CN109726279A - A kind of data processing method and device - Google Patents
A kind of data processing method and device Download PDFInfo
- Publication number
- CN109726279A CN109726279A CN201811648577.8A CN201811648577A CN109726279A CN 109726279 A CN109726279 A CN 109726279A CN 201811648577 A CN201811648577 A CN 201811648577A CN 109726279 A CN109726279 A CN 109726279A
- Authority
- CN
- China
- Prior art keywords
- content
- node
- information
- attribute
- knowledge mapping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a kind of data processing method and device, after obtaining input information, obtain with input information matches first content and with associated second content of first content, based on first content and with associated second content of first content, obtain feedback information, feedback information is exported, is achieved in and the expected desired feedback information of user is met based on input information output.And first content and the second content are related to input information, and it inputs information and can be the customer problem that user proposes with regard to a user demand, first content and the second content are matched with user demand, that is the feedback information obtained based on first content and the second content can effectively solve the problem that the user demand of the input information of this man-machine conversation, and then the user demand of this man-machine conversation can be effectively solved, it improves the user experience.
Description
Technical field
The application belongs to technical field of data processing, more specifically more particularly to a kind of data processing method and device.
Background technique
With the development of information technology, intelligent customer service can provide more and more abundant online business service for user, such as
On-line intelligence question and answer service answers customer problem (the problem of i.e. user proposes) to service by on-line intelligence question and answer automatically,
Such as call online intelligent answer service when, provide online business consultation interface for user, by online business consultation interface come
Customer problem is obtained, the answer of the customer problem is provided after analyzing customer problem.
Such as customer problem are as follows: the screen size of mobile phone products, then it can be from product know-how by the service of on-line intelligence question and answer
The screen size of the mobile phone products is found in library, and the parameter for indicating the screen size of the mobile phone products is then spliced to answer
In template, the answer of the customer problem and output are obtained.
But user's parameter that it may be not only to expect product when proposing some problem, as mobile phone products
For this customer problem of screen size, user is practical it is intended that understand the mobile phone if appropriate for being hold by one hand, if therefore
The answer for only providing screen size also needs user voluntarily the data of searching judges.
Summary of the invention
In view of this, the purpose of the application is to disclose a kind of data processing method and device, for based on input information,
It obtains meeting the expected desired feedback information of user, effectively solves the user demand of this man-machine conversation, improve user and use body
It tests.
The application discloses a kind of data processing method, is applied to intelligent conversational system, the intelligence conversational system can be right
The input information received is responded and provides feedback information, which comprises
Input information is obtained, the input information is that user during man-machine conversation is carried out by the intelligent conversational system
The information of input;
Obtain with it is described input information matches first content and with associated second content of the first content;
Based on the first content and with associated second content of the first content, obtain feedback information;
Export the feedback information.
Preferably, one of the first content and second content are the attribute of the first attribute of the first object
Value, another is the attribute value of the second attribute of first object, and first attribute and second attribute are different;
Or
One of the first content and second content are the attribute value of the first attribute of the first object, another
For the attribute value of the first attribute of the second object, first object and the second object difference.
Preferably, the first content of the acquisition and the input information matches and with the first content associated the
Two contents include:
The first content with the input information matches is obtained from default knowledge mapping;
Obtain the content that there is particular kind of relationship with the first content in the default knowledge mapping, it is described with described the
One content have characteristic relation content be and associated second content of the first content.
Preferably, described that the content packet that there is particular kind of relationship with the first content is obtained in the default knowledge mapping
It includes:
The corresponding first node of the first content, the default knowledge mapping note are obtained in the default knowledge mapping
Record the attribute value of at least one attribute of at least one object, and the default knowledge mapping is with object id information and object
The attribute value of every attribute is node, using the attribute of object as side connecting node;
The second node that first node corresponding with the first content is direct-connected in the default knowledge mapping is obtained, it is described
It is direct-connected to show to connect between first node and second node by a line;
Third node direct-connected with the second node in the default knowledge mapping is obtained, the third node is corresponding
Content as with the first content have particular kind of relationship content.
Preferably, described that the content packet that there is particular kind of relationship with the first content is obtained in the default knowledge mapping
It includes:
One of first knowledge mapping and the second knowledge mapping in the default knowledge mapping obtain and described the
The corresponding first node of one content, one of first knowledge mapping and second knowledge mapping record the first object
The attribute value of at least one attribute, the attribute value of at least one attribute of another the second object of record, and each knowledge mapping
Using the attribute value of the object id information of each self-recording object and every attribute of object as node, using the attribute of object as side
Connecting node;
The label information of the first node is obtained, the label information of the first node is for showing that first content is adapted to
Object-oriented requirements;
Based on the label information of the first node, in the another of first knowledge mapping and second knowledge mapping
Second node is obtained in a map, there is particular kind of relationship using the corresponding content of the second node as with the first content
The label information of content, the label information of the second node and the first node shows the same meaning.
Preferably, the method also includes: the first node is connected with the second node by a line, will
First knowledge mapping and second knowledge mapping constitute a knowledge mapping.
Preferably, it is described based on the first content and with associated second content of the first content, obtain feedback letter
Breath includes:
The feedback information is formed by the first content and second content;
Or
The first content and second content are modified, the feedback information is obtained.
A kind of data processing equipment is also disclosed in the application, is applied to intelligent conversational system, the intelligence conversational system can
Feedback information is responded and is provided to the input information received, described device includes:
First obtains unit, for obtaining input information, the input information is carried out by the intelligent conversational system
The information that user inputs during man-machine conversation;
Second obtaining unit, for obtaining with the first content of the input information matches and being closed with the first content
Second content of connection;
Third obtaining unit, for based on the first content and with associated second content of the first content, obtain
Feedback information;
Output unit, for exporting the feedback information.
Preferably, one of the first content and second content are the attribute of the first attribute of the first object
Value, another is the attribute value of the second attribute of first object, and first attribute and second attribute are different;
Or
One of the first content and second content are the attribute value of the first attribute of the first object, another
For the attribute value of the first attribute of the second object, first object and the second object difference.
Preferably, second obtaining unit includes:
First obtains module, for obtaining the first content with the input information matches from default knowledge mapping;
Second obtains module, has particular kind of relationship with the first content for obtaining in the default knowledge mapping
Content, it is described to be and associated second content of the first content with the first content has characteristic relation content.
Preferably, the second acquisition module includes:
First acquisition submodule, for obtaining the corresponding first segment of the first content in the default knowledge mapping
Point, the default knowledge mapping record the attribute value of at least one attribute of at least one object, and the default knowledge mapping
Using the attribute value of object id information and every attribute of object as node, using the attribute of object as side connecting node;
Second acquisition submodule, for obtaining first node corresponding with the first content in the default knowledge mapping
Direct-connected second node, it is described it is direct-connected show between first node and second node by a line connect;
Third acquisition submodule, for obtaining third section direct-connected with the second node in the default knowledge mapping
Point, using the corresponding content of the third node as the content with the first content with particular kind of relationship.
Preferably, the second acquisition module includes:
First obtain submodule, in the default knowledge mapping the first knowledge mapping and the second knowledge mapping its
One of obtain corresponding with first content first node, first knowledge mapping and second knowledge mapping are wherein
One of record the first object at least one attribute attribute value, another record the second object at least one attribute attribute
Value, and each knowledge mapping is section with the attribute value of the object id information of each self-recording object and every attribute of object
Point, using the attribute of object as side connecting node;
Second obtains submodule, for obtaining the label information of the first node, the label information of the first node
For showing the object-oriented requirements of first content adaptation;
Third obtains submodule, for the label information based on the first node, in first knowledge mapping and institute
It states in another map of the second knowledge mapping and obtains second node, using the corresponding content of the second node as with described
One content has the content of particular kind of relationship, and the label information of the second node and the label information of the first node show together
One meaning.
Preferably, second obtaining unit further include:
Merging module, for connecting the first node by a line with the second node, by described first
Knowledge mapping and second knowledge mapping constitute a knowledge mapping.
Preferably, the third obtaining unit is specifically used for the first content and second content composition is described anti-
Feedforward information;
Or
The first content and second content are modified, the feedback information is obtained.
A kind of electronic equipment, including processor and display component is also disclosed in the application, and the processor has intelligent session
System, the intelligence conversational system can respond the input information received and provide feedback information;
The processor, for obtain input information, obtain with it is described input information matches first content and with institute
State associated second content of first content, and based on the first content and with associated second content of the first content, obtain
Feedback information is obtained, the input information is the letter that user inputs during carrying out man-machine conversation by the intelligent conversational system
Breath;
The display component, for exporting the feedback information.
A kind of storage medium is also disclosed in the application, and computer program code, the calculating are stored on the storage medium
Machine program code realizes above-mentioned data processing method when executing.
From above-mentioned technical proposal it is found that obtain input information after, obtain with input information matches first content and
With associated second content of first content, based on first content and with associated second content of first content, obtain feedback information,
Feedback information is exported, is achieved in and the expected desired feedback information of user is met based on input information output.And first content
It is related to input information to the second content, and input information and can be the customer problem that user proposes with regard to a user demand, make
Obtaining first content and the second content can match with user demand, that is to say, that be obtained based on first content and the second content anti-
Feedforward information can effectively solve the problem that the user demand of the input information of this man-machine conversation, and then can effectively solve this man-machine meeting
The user demand of words, improves the user experience.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the application
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is the schematic diagram of intelligence conversational system disclosed in the embodiment of the present application;
Fig. 2 is a kind of flow chart of data processing method disclosed in the embodiment of the present application;
Fig. 3 is a kind of flow chart for obtaining the second content disclosed in the embodiment of the present application;
Fig. 4 a is a kind of schematic diagram of default knowledge mapping disclosed in the embodiment of the present application;
Fig. 4 b is a kind of schematic diagram of the form of feedback information disclosed in the embodiment of the present application;
Fig. 4 c is the schematic diagram that another kind disclosed in the embodiment of the present application presets knowledge mapping;
Fig. 4 d is the schematic diagram of the form of another kind feedback information disclosed in the embodiment of the present application;
Fig. 5 is the schematic diagram that another kind disclosed in the embodiment of the present application obtains the second content;
Fig. 6 a is a kind of schematic diagram of default knowledge mapping disclosed in the embodiment of the present application;
Fig. 6 b is a kind of schematic diagram of the form of feedback information disclosed in the embodiment of the present application;
Fig. 6 c is the schematic diagram that another kind disclosed in the embodiment of the present application presets knowledge mapping;
Fig. 6 d is the schematic diagram of the form of another kind feedback information disclosed in the embodiment of the present application;
Fig. 6 e is the schematic diagram of another disclosed default knowledge mapping of the embodiment of the present application;
Fig. 7 is a kind of structural schematic diagram of data processing equipment disclosed in the embodiment of the present application;
Fig. 8 is the structural schematic diagram of another kind data processing equipment disclosed in the embodiment of the present application;
Fig. 9 is the structural schematic diagram of another data processing equipment disclosed in the embodiment of the present application;
Figure 10 is the configuration diagram of another data processing equipment disclosed in the embodiment of the present application.
Specific embodiment
During carrying out man-machine conversation by intelligent conversational system at present, the feedback information that is exported by intelligent conversational system
It may it is expected that the reply received has gap with user, such as intelligent conversational system is applied in electronic equipment, it such as will be intelligent
Conversational system is installed in the electronic device in a manner of APP (application program), may make user when electronic equipment runs the APP
Man-machine conversation is carried out by electronic equipment.
As shown in Figure 1, user pass through intelligent conversational system inquiry " screen size of product A " this customer problem (can be with
It is considered as a kind of input information) when, the feedback information of intelligent conversational system output " 5.5 inches ".But user inquires " product A's
The actual demand of screen size " is intended to understand the model mobile phone if appropriate for single hand grip, it is clear that intelligent conversational system output
Feedback information can not solve this demand of user, may receive in the case because of feedback information and user's expectation
Reply has gap and this man-machine conversation is caused to terminate, or intelligent conversational system may receive that " I is just intended to again
The size for solving mobile phone screen sees whether to be suitble to single hand grip " input information, as shown in Figure 1, by the way that repeatedly interaction can
So that the feedback information of intelligent conversational system output is able to solve user demand, but this mode will affect the use body of user
It tests.
The embodiment of the present application discloses a kind of data processing method and device thus, defeated for based on the user's input information
Meet the expected desired feedback information of user out, that is to say, that the feedback information obtained based on first content and the second content can
The user demand of this man-machine conversation is effectively solved, to reduce interaction times, is improved the user experience.
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall in the protection scope of this application.
Referring to Fig. 2, a kind of data processing method provided by the embodiments of the present application is shown, the data processing method application
In intelligent conversational system, intelligent conversational system can be responded and be provided feedback information to the input information received, can be with
The following steps are included:
201: obtaining input information.
In the present embodiment, input information is the letter that user inputs during carrying out man-machine conversation by intelligent conversational system
Breath, input information refer to the information that user inputs according to the expectation of itself to intelligent conversational system.
Inputting information may include at least one of problem information, data information and demand information etc..Wherein problem is believed
Breath inquired for an object, for example, the value of at least one attribute of one object of inquiry, data information be then provide to
The value of a few attribute is then to directly give the current need of user with object, the demand information of inquiring the value with the attribute
It asks.Separately below for example: on the one hand, for electronic product, such as correspondence problem information, the input of user may be
" battery of certain model mobile phone is great? " the problems such as information, such as corresponding data information, the input of user may be
Data informations such as " 2000mA, 64G, 5 inches ", such as input information can be that " battery 2000mA, stored memory are 64G's
Have which product ", in another example corresponding demand information, the input of user may be demand informations such as " continuations of the journey, large-size screen monitors ", such as input letter
Breath can be " cruising ability of which product is strong " or " cruising ability of product A has much ".
On the other hand, for dress-goods, correspondence problem information, the input of user may be that " whether size is normal
Size, if appropriate for the partially fertile people of figure " the problems such as information, corresponding data information, the input of user may be " height 180, body
The data informations such as weight 70kg ", corresponding demand information, the input of user may be demand informations such as " warming, antistatics ".
According to the content in above-mentioned two aspect, the particular content for inputting information may be for different types of object
Different, but the type for inputting information includes but are not limited in above-mentioned two aspect the problem of being related to information, data letter
Breath and demand information, remaining information and type will not enumerate.
In practical applications, intelligent conversational system can be by NLUPipeline (natural language understanding model) to input
Information is handled, which may include: to be converted to using such as Word2Vec (term vector model) to input information
Term vector segments input information, part of speech note, Sentence analysis, is intended to the work such as understanding and key message extraction, finally
The key-value being made of processing result mark pair is obtained, conversational system intelligent in this way can be based on key-value mark pair
Obtain the feedback information to match with input information.
Such as during carrying out man-machine conversation, input information is M, is handled in NLUPipeline M, first
It obtains M={ m1, m2 ..., mn }, each of M element representation inputs a word in information, then obtains processing result O
={ o1, o2 ..., on }, each of O element representation carry out a pair of of the mark obtained after language understanding to input information, i.e.,
The above-mentioned key-value mark pair being related to.It should be noted that the above-mentioned process handled input information being related to
Above-mentioned several ways are included but are not limited to, remaining mode will not enumerate herein.
202: obtain with input information matches first content and with associated second content of first content.Namely
It says, first content is to be directly obtained by inputting information, and the second content needs to obtain by first content, makes in second indirectly
Appearance is related to input information, so that content relevant to input information is expanded.
In the present embodiment, first content and the first feasible pattern of the second content are: first content and the second content
One of them for the first object the first attribute attribute value, another for the first object the second attribute attribute value, first
Attribute and the second attribute are different.Include following several situations: first content is the attribute value of the first attribute of the first object, the
Two contents are the attribute value of the second attribute of the first object.Or first content is the attribute value of the second attribute of the first object,
Second content is the attribute value of the first attribute of the first object.Such as when input information is " screen size of product A is how many "
When, intelligent conversational system exports " screen size: 5.5 inches, resolution ratio are as follows: the feedback information of 2248*1080 ", wherein screen is big
Small is the above-mentioned first content being related to, and resolution ratio is above-mentioned the second content being related to.
The second way of first content and the second content is: one of first content and the second content are the first object
The first attribute attribute value, another is the attribute value of the first attribute of the second object, and the first object and the second object are different.
Include following several situations: first content is the attribute value of the first attribute of the first object, and the second content is the second object
The attribute value of first attribute.Alternatively, first content is the attribute value of the first attribute of the second object, the second content is the first object
The first attribute attribute value.Such as when inputting information is " screen size of product A is how many ", intelligent conversational system output
" screen size are as follows: 5.5 inches, suitable hand length are as follows: the feedback message of 18cm ", wherein screen size is above-mentioned is related to
First content, being suitble to hand length is above-mentioned the second content being related to.
The third mode of first content and the second content is: one of first content and the second content are with first
The object id information of first object of the attribute value of attribute, another is the second object of the attribute value with the first attribute
Object id information.Such as when input information is " mobile phones of 5.5 inch screens ", intelligent conversational system, which can export, " has 5.5 English
The mobile phone of very little screen is product A and product B ", and wherein product A is the above-mentioned first content being related to, and product B is above-mentioned is related to
The second content.
Above-mentioned three kinds of modes are only explanations, other content can also be provided in practical applications, such as first way
For, can also obtain with the associated third content of first content, the third content can be with first content belongs to a pair
The attribute value of another attribute of elephant can also be in the object id information for obtaining different objects for the third mode
Except obtain each object other attributes attribute value, such as first content and the second content show the attribute with attribute B
The object 1 and object 2 of value, then can also obtain the attribute value of the attribute C of object 1 and/or object 2 on this basis.
In the present embodiment, the mode for obtaining first content and the second content associated with first content is but unlimited
In are as follows: the first content that information matches are inputted with user is obtained from default knowledge mapping first, then in default knowledge mapping
Middle acquisition has the content of particular kind of relationship with first content, should have the content of characteristic relation as with first content and first content
Associated second content.
Such as according to it is above-mentioned to as user input information M handled processing result O=o1, o2 ...,
On }, it is obtained in default knowledge mapping and the fact that input information matches knowledge graph (a part in default knowledge mapping) K=
<E, R>, wherein E={ e1, e2 ..., en }, an entity in each of E element representation fact knowledge figure, R=r1,
R2 ..., rn }, the relationship in each of R element representation fact knowledge figure between certain two entity is known from there through the fact
E can obtain first content in knowledge figure.The result figure that there is particular kind of relationship with fact knowledge figure K is obtained from default knowledge mapping
G=<E ', R '>, wherein E '={ e ' 1, e ' 2 ..., e ' n }, an entity in each of E ' element representation result figure, R '
={ r ' 1, r ' 2 ..., r ' n }, the relationship in each of R ' element representation result figure between certain two entity, from there through
E ' in result figure G can obtain first content and the second content, as contained in the E of the E ' of result figure G and fact knowledge figure K
Identical entity can be considered as first content, with the related entity of the first content be then the second content in result figure G,
It can be indicated by the R ' in result figure with the relationship of first content.
203: based on first content and with associated second content of first content, obtain feedback information.
In the present embodiment, first content and the second content are formed into feedback information, alternatively, modification first content and second
Content obtains feedback information.
Such as first content and the second content are not modified, first content and the second content are directly formed into feedback information A=
{ a1, a2 ..., an }, a word of each of A element representation feedback information, and each word is in first content and second
Word in appearance, such as first content and the second content are as follows: 5.5 inches and 64G, then feedback information may is that 5.5 inches, 64G,
So that the description of feedback information is simple, but have a problem that: it is possible that user does not know feedback after output feedback information
What each part of information indicates, such as above-mentioned feedback information " 5.5 inches, 64G ", if user is to product
Attribute is unclear, then user cannot be understood by one mesh of feedback information knows what the two numerical value indicate.
In another example the mode of modification first content and the second content is: selecting corresponding language according to the type of input information
Template inserts first content and the second content in the language template, corrects language template by the matched mode of preset rules
In syntax error, obtain feedback information.The type for wherein inputting information shows the tone that the input information uses, such as input letter
Breath can be any one in declarative sentence, interrogative sentence and double negative sentences, can be pre-configured with language for the sentence of these types
Say that template, the language template repair first content and the second content after obtaining first content and the second content for showing
Change mode.Such as in language template reserve filling first content and the second content position, in this way obtain first content and
It, can be by first content and the second fills to opposite position, to obtain feedback information after second content.
Additionally the language can be modified by preset rules to the language template for being filled with first content and the second content
The syntax error of template, so that the feedback information finally obtained meets language requirement, wherein preset rules can be according to being actually subjected to
Setting is asked, the present embodiment no longer illustrates.Obtaining above-mentioned first content and the second content " 5.5 inches and 64G " in this way
Afterwards, the feedback information of acquisition can be but be not limited to: the screen size of product A is 5.5 inches, stored memory 64G, can
Meet user large screen and high memory requirements.
204: output feedback information.
In the present embodiment, the mode for exporting feedback information, which may is that, exports electron equipment by intelligent conversational system,
Feedback information is shown by electronic equipment again, such as shows that the display area may belong to electricity on the display region by electronic equipment
Sub- equipment or the display area extended for electronic equipment.Such as feedback information is exported to mobile phone or is put down by intelligent conversational system
On panel device, feedback information is shown by the screen of mobile phone or tablet device.In another example mobile phone or tablet device with
Projector connection, intelligent conversational system export feedback information to mobile phone or tablet device, then will be anti-by mobile phone or tablet device
Feedforward information is exported to projector, shows feedback information by the projecting function of projector, user is made to check feedback information.
From above-mentioned technical proposal it is found that obtain input information after, obtain with input information matches first content and
With associated second content of first content, based on first content and with associated second content of first content, obtain feedback information,
Feedback information is exported, is achieved in and the expected desired feedback information of user is met based on input information output.And first content
It is related to input information to the second content, and input information and can be the customer problem that user proposes with regard to a user demand, make
Obtaining first content and the second content can match with user demand, that is to say, that be obtained based on first content and the second content anti-
Feedforward information can effectively solve the problem that the user demand of the input information of this man-machine conversation, and then can effectively solve this man-machine meeting
The user demand of words, it is possible to reduce interaction times improve the user experience.
For obtaining the process for the content that there is particular kind of relationship with first content in default knowledge mapping, referring to Fig. 3,
A kind of feasible pattern for obtaining the second content disclosed in the embodiment of the present application is shown, may comprise steps of:
301: the first node of first content is obtained in default knowledge mapping.
In the present embodiment, default knowledge mapping records the attribute value of at least one attribute of at least one object, and pre-
If knowledge mapping using the attribute value of object id information and every attribute of object as node, is connected as side using the attribute of object and is saved
Point.
That is, default knowledge mapping can be made of the attribute value of a variety of attributes of an object, be also possible to by
The attribute value of a variety of attributes of multiple objects is constituted, and the type of multiple objects can be identical when being made of multiple objects,
It can also be different.For knowledge mapping default for one, the node having in all nodes of the default knowledge mapping is with object
Object id information name, some nodes are named with the attribute value of the attribute of object, and the side between node is with the category of object
Life name.
For different object, attribute and the corresponding attribute value of attribute be not also identical.Such as mobile phone products,
Attribute is the attributes such as screen, system version, running memory, stored memory, battery and the system level chip (Soc) of mobile phone products.
For dress-goods, attribute is the attributes such as materials, size and the mode of washing of clothes.In the present embodiment, the production of remaining type
The attribute of product is just not listed one by one.
302: obtaining the second node that first node corresponding with first content is direct-connected in default knowledge mapping.
In the present embodiment, direct-connected to show to connect between first node and second node by a line, such as default
In knowledge mapping, it is connected using attribute as side between node, then one of first node and second node are the category of attribute
Property value, another is object id information.
303: obtaining third node direct-connected with second node in default knowledge mapping, the corresponding content of third node is made
To have the content of particular kind of relationship with first content, it is possible thereby to using the second node in default knowledge mapping as connection matchmaker
It is situated between, the content that there is particular kind of relationship with first content is obtained by second node.
In the present embodiment, when default knowledge mapping is made of the attribute of an object, if the object includes more
A attribute, the third node direct-connected with second node include the attribute value of more than one attribute, then can by second node
The attribute value of multiple attributes in addition to first node corresponds to attribute is obtained, but if the object for forming default knowledge mapping can
An attribute can be only existed, using the attribute as first node after, the third node direct-connected with object B may be 0, at this time
The content with first content with particular kind of relationship obtained is then empty.
When default knowledge mapping is made of the attribute of multiple objects, then third node is the category of the attribute of different objects
Property value.Such as when inputting information is " mobile phone that screen size is 5.5 inches ", the attribute for obtaining first node is screen, is belonged to
Property value be 5.5 inches, it is assumed that the corresponding mobile phone of 5.5 inch screens have product A and product B, product A and product B be second node,
Then third node is the attribute value of the attribute direct-connected with product A and the attribute value of the attribute direct-connected with product B.
It can be seen from the above, if default knowledge mapping is made of an object, and the object includes multiple attributes, then passes through
It is that second node obtains to be with first content has particular kind of relationship content: other categories of an object are belonged to first content
Property attribute value, if default knowledge mapping is made of multiple objects, and multiple objects respectively include multiple attributes, then by the
It is that two nodes obtain to be with first content has particular kind of relationship content: other attributes of an object are belonged to first content
Attribute value, and belong to from first content the attribute value of other attributes of different objects, wherein other attributes refer to except first
Attribute except the corresponding attribute of content.
In the present embodiment, it is how more preferable illustrate is obtained in default knowledge mapping with first content with specific
The content of relationship is illustrated in conjunction with Fig. 4 a to Fig. 4 d:
Example one:
It is that one is made of the attribute value of multiple attributes of product A in advance by taking product A as an example that knowledge mapping is preset shown in Fig. 4 a
If knowledge mapping, in fig.4: the attribute of node 401 is screen, and attribute value is 5.5 inches (inch), node 402 is object,
Object id information is product A, the attribute of node 4031 is system version, attribute value 6.0.1, node 4032 attribute be electricity
Pond, attribute value be 2000 milliampere hour (mAh), node 4033 attribute be stored memory, the category of attribute value 64G, node 4034
Property is Soc, attribute value 1.8GHz, node 4035 attribute be stored memory, attribute value 4GB.
Assuming that obtaining the corresponding first node 401 of first content in default knowledge mapping, obtain straight with first node 401
Second node 402 even, first node 401 and second node 402 are Bian Zhilian with the attribute of first node 401.It obtains and the
The direct-connected third node of two nodes 402, third node are as follows: node 4031, node 4032, node 4033, node 4034 and node
4035.By the corresponding content of third node, i.e. the attribute of third node and attribute value has particular kind of relationship as with first content
Content.
With reference to Fig. 4 b, such as when user inputs " the mobile phone screen size of product A " information, intelligent conversational system can be from pre-
If the attribute value of screen size and the attribute value of the attribute in addition to screen size are obtained in knowledge mapping, as obtained in Fig. 4 b
Attribute value have: " screen size: 5.5 inches, system version: in 6.0.1, battery: 2000mAh, stored memory: 64GB, operation
Deposit: 4GB, System on Chip/SoC information: 1.8GHz " etc. obtain feedback information output, specific feedback system then in conjunction with these attribute values
Above- mentioned information directly can be sent to user.Above- mentioned information can also be inserted in language template, it is defeated to form complete sentence
Out, for example, by above- mentioned information be combined into sentence " A mobile phone belongs to large-screen mobile phone, screen size are as follows: 5.5 inches, system version are as follows:
6.0.1, battery sizes are as follows: 2000mAh, stored memory are as follows: 64GB, running memory are as follows: 4GB, System on Chip/SoC information are as follows:
1.8GHz ", and above-mentioned sentence is passed through into intelligent conversational system output.
Example two:
It is by taking product A and product B as an example, by the category of multiple attributes of product A and product B that knowledge mapping is preset shown in Fig. 4 c
Property value constitute a default knowledge mapping.In Fig. 4 c, the attribute of node 401 is screen, attribute value is 5.5 inches (inch),
Node 402 is object, and object id information is product A, the attribute of node 4021 is system version, attribute value 6.0.1, node
4022 attribute be running memory, attribute value 4GB, node 4023 attribute be stored memory, attribute value 64GB, node
403 be object, and object id information is product B, the attribute of node 4031 is system version, attribute value 6.0.1, node 4032
Attribute be running memory, attribute value 6GB, node 4033 attribute be stored memory, attribute value 128GB.
Assuming that obtaining the corresponding first node 401 of first content in default knowledge mapping, obtain straight with first node 401
Second node even is node 402 and node 403, and obtaining with the direct-connected third node of node 402 is node 4021, node 4022
With node 4023, obtaining with the direct-connected third node of node 403 is node 4031, node 4032 and node 4033.By third section
The corresponding content of point, the i.e. attribute of third node and attribute value are as the content with first content with particular kind of relationship.
With reference to Fig. 4 d, such as when user's input " mobile phone that screen is 5.5 inches " information, intelligent conversational system can be incited somebody to action
" mobile phone model: the screen size of product A and product B, product A and product B are as follows: 5.5 inches, the system version of product A and product B
This are as follows: the running memory of 6.0.1, product A are as follows: the stored memory of 4GB, product A are as follows: the running memory of 64GB, product B are as follows:
The stored memory of 6GB, product B are as follows: the output such as 128GB ", specific feedback system can be directly using above- mentioned information as feedback information
Output.Above- mentioned information can also be inserted in language template, form complete sentence output, for example above- mentioned information combination is formed a complete sentence
" mobile phone screen is that 5.5 inches of mobile phone has product A and product B, the product parameters of product A to son are as follows: system version 6.0.1, fortune
It is 64GB, the product parameters of product B that 4GB and stored memory are saved as in row are as follows: system version 6.0.1, running memory be 6GB and
Stored memory is 128GB.", and above-mentioned sentence is exported.
It should be noted that the content shown in above-mentioned two example is used only for illustrating.
From above-mentioned technical proposal it is found that obtaining input information during carrying out man-machine conversation by intelligent conversational system
Afterwards, the first node of first content is obtained in default knowledge mapping, and acquisition is straight with first node in default knowledge mapping
Second node even, using in default knowledge mapping content corresponding with the third node that second node is direct-connected as in first
Container has the second content of particular kind of relationship, obtains feedback information, exports feedback information.It is achieved in based on input information, output
Meet the expected desired feedback information of user, and first content and the second content can be matched with user demand, that is to say, that
First content and the second content can effectively solve the problem that the user demand of this man-machine conversation, it is possible to reduce interaction times are improved and used
Family usage experience.
For obtaining the another way such as Fig. 5 for the content that there is particular kind of relationship with first content in default knowledge mapping
It is shown, it may comprise steps of:
501: being obtained in first in one of the first knowledge mapping and the second knowledge mapping preset in knowledge mapping
Hold corresponding first node.
In the present embodiment, one of the first knowledge mapping and the second knowledge mapping record at least one of the first object
The attribute value of attribute, the attribute value of at least one attribute of another the second object of record, and each knowledge mapping is respectively to remember
The attribute value of every attribute of the object id information and object of the object of record is node, is connected and is saved as side using the attribute of object
Point.
502: obtaining the label information of first node.
In the present embodiment, the label information of first node is used to show the object-oriented requirements of first content adaptation.This first
The label information of node can be arranged in the first knowledge mapping and the second knowledge mapping building process, such as according to the first knowledge graph
The information that can embody object-oriented requirements is got in the evaluation information that spectrum and the second knowledge mapping building process use, based on embodiment
The information acquisition label information of object-oriented requirements, such as the information of object-oriented requirements will be directly embodied as label information.Such as it embodies
The information of object-oriented requirements is: being suitable for large screen demand, then " can will be suitable for large screen demand " is used as label information.
503: the label information based on first node, in another map of the first knowledge mapping and the second knowledge mapping
Second node is obtained, using the corresponding content of second node as the content with first content with particular kind of relationship.
In the present embodiment, the label information of second node and the label information of first node show the same meaning, by
This obtains content corresponding with the second node of first node label information having the same, realizes and obtains the corresponding same object
The both sides content of demand.Obtaining that there is particular kind of relationship with first content based on the first knowledge mapping and the second knowledge mapping
Content after, data processing method provided in this embodiment can be with: first node and second node being passed through a line and connected
Connect, the first knowledge mapping and the second knowledge mapping constituted into a knowledge mapping, be again based in this way the first knowledge mapping and
Second knowledge mapping can be obtained when obtaining the content for having particular kind of relationship with first content using mode shown in Fig. 3.
In the present embodiment, it is how more preferable illustrate is obtained in default knowledge mapping with first content with specific
The content of relationship specifically combines Fig. 6 a to Fig. 6 e, and is illustrated by following example:
Example one: corresponding with first content first is obtained from preset first knowledge mapping or the second knowledge mapping
Node obtains the label information of first node, according to the label information of first node, obtains label in another knowledge mapping
The consistent second node of the label information meaning of information and first node, using the corresponding content of second node as with first content
There is the content of particular kind of relationship.
Such as with reference to Fig. 6 a, in the first knowledge mapping, node 601 be object, object id information A, node 602
Attribute is screen size, and attribute value is 5.5 inches, and 603 be label, and label information is large screen.In the second knowledge mapping, section
The attribute of point 604 is hand sizes, and attribute value 18cm, 605 are label, and label information is large screen.Wherein, in Fig. 6 a
Second knowledge mapping is the knowledge mapping comprising user information constructed in advance, and user information includes but are not limited to: height, body
Weight, gender, measurements of the chest, waist and hips, hand sizes and sole size etc..
In conjunction with Fig. 6 a, it is assumed that when input information is " screen size of A ", obtained from the first knowledge mapping and the information phase
Corresponding first node 602, obtains the label information of the first node 602, is label letter if the corresponding content of label 603
Breath.Based on label 603, label information identical with the Meaning of Information of label 603 and corresponding is found from the second knowledge mapping
Node obtains second node 604 and label 605.There is specific pass using the corresponding content of second node 604 as with first content
The content of system.
I.e. when user input " screen size of A " information when, intelligent conversational system can will " screen size: 5.5 inches,
Hand size: the information such as 18cm " feed back to user, and specific feedback system can directly export above- mentioned information.It can also will be upper
It states information to be inserted in language template, forms complete sentence output, for example in figure 6b, above- mentioned information are combined into sentence " root
According to the information of your input, the screen size of A are as follows: 5.5 inches, be suitble to the size of palm are as follows: 18cm ", and above-mentioned sentence is defeated
Out.
Another example: it is obtained from preset first knowledge mapping or the second knowledge mapping corresponding with first content
First node obtains the label of first node, according to the label information of the label of first node, obtains in another knowledge mapping
Label information and the consistent label of label information meaning of first node and corresponding second node are taken, in another knowledge mapping
Middle at least one third node direct-connected with second node of selection, by content corresponding with third node in another knowledge mapping
And/or content corresponding with the direct-connected fourth node of third node is as the content with first content with particular kind of relationship.
Wherein chosen in another knowledge mapping at least one third node direct-connected with second node mode have but
Be not limited only to: optional one is used as third node from all nodes direct-connected with second node, or will be all and second
The corresponding object id information output of the direct-connected node of node, prompts user to select from all object id informations, will
The corresponding node of the object id information selected is as third node.Third node selected with user below and by third section
The direct-connected corresponding content of fourth node of point for the content with first content with particular kind of relationship as being illustrated:
Such as with reference to Fig. 6 c, in the first knowledge mapping, the attribute of node 601 is hand sizes, attribute value 18cm, mark
The label information of label 602 is large screen.In the second knowledge mapping, the label information of label 603 is large screen, node 604
Attribute is screen size, and attribute value is 5.5 inches, node 605 is object, and object id information is product A, node 606 is pair
As object id information is product B, node 607 is object, and object id information is products C.
In conjunction with Fig. 6 c, it is assumed that when user inputs the information of " pushing 18cm ", obtain first node from the first knowledge mapping
601, and the label 602 of first node 601 is obtained, it is obtained from the second knowledge mapping according to label 602 and the mark of label 602
The consistent label 603 of Meaning of Information is signed, second node 604 is obtained from the second knowledge mapping according to label 603.Obtain it is all with
The direct-connected object of second node 604: node 605, node 606 and node 607, and with any type in model series in Fig. 6 d
Mark is exported in a manner of combobox, and user can choose a model identification from combobox.Such as user select product A this
A type, that is, the third node selected is node 605, using the node direct-connected with third node 605 as fourth node, by the 4th
The corresponding content of node can be obtained as the content with first content with particular kind of relationship such as the content in Fig. 6 e.
In Fig. 6 e, node 605 is object, and object id information is product A, the attribute of node 6051 is system version,
Attribute value is 6.0.1, the attribute of node 6052 is battery, attribute value be 2000 milliampere hour (mAh), node 6053 attribute be to deposit
Store up memory, attribute value 64G, node 6054 attribute be Soc, attribute value 1.8GHz, node 6055 attribute be in storage
It deposits, attribute value 4GB.By node 6051 to node 6055 be used as fourth node, will content corresponding with fourth node as with
First content has the content of particular kind of relationship.
In conjunction with Fig. 6 d, such as when user inputs the information of " pushing 18cm ", intelligent conversational system can to user feedback " according to
Your pushing information, the mobile phone screen size that you are suitble to are 5.5 inches, and the screen size of following mobile phone model is 5.5 inches, are asked
You select the type wanted to know about: product A, product B and products C ", when user selects " product A ", intelligent conversational system can be incited somebody to action
" mobile phone model: A, screen size: 5.5 inches, system version: in 6.0.1, battery: 2000mAh, stored memory: 64GB, operation
Deposit: 4GB, System on Chip/SoC information: 1.8GHz, feature: large-screen mobile phone " etc. information output, specific feedback system can directly will be upper
State information output.Above- mentioned information can also be inserted in language template, form the output of complete sentence, such as by above- mentioned information group
Synthesis sentence " type model A, the A mobile phone that you select belongs to large-screen mobile phone, screen size are as follows: 5.5 inches, system version are as follows:
6.0.1, battery sizes are as follows: 2000mAh, stored memory are as follows: 64GB, running memory are as follows: 4GB, System on Chip/SoC information are as follows:
1.8GHz ", and above-mentioned sentence is exported.
It should be noted that the content being related in above two example is only used for illustrating.
From above-mentioned technical proposal it is found that obtaining input information during carrying out man-machine conversation by intelligent conversational system
Afterwards, one of the first knowledge mapping in default knowledge mapping and the second knowledge mapping acquisition corresponding with first content the
The label information of one node and first node, the label information based on first node, in the first knowledge mapping and the second knowledge graph
Second node is obtained in another map of spectrum, there is particular kind of relationship using the corresponding content of second node as with first content
The second content, obtain feedback information, export feedback information.It is achieved in based on input information, output meets user's expected phase
The feedback information of prestige, and first content and the second content can be matched with user demand, that is to say, that first content and second
Content can effectively solve the problem that the user demand of this man-machine conversation, it is possible to reduce interaction times improve the user experience.
For the various method embodiments described above, for simple description, therefore, it is stated as a series of action combinations, but
Be those skilled in the art should understand that, the application is not limited by the described action sequence because according to the application, certain
A little steps can be performed in other orders or simultaneously.Secondly, those skilled in the art should also know that, it is retouched in specification
The embodiment stated belongs to preferred embodiment, necessary to related actions and modules not necessarily the application.
Corresponding with above method embodiment, the embodiment of the present application discloses a kind of data processing equipment, is applied to intelligence meeting
Telephone system, the intelligence conversational system can be responded and be provided feedback information to the input information received, wherein data
The structure of processing unit is as shown in fig. 7, may include: first obtains unit 701, the second obtaining unit 702, third obtaining unit
703 and output unit 704.
First obtains unit 701, for obtaining input information, the input information is to carry out people by intelligent conversational system
The information that user inputs during chance words.
In the present embodiment, input information is the letter that user inputs during carrying out man-machine conversation by intelligent conversational system
Breath, input information refer to the information that user inputs according to the expectation of itself to intelligent conversational system.Jie of specific input information
Continue the explanation please referred in above method embodiment, and to this, this embodiment is not repeated.
Second obtaining unit 702, for obtain and input the first content of information matches and with first content it is associated
Second content.
In the present embodiment, one of first content and the second content are the attribute value of the first attribute of the first object,
Another is the attribute value of the second attribute of the first object, and the first attribute and the second attribute are different.Alternatively, first content and second
One of content is the attribute value of the first attribute of the first object, another is the attribute value of the first attribute of the second object,
First object and the second object are different.Or one of first content and the second content are the attribute value with the first attribute
The object id information of first object, another is the object id information of the second object of the attribute value with the first attribute.
Specific to introduce the explanation please referred in above method embodiment, to this, this embodiment is not repeated.
Third obtaining unit 703, for based on first content and with associated second content of first content, obtain feedback letter
Breath.
In the present embodiment, first content and the second content are formed into feedback information, alternatively, modification first content and second
Content obtains feedback information.Specific to introduce the explanation please referred in above method embodiment, to this, this embodiment is not repeated.
Output unit 704, for exporting feedback information.
In the present embodiment, the mode for exporting feedback information, which may is that, exports electron equipment by intelligent conversational system,
Feedback information is shown by electronic equipment again, such as shows that the display area may belong to electricity on the display region by electronic equipment
Sub- equipment or the display area extended for electronic equipment.Such as feedback information is exported to mobile phone or is put down by intelligent conversational system
On panel device, feedback information is shown by the screen of mobile phone or tablet device.In another example mobile phone or tablet device with
Projector connection, intelligent conversational system export feedback information to mobile phone or tablet device, then will be anti-by mobile phone or tablet device
Feedforward information is exported to projector, shows feedback information by the projecting function of projector, user is made to check feedback information.
From above-mentioned technical proposal it is found that obtain input information after, obtain with input information matches first content and
With associated second content of first content.Based on first content and with associated second content of first content, obtain feedback information,
Export feedback information.It is achieved in based on the user's input information, output meets the expected desired feedback information of user.In first
Hold and the second content is related to user's input information, that is to say, that first content and the second content can effectively solve the problem that this is man-machine
The user demand of the input information of session, and then can effectively solve the problems, such as that user proposes, it improves the user experience.
In the present embodiment, above-mentioned second obtaining unit 702, which obtains a kind of mode of the second content, is: from default knowledge graph
The first content with input information matches is obtained in spectrum.Obtaining in default knowledge mapping has particular kind of relationship with first content
Content, with first content have characteristic relation content be and associated second content of first content.Corresponding second obtains
Two kinds of alternative constructions of unit 702 are as shown in figure 8, wherein Fig. 8 shows another kind data processing disclosed in the embodiment of the present application
The structural block diagram of device, the second obtaining unit 702 includes: that the first acquisition module 7021 and second obtains module 7022 in fig. 8.
First obtains module 7021, for obtaining the first content with input information matches from default knowledge mapping.
Second obtains module 7022, has in particular kind of relationship for obtaining in default knowledge mapping with first content
Hold, with first content have characteristic relation content be and associated second content of first content.
In one embodiment, the second acquisition module 7022 is specifically used for obtaining first content in default knowledge mapping
Corresponding first node, default knowledge mapping record the attribute value of at least one attribute of at least one object, and default knowledge
Map is using the attribute value of object id information and every attribute of object as node, using the attribute of object as side connecting node.With
It is direct-connected to show first and for obtaining the second node that first node corresponding with institute's first content is direct-connected in default knowledge mapping
It is connected between node and second node by a line.And for obtaining direct-connected with second node in default knowledge mapping
Three nodes, using the corresponding content of third node as the content with first content with particular kind of relationship.Specific introduce please refers to
The explanation in embodiment of the method is stated, this embodiment is not repeated to this.
In another embodiment, second the first knowledge that module 7022 is specifically used in default knowledge mapping is obtained
One of map and the second knowledge mapping acquisition first node corresponding with first content, the first knowledge mapping and the second knowledge
One of map records the attribute value of at least one attribute of the first object, and at least one of another the second object of record belongs to
The attribute value of property, and each knowledge mapping is with the attribute of the object id information of each self-recording object and every attribute of object
Value is node, using the attribute of object as side connecting node.And the label information for obtaining first node, the mark of first node
Label information is used to show the object-oriented requirements of first content adaptation, and for the label information based on first node, knows first
Know in another map of map and the second knowledge mapping and obtain second node, using the corresponding content of second node as with first
Content has the content of particular kind of relationship, and the label information of second node and the label information of first node show the same meaning.
Specific to introduce the explanation please referred in above method embodiment, to this, this embodiment is not repeated.
And for for another embodiment, data processing equipment disclosed in the embodiment of the present application can also be in Fig. 8
On the basis of, the second obtaining unit 702 further includes merging module 7023, as shown in figure 9, wherein merging module 7023 is used for first
Node is connected with second node by a line, and the first knowledge mapping and the second knowledge mapping are constituted a knowledge mapping.
More preferably to illustrate the data processing equipment that above-described embodiment is related to, the present embodiment shows data processing equipment
A kind of feasible architecture diagram in practical applications may include: as shown in Figure 10 input module 1001, language understanding module
1002, knowledge mapping uses module 1003, knowledge relation reasoning module 1004, language generation module 1005 and output module
1006.Wherein input module 1001 and the corresponding above-mentioned first obtains unit 701 of language understanding module 1002, knowledge mapping use mould
Block 1003 and corresponding above-mentioned second obtaining unit 702 of knowledge relation reasoning module 1004, the correspondence of language generation module 1005 are above-mentioned
Third obtaining unit 703, the corresponding above-mentioned output unit 704 of output module 1006.
In practical applications, input module 1001 inputs problem M (a kind of input information), language into intelligent conversational system
Understanding Module 1002 extracts characteristic information from problem M, and marks emphasis word, for example, emphasis word mark product information,
Problem types etc. obtain recognition result O.
Knowledge mapping, according to recognition result O, is drawn using module 1003 from the fact that build knowledge data map and user
As searching the relationship between entity relevant to problem M and entity, such as weight, the screen of lookup mobile phone products in data map
The information such as the relevant parameter of size, the gender of user, height, weight, hand sizes.Based on by knowledge mapping use module 1003
Relationship R between the entity E and entity of acquisition, knowledge relation reasoning module 1004 make inferences to obtain the result for generating answer
Scheme G.Wherein fact knowledge data picture and user's representation data map are two knowledge mappings in default knowledge mapping.
Language generation module 1005 selects suitable language template according to result figure G, and the content in result figure G is filled out
Enter to generate the text answer A (i.e. feedback information) of structuring in the template, and passes through 1006 output character answer A of output module.
Above method embodiment is please referred to for the specific execution content of above-mentioned module, this present embodiment is no longer illustrated.
Furthermore a kind of electronic equipment, including processor and display component is also disclosed in the embodiment of the present application, and processor has intelligence
Energy conversational system, intelligent conversational system can respond the input information received and provide feedback information.
Processor, for obtaining input information, obtaining with the first content of input information matches and being closed with first content
Connection the second content, and based on first content and with associated second content of first content, obtain feedback information, input information be
The information of user's input during man-machine conversation is carried out by intelligent conversational system.Illustrating for function in processor is asked
Refering to above method embodiment, this present embodiment is no longer illustrated.
Display component, for exporting feedback information.
A kind of storage medium is also disclosed in the embodiment of the present application, and computer program code, computer are stored on storage medium
Program code realizes above-mentioned data processing method when executing.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
For device class embodiment, since it is basically similar to the method embodiment, so being described relatively simple, related place ginseng
See the part explanation of embodiment of the method.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
The foregoing description of the disclosed embodiments makes those skilled in the art can be realized or use the application.To this
A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can
Without departing from the spirit or scope of the application, to realize in other embodiments.Therefore, the application will not be limited
It is formed on the embodiments shown herein, and is to fit to consistent with the principles and novel features disclosed in this article widest
Range.
The above is only the preferred embodiment of the application, it is noted that for the ordinary skill people of the art
For member, under the premise of not departing from the application principle, several improvements and modifications can also be made, these improvements and modifications are also answered
It is considered as the protection scope of the application.
Claims (10)
1. a kind of data processing method, it is applied to intelligent conversational system, the intelligence conversational system being capable of input to receiving
Information is responded and provides feedback information, which comprises
Input information is obtained, the input information is that user inputs during carrying out man-machine conversation by the intelligent conversational system
Information;
Obtain with it is described input information matches first content and with associated second content of the first content;
Based on the first content and with associated second content of the first content, obtain feedback information;
Export the feedback information.
2. according to the method described in claim 1, one of the first content and second content are the first object
The attribute value of first attribute, another attribute value for the second attribute of first object, first attribute and described the
Two attributes are different;
Or
One of the first content and second content are the attribute value of the first attribute of the first object, another is the
The attribute value of first attribute of two objects, first object are different with second object.
3. according to the method described in claim 1, it is described acquisition with it is described input information matches first content and with it is described
Associated second content of first content includes:
The first content with the input information matches is obtained from default knowledge mapping;
In the default knowledge mapping obtain with the first content have particular kind of relationship content, it is described with described first in
Container have characteristic relation content be and associated second content of the first content.
4. according to the method described in claim 3, described obtain in the default knowledge mapping has with the first content
The content of particular kind of relationship includes:
Obtain the corresponding first node of the first content in the default knowledge mapping, the default knowledge mapping record to
The attribute value of at least one attribute of a few object, and the default knowledge mapping is with every kind of object id information and object
The attribute value of attribute is node, using the attribute of object as side connecting node;
The second node that first node corresponding with the first content is direct-connected in the default knowledge mapping is obtained, it is described direct-connected
Show to connect between first node and second node by a line;
Third node direct-connected with the second node in the default knowledge mapping is obtained, the third node is corresponding interior
Hold as the content with the first content with particular kind of relationship.
5. according to the method described in claim 3, described obtain in the default knowledge mapping has with the first content
The content of particular kind of relationship includes:
In the default knowledge mapping the first knowledge mapping and one of the second knowledge mapping obtain in described first
Hold corresponding first node, one of first knowledge mapping and second knowledge mapping record the first object at least
The attribute value of one attribute, the attribute value of at least one attribute of another the second object of record, and each knowledge mapping is with each
The attribute value of every attribute of the object id information and object of self-recording object is node, is connected by side of the attribute of object
Node;
The label information of the first node is obtained, the label information of the first node is used to show pair of first content adaptation
As demand;
Based on the label information of the first node, in another figure of first knowledge mapping and second knowledge mapping
Second node is obtained in spectrum, is had in particular kind of relationship using the corresponding content of the second node as with the first content
Hold, the label information of the label information of the second node and the first node shows the same meaning.
6. according to the method described in claim 5, the method also includes: the first node and the second node are passed through
First knowledge mapping and second knowledge mapping are constituted a knowledge mapping by a line connection.
7. according to the method described in claim 1, it is described based on the first content and with the first content associated second
Content, obtaining feedback information includes:
The feedback information is formed by the first content and second content;
Or
The first content and second content are modified, the feedback information is obtained.
8. a kind of data processing equipment, it is applied to intelligent conversational system, the intelligence conversational system being capable of input to receiving
Information is responded and is provided feedback information, and described device includes:
First obtains unit, for obtaining input information, the input information is man-machine by the intelligent conversational system progress
The information that user inputs in conversation procedure;
Second obtaining unit, for obtain with it is described input information matches first content and with the first content it is associated
Second content;
Third obtaining unit, for based on the first content and with associated second content of the first content, fed back
Information;
Output unit, for exporting the feedback information.
9. device according to claim 8, one of the first content and second content are the first object
The attribute value of first attribute, another attribute value for the second attribute of first object, first attribute and described the
Two attributes are different;
Or
One of the first content and second content are the attribute value of the first attribute of the first object, another is the
The attribute value of first attribute of two objects, first object are different with second object.
10. a kind of electronic equipment, including processor and display component, the processor has intelligent conversational system, the intelligence
Conversational system can respond the input information received and provide feedback information;
The processor is obtained with the first content of the input information matches and with described for obtaining input information
Associated second content of one content, and based on the first content and with associated second content of the first content, obtain anti-
Feedforward information, the input information are the information that user inputs during carrying out man-machine conversation by the intelligent conversational system;
The display component, for exporting the feedback information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811648577.8A CN109726279B (en) | 2018-12-30 | 2018-12-30 | Data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811648577.8A CN109726279B (en) | 2018-12-30 | 2018-12-30 | Data processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109726279A true CN109726279A (en) | 2019-05-07 |
CN109726279B CN109726279B (en) | 2021-05-18 |
Family
ID=66298552
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811648577.8A Active CN109726279B (en) | 2018-12-30 | 2018-12-30 | Data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109726279B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116795452A (en) * | 2023-07-20 | 2023-09-22 | 龙芯中科(北京)信息技术有限公司 | Method, device and equipment for determining compatibility of driving program |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104462505A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Search method and device |
CN104462504A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing reasoning process data in search |
CN104699845A (en) * | 2015-03-31 | 2015-06-10 | 北京奇虎科技有限公司 | Question-style search word based providing method and device of search results |
CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术(北京)有限公司 | Man-machine interaction method and system based on artificial intelligence |
CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
CN107066628A (en) * | 2017-05-15 | 2017-08-18 | 京东方科技集团股份有限公司 | Wear the clothes recommendation method and device |
CN107644093A (en) * | 2017-09-27 | 2018-01-30 | 上海斐讯数据通信技术有限公司 | Method and system are recommended in wearing the clothes for intelligent terminal |
CN107729493A (en) * | 2017-09-29 | 2018-02-23 | 北京创鑫旅程网络技术有限公司 | Travel the construction method of knowledge mapping, device and travelling answering method, device |
CN108108449A (en) * | 2017-12-27 | 2018-06-01 | 哈尔滨福满科技有限责任公司 | A kind of implementation method based on multi-source heterogeneous data question answering system and the system towards medical field |
CN108268582A (en) * | 2017-07-14 | 2018-07-10 | 广东神马搜索科技有限公司 | Information query method and device |
CN108595494A (en) * | 2018-03-15 | 2018-09-28 | 腾讯科技(深圳)有限公司 | The acquisition methods and device of reply message |
CN109002516A (en) * | 2018-07-06 | 2018-12-14 | 国网电子商务有限公司 | A kind of searching method and device |
-
2018
- 2018-12-30 CN CN201811648577.8A patent/CN109726279B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104462505A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Search method and device |
CN104462504A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing reasoning process data in search |
CN104699845A (en) * | 2015-03-31 | 2015-06-10 | 北京奇虎科技有限公司 | Question-style search word based providing method and device of search results |
CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术(北京)有限公司 | Man-machine interaction method and system based on artificial intelligence |
CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
CN107066628A (en) * | 2017-05-15 | 2017-08-18 | 京东方科技集团股份有限公司 | Wear the clothes recommendation method and device |
CN108268582A (en) * | 2017-07-14 | 2018-07-10 | 广东神马搜索科技有限公司 | Information query method and device |
CN107644093A (en) * | 2017-09-27 | 2018-01-30 | 上海斐讯数据通信技术有限公司 | Method and system are recommended in wearing the clothes for intelligent terminal |
CN107729493A (en) * | 2017-09-29 | 2018-02-23 | 北京创鑫旅程网络技术有限公司 | Travel the construction method of knowledge mapping, device and travelling answering method, device |
CN108108449A (en) * | 2017-12-27 | 2018-06-01 | 哈尔滨福满科技有限责任公司 | A kind of implementation method based on multi-source heterogeneous data question answering system and the system towards medical field |
CN108595494A (en) * | 2018-03-15 | 2018-09-28 | 腾讯科技(深圳)有限公司 | The acquisition methods and device of reply message |
CN109002516A (en) * | 2018-07-06 | 2018-12-14 | 国网电子商务有限公司 | A kind of searching method and device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116795452A (en) * | 2023-07-20 | 2023-09-22 | 龙芯中科(北京)信息技术有限公司 | Method, device and equipment for determining compatibility of driving program |
CN116795452B (en) * | 2023-07-20 | 2024-04-02 | 龙芯中科(北京)信息技术有限公司 | Method, device and equipment for determining compatibility of driving program |
Also Published As
Publication number | Publication date |
---|---|
CN109726279B (en) | 2021-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10635408B2 (en) | Method and apparatus for enabling agile development of services in cloud computing and traditional environments | |
CN107608652A (en) | A kind of method and apparatus of Voice command graphical interfaces | |
CN108664406A (en) | Exploitation test method, terminal device and the medium of mobile application | |
CN103294805B (en) | The creation method of data warehouse personalization dimension table and device | |
CN109815486A (en) | Spatial term method, apparatus, equipment and readable storage medium storing program for executing | |
CN110019838A (en) | Intelligent Answer System and intelligent terminal | |
CN110532487A (en) | The generation method and device of label | |
CN112180776A (en) | OPC UA information modeling method and device based on equipment component module | |
CN107704331A (en) | Electrical equipment fault analyzes and processes householder method and device | |
CN117271791A (en) | Method and system for optimizing knowledge graph based on large language model | |
CN106020601A (en) | Interface display management method and device | |
CN112149838A (en) | Method, device, electronic equipment and storage medium for realizing automatic model building | |
CN107301189A (en) | A kind of method for exhibiting data and device | |
CN109726279A (en) | A kind of data processing method and device | |
CN103593188B (en) | System and method for managing what-you-see-is-what-you-get menus of instrument | |
CN100390731C (en) | Interaction design system | |
CN110109675A (en) | Intelligent contract processing method, device and computer readable storage medium | |
CN103955369B (en) | Software fuzzy self-adaptation modeling tool construction method based on expanded UML (Unified Modeling Language) | |
CN102509019A (en) | Web television based fitting and shopping simulation method and system | |
CN106911634A (en) | A kind of login method and equipment | |
CN109919762A (en) | Reporting method, device, equipment and the storage medium of customer information | |
CN110019730A (en) | Automatic interaction system and intelligent terminal | |
CN106991124A (en) | Answer method and system based on how interactive situation editor | |
CN110287304A (en) | Question and answer information processing method, device and computer equipment | |
Kumar et al. | An approach for services design by understanding value requirements, identifying value carriers, developing value proposition, and subsequently realizing value |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |