en) | Intelligent chatting method and device based on social relationship and related products | |
| CN111859191A (en) | GIS service aggregation method, device, computer equipment and storage medium | |
| CN111046151A (en) | Message processing method and device | |
| CN113837322B (en) | Course classification processing method, device, equipment and medium | |
| CN117422473A (en) | Transaction switching method and device, storage medium and electronic equipment |
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 |