CN109684453A - A kind of information processing method and electronic equipment - Google Patents
A kind of information processing method and electronic equipment Download PDFInfo
- Publication number
- CN109684453A CN109684453A CN201811601011.XA CN201811601011A CN109684453A CN 109684453 A CN109684453 A CN 109684453A CN 201811601011 A CN201811601011 A CN 201811601011A CN 109684453 A CN109684453 A CN 109684453A
- Authority
- CN
- China
- Prior art keywords
- information
- keyword
- described problem
- parsing result
- problem information
- 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
Abstract
The present disclosure discloses a kind of information processing method, it is applied in intelligent conversational system, intelligent conversational system to the input information response received and can provide feedback information；Method includes: information the problem of obtaining user's input, and response problem information exports feedback information；Wherein, feedback information includes at least the both keyword in problem information, and the sequence in feedback information between keyword is identical as sequence of the keyword in problem information.The problem of disclosure can be inputted according to user information exports diversified feedback information, effectively improves user experience.The disclosure also discloses a kind of intelligent conversational system.
Description
Technical field
This disclosure relates to electronic technology field more particularly to a kind of information processing method and electronic equipment.
Background technique
With the continuous development of electronic technology, human-computer interaction is also widely used.Currently, in intelligent customer service system
In system, it can be parsed first with natural language understanding for customer problem for true class problem, obtain corresponding structure
Neutralizing analysis is as a result, be then based on the reply that the structural texture combines words art template generation customer problem.
Existing this mode, the reply of generation is often excessively stiff, can not according to the specific question formulation of user to return
Multiple content is effectively changed, excessively mechanization, can not bring user's hommization, personalized experience.
Summary of the invention
In view of this, the disclosure provides a kind of information processing method, diversification can be generated according to the question formulation of user
Reply, effectively improve user experience.
Present disclose provides a kind of information processing methods, are applied in intelligent conversational system, the intelligence conversational system energy
It reaches to the input information response received and feedback information is provided；The described method includes:
The problem of obtaining user's input information；
Described problem information is responded, feedback information is exported；Wherein, the feedback information includes at least described problem information
In both keyword, and the sequence in the feedback information between keyword with the keyword in described problem information
Sequence it is identical.
Preferably, the response described problem information, output feedback information include:
Natural language understanding is carried out to described problem information, obtains corresponding structuring parsing result；
KnowledgeBase-query is carried out based on the structuring parsing result, obtains answer letter corresponding with described problem information
Breath；
Described problem information is handled；
The structuring parsing result is handled；
The answer information is handled；
Based on the processing result of described problem information, structuring parsing result and the answer information, described in output
Feedback information.
Preferably, it is described to described problem information carry out processing include:
Coded representation is carried out to described problem information, obtains the corresponding coding result of described problem information.
Preferably, it is described to the structuring parsing result carry out processing include:
Coded representation is carried out to the structuring parsing result, the structuring parsing result is obtained and encodes knot accordingly
Fruit.
Preferably, it is described to the answer information carry out processing include:
Coded representation is carried out to the answer information, obtains the corresponding coding result of the answer information.
Preferably, described that natural language understanding is carried out to described problem information, obtain corresponding structuring parsing result packet
It includes:
The semantic analytic modell analytical model obtained based on preparatory training carries out natural language understanding to described problem information, obtains phase
The structuring parsing result answered.
A kind of intelligence conversational system, comprising: memory, for storing caused by application program and application program operation
Data；
Input unit, for obtaining the problem of user inputs information；
Processor exports feedback information for running the application program to respond described problem information；Wherein, described
Feedback information includes at least the both keyword in described problem information, and the sequence in the feedback information between keyword
It is identical as sequence of the keyword in described problem information.
Preferably, the processor is specifically used for when exporting feedback information in response described problem information:
Natural language understanding is carried out to described problem information, obtains corresponding structuring parsing result；
KnowledgeBase-query is carried out based on the structuring parsing result, obtains answer letter corresponding with described problem information
Breath；
Described problem information is handled；
The structuring parsing result is handled；
The answer information is handled；
Based on the processing result of described problem information, structuring parsing result and the answer information, described in output
Feedback information.
Preferably, the processor is specifically used for when handling described problem information:
Coded representation is carried out to described problem information, obtains the corresponding coding result of described problem information.
Preferably, the processor is specifically used for when handling the structuring parsing result:
Coded representation is carried out to the structuring parsing result, the structuring parsing result is obtained and encodes knot accordingly
Fruit.
Preferably, the processor is specifically used for when handling the answer information:
Coded representation is carried out to the answer information, obtains the corresponding coding result of the answer information.
Preferably, the processor is carrying out natural language understanding to described problem information, obtains corresponding structure neutralizing
When analysing result, it is specifically used for:
The semantic analytic modell analytical model obtained based on preparatory training carries out natural language understanding to described problem information, obtains phase
The structuring parsing result answered.
It can be seen from the above technical proposal that a kind of data processing method disclosed in the disclosure, is applied to intelligent session system
In system, wherein intelligent conversational system to the input information response received and can provide feedback information；When needing to user
When the information of input is handled, the problem of user inputs information is obtained first, is then responding to problem information, exports feedback letter
Breath, wherein the feedback information of output includes at least the both keyword in problem information, and in feedback information between keyword
Sequence it is identical as sequence of the keyword in problem information.The problem of capable of being inputted according to user information output multiplicity
The feedback information of change, effectively improves user experience.
Detailed description of the invention
In order to illustrate more clearly of the embodiment of the present disclosure or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
Some embodiments of the present disclosure, for those of ordinary skill in the art, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 is a kind of flow chart of information processing method embodiment 1 disclosed in the disclosure；
Fig. 2 is a kind of flow chart of information processing method embodiment 2 disclosed in the disclosure；
Fig. 3 is a kind of structural schematic diagram of intelligent conversational system embodiment 1 disclosed in the disclosure；
Fig. 4 is a kind of structural schematic diagram of intelligent conversational system embodiment 2 disclosed in the disclosure.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present disclosure, the technical solution in the embodiment of the present disclosure is carried out clear, complete
Site preparation description, it is clear that described embodiment is only disclosure a part of the embodiment, instead of all the embodiments.It is based on
Embodiment in the disclosure, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment belongs to the range of disclosure protection.
As shown in Figure 1, for a kind of flow chart of information processing method embodiment 1 disclosed in the disclosure, the method application
In intelligent conversational system, that is, being applied to the input information response received and can provide the conversational system of feedback；It is described
Method may comprise steps of:
S101, the problem of user's input information is obtained；
When needing the problem of inputting to user information to handle, the problem of user inputs information is obtained first.It is obtaining
When the problem of taking family input information, the problem of equipment obtains user information can be obtained by sound and e.g. passes through microphone
The problem of obtaining user information.
Wherein, the problem of user inputs information can be the English problem information of user's input, be also possible to user's input
Chinese charater problem information etc..For example, the problem of user inputs information may is that " Does z2 work with mods? ",
" Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible with
Z2? "；Be also possible to: " z2 is suitable for mods? ", " z2 supports mods? ", " supporting z2mods? " or " mods
It is compatible with z2? ".
S102, response problem information export feedback information；Wherein, feedback information includes at least two in problem information
Keyword, and the sequence in feedback information between keyword is identical as sequence of the keyword in problem information.
When get user input the problem of information after, further to user input the problem of information respond, then
Export corresponding feedback information.Wherein, the feedback information of output includes at least the both keyword in problem information, and feeds back
Sequence in information between keyword is identical as sequence of the keyword in problem information.
By taking above-mentioned example as an example, when user input the problem of information be " Does z2work with mods? " when, response
The problem information, the feedback information of output are " Yes.Z2can work with mods. ", key in the feedback information of output
Word " work with mods " is included in problem information, and sequence between keyword and keyword are in problem information
It is sequentially identical；
When user input the problem of information be " Does z2support mods? " when, the problem information is responded, output
Feedback information is " Yes.Z2support mods. ", and keyword " Z2 support mods " is included in the feedback information of output
Sequence in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is z2mods supported? " when, the problem information is responded, output
Feedback information is " Yes.Z2is mods supported. ", keyword " mods supported " in the feedback information of output
Sequence included in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is mods compatible with z2? " when, the feedback information of output
For " Yes.Mods is compatible with Z2. ", keyword " compatible in the feedback information of output
Z2 " is included in problem information, and the sequence between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 is suitable for mods? " when, the feedback information of output is " yes.Z2
It can be adapted for mods.", keyword " suitable for mods " is included in problem information in the feedback information of output, and keyword
Between sequence it is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 supports mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " supporting mods " is included in problem information in the feedback information of output, and the sequence between keyword
It is identical as sequence of the keyword in problem information；
When user input the problem of information be " supporting z2mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " support " is included in problem information in the feedback information of output, and the sequence between keyword and pass
Sequence of the key word in problem information is identical；
When user input the problem of information be " mods is compatible with z2? " when, the feedback information of output is " yes.Mods
It is compatible with Z2 ", keyword " Mods is compatible with Z2 " is included in problem information in the feedback information of output, and between keyword
Sequence it is identical as sequence of the keyword in problem information.
And existing information processing manner is, when user input the problem of information be " Does z2work with
Mods? ", " Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible
With z2? " when, the feedback information of output is unified for " Yes, Z2supports mods. ".
In conclusion in the above-described embodiments, when needing the information inputted to user to handle, obtaining use first
The problem of family inputs information, is then responding to problem information, exports feedback information, wherein the feedback information of output is included at least and asked
Inscribe information in both keyword, and the sequence in feedback information between keyword with the keyword in problem information
It is sequentially identical.The problem of capable of being inputted according to user information exports diversified feedback information, effectively improves user experience.
As shown in Fig. 2, for a kind of flow chart of information processing method embodiment 2 disclosed in the disclosure, the method application
In intelligent conversational system, that is, being applied to the input information response received and can provide the conversational system of feedback；It is described
Method may comprise steps of:
S201, the problem of user's input information is obtained；
When needing the problem of inputting to user information to handle, the problem of user inputs information is obtained first.It is obtaining
When the problem of taking family input information, the problem of equipment obtains user information can be obtained by sound and e.g. passes through microphone
The problem of obtaining user information.
Wherein, the problem of user inputs information can be the English problem information of user's input, be also possible to user's input
Chinese charater problem information etc..For example, the problem of user inputs information may is that " Does z2 work with mods? ",
" Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible with
Z2? "；Be also possible to: " z2 is suitable for mods? ", " z2 supports mods? ", " supporting z2mods? " or " mods
It is compatible with z2? ".
S202, natural language understanding is carried out to problem information, obtains corresponding structuring parsing result；
After getting the problem of user inputs information, natural-sounding reason further is carried out to information the problem of getting
Solution, obtains corresponding structuring parsing result.
Specifically, when obtaining corresponding structured result, can be used carrying out natural language understanding to problem information
Major Natural language processing techniques have Text Classification and information extraction technique.In order to accomplish accurate user semantic solution
Analysis, it is necessary first to which information is manually marked aiming at the problem that historical user's input, inputs information for each user
Mark out corresponding attribute-name: Domain (domain name), Intent (intention), Sub-Intent (sub- intention) and Question-
Type (problem types) and Slot (slot).After completing data mark, need to start the training of semantic analytic modell analytical model, the mould
The input of type is originally inputted for user, exports as corresponding Domain, Intent, Sub-Intent and Question-
Type and Slot.For example, the problem of user inputs information is " Does Z2play support 5G? ", managed by natural language
Xie Hou can obtain structured result as shown in Table 1:
1 structuring parsing result of table
Semantic analytic modell analytical model can be with textual classification model be used, such as using SVM (Support Vector Machine, branch
Hold vector machine), CNN (Convolutional Neural Networks, convolutional neural networks) or LSTM (Long
Short-Term Memory, shot and long term memory network) etc.；Information Extraction Model can also be selected, such as CRF (conditional
Random field algorithm, condition random field), LSTM+CRF etc..In the disclosure, in order to obtain preferably semantic solution
Effect is analysed, textual classification model and Information Extraction Model can be subjected to effective fusion.For customer problem Q, by certainly
Right language understanding module can obtain a corresponding structuring parsing result U.
S203, KnowledgeBase-query is carried out based on structuring parsing result, obtains answer information corresponding with problem information；
The structuring that the problem of user exports information can obtain after natural language understanding understands as a result, based on the knot
Fruit can obtain the answer of customer problem by KnowledgeBase-query.As customer problem " Does Z2play support 5G? " through
After crossing the parsing of natural language understanding module, the result that search knowledge base obtains is Yes.It should be noted that due to knowledge inventory
The answer of storage be it is fine-grained, can be used as the component part for generating and replying, but reply customer problem cannot be directly used in
's.
S204, problem information is handled；
Then, the problem of further inputting to user information is handled.Specifically, the problem of being inputted to user
Information carries out coded representation, obtains the corresponding coding result of problem information.
S205, structuring parsing result is handled；
Meanwhile further obtained structuring parsing result is handled.It is tied specifically, can be parsed to structuring
Fruit carries out coded representation, obtains the corresponding coding result of structuring parsing result.
S206, answer information is handled；
Meanwhile it further handling to obtaining answer information.Specifically, coded representation can be carried out to answer information,
Obtain the corresponding coding result of answer information.
S207, the processing result for being based on problem information, structuring parsing result and answer information export feedback letter
Breath, wherein feedback information includes at least the both keyword in problem information, and the sequence in feedback information between keyword
It is identical as sequence of the keyword in described problem information.
Finally, by the corresponding coding result of problem information, the corresponding coding result of structuring parsing result and answer
The corresponding coding result of information exports feedback information as input.
By taking above-mentioned example as an example, when user input the problem of information be " Does z2work with mods? " when, response
The problem information, the feedback information of output are " Yes.Z2can work with mods. ", key in the feedback information of output
Word " work with mods " is included in problem information, and sequence between keyword and keyword are in problem information
It is sequentially identical；
When user input the problem of information be " Does z2support mods? " when, the problem information is responded, output
Feedback information is " Yes.Z2support mods. ", and keyword " Z2 support mods " is included in the feedback information of output
Sequence in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is z2mods supported? " when, the problem information is responded, output
Feedback information is " Yes.Z2is mods supported. ", keyword " mods supported " in the feedback information of output
Sequence included in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is mods compatible with z2? " when, the feedback information of output
For " Yes.Mods is compatible with Z2. ", keyword " compatible in the feedback information of output
Z2 " is included in problem information, and the sequence between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 is suitable for mods? " when, the feedback information of output is " yes.Z2
It can be adapted for mods.", keyword " suitable for mods " is included in problem information in the feedback information of output, and keyword
Between sequence it is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 supports mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " supporting mods " is included in problem information in the feedback information of output, and the sequence between keyword
It is identical as sequence of the keyword in problem information；
When user input the problem of information be " supporting z2mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " support " is included in problem information in the feedback information of output, and the sequence between keyword and pass
Sequence of the key word in problem information is identical；
When user input the problem of information be " mods is compatible with z2? " when, the feedback information of output is " yes.Mods
It is compatible with Z2 ", keyword " Mods is compatible with Z2 " is included in problem information in the feedback information of output, and between keyword
Sequence it is identical as sequence of the keyword in problem information.
And existing information processing manner is, when user input the problem of information be " Does z2work with
Mods? ", " Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible
With z2? " when, the feedback information of output is unified for " Yes, Z2supports mods. ".
In conclusion the reply different from the past based on template of technical solution disclosed in above-described embodiment generates scheme, it is first
It first passes through the parsing of natural language understanding module to be parsed to obtain the parsing result of structuring for customer problem, then pass through
Search knowledge base obtains the corresponding answer of customer problem, customer problem is finally combined, to generate the reply of customer problem.It should
Scheme can effectively avoid based on template generation reply mode it is single, excessively mechanization the shortcomings that.The program is asked in correct answer
On the basis of topic, reply mode can greatly be enriched, brings user's hommization, personalized experience.
In addition, it is three parts: customer problem, the structure of natural language understanding that the reply that the program proposes, which generates model,
Change parsing result and KnowledgeBase-query result.Various pieces play a crucial role the generation of reply, first
It is customer problem, directly using customer problem as one of the input generated is replied, has can establish customer problem and reply content
Direct correlation, it is ensured that reply mode and problem consistency；Structuring parsing result and KnowledgeBase-query result are made
The correctness for generating and replying can be effectively ensured to input.And reply generate model can by the accumulation of data, constantly into
Row learning improvement, therefore the program has good practicability and validity.
As shown in figure 3, for a kind of structural schematic diagram of intelligent conversational system embodiment 1 disclosed in the disclosure, the intelligence
Conversational system to the input information response received and can provide the conversational system of feedback；The intelligence conversational system can be with
Include:
Memory 301 runs generated data for storing application program and application program；
Input unit 302, for obtaining the problem of user inputs information；
When needing the problem of inputting to user information to handle, the problem of user inputs information is obtained first.It is obtaining
When the problem of taking family input information, the problem of equipment obtains user information can be obtained by sound and e.g. passes through microphone
The problem of obtaining user information.
Wherein, the problem of user inputs information can be the English problem information of user's input, be also possible to user's input
Chinese charater problem information etc..For example, the problem of user inputs information may is that " Does z2 work with mods? ",
" Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible with
Z2? "；Be also possible to: " z2 is suitable for mods? ", " z2 supports mods? ", " supporting z2mods? " or " mods
It is compatible with z2? ".
Processor 303 exports feedback information for running the application program with response problem information；Wherein, it feeds back
Information includes at least the both keyword in problem information, and sequence in feedback information between keyword and keyword are being asked
The sequence inscribed in information is identical.
When get user input the problem of information after, further to user input the problem of information respond, then
Export corresponding feedback information.Wherein, the feedback information of output includes at least the both keyword in problem information, and feeds back
Sequence in information between keyword is identical as sequence of the keyword in problem information.
By taking above-mentioned example as an example, when user input the problem of information be " Does z2work with mods? " when, response
The problem information, the feedback information of output are " Yes.Z2can work with mods. ", key in the feedback information of output
Word " work with mods " is included in problem information, and sequence between keyword and keyword are in problem information
It is sequentially identical；
When user input the problem of information be " Does z2support mods? " when, the problem information is responded, output
Feedback information is " Yes.Z2support mods. ", and keyword " Z2 support mods " is included in the feedback information of output
Sequence in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is z2mods supported? " when, the problem information is responded, output
Feedback information is " Yes.Z2is mods supported. ", keyword " mods supported " in the feedback information of output
Sequence included in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is mods compatible with z2? " when, the feedback information of output
For " Yes.Mods is compatible with Z2. ", keyword " compatible in the feedback information of output
Z2 " is included in problem information, and the sequence between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 is suitable for mods? " when, the feedback information of output is " yes.Z2
It can be adapted for mods.", keyword " suitable for mods " is included in problem information in the feedback information of output, and keyword
Between sequence it is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 supports mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " supporting mods " is included in problem information in the feedback information of output, and the sequence between keyword
It is identical as sequence of the keyword in problem information；
When user input the problem of information be " supporting z2mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " support " is included in problem information in the feedback information of output, and the sequence between keyword and pass
Sequence of the key word in problem information is identical；
When user input the problem of information be " mods is compatible with z2? " when, the feedback information of output is " yes.Mods
It is compatible with Z2 ", keyword " Mods is compatible with Z2 " is included in problem information in the feedback information of output, and between keyword
Sequence it is identical as sequence of the keyword in problem information.
And existing information processing manner is, when user input the problem of information be " Does z2work with
Mods? ", " Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible
With z2? " when, the feedback information of output is unified for " Yes, Z2supports mods. ".
In conclusion in the above-described embodiments, when needing the information inputted to user to handle, obtaining use first
The problem of family inputs information, is then responding to problem information, exports feedback information, wherein the feedback information of output is included at least and asked
Inscribe information in both keyword, and the sequence in feedback information between keyword with the keyword in problem information
It is sequentially identical.The problem of capable of being inputted according to user information exports diversified feedback information, effectively improves user experience.
As shown in figure 4, for a kind of structural schematic diagram of intelligent conversational system embodiment 2 disclosed in the disclosure, the intelligence
Conversational system to the input information response received and can provide the conversational system of feedback；The intelligence conversational system can be with
Include:
Memory 401 runs generated data for storing application program and application program；
Input unit 402, for obtaining the problem of user inputs information；
When needing the problem of inputting to user information to handle, the problem of user inputs information is obtained first.It is obtaining
When the problem of taking family input information, the problem of equipment obtains user information can be obtained by sound and e.g. passes through microphone
The problem of obtaining user information.
Wherein, the problem of user inputs information can be the English problem information of user's input, be also possible to user's input
Chinese charater problem information etc..For example, the problem of user inputs information may is that " Does z2 work with mods? ",
" Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible with
Z2? "；Be also possible to: " z2 is suitable for mods? ", " z2 supports mods? ", " supporting z2mods? " or " mods
It is compatible with z2? ".
Processor 403 obtains corresponding for running the application program to carry out natural language understanding to problem information
Structuring parsing result；
After getting the problem of user inputs information, natural-sounding reason further is carried out to information the problem of getting
Solution, obtains corresponding structuring parsing result.
Specifically, when obtaining corresponding structured result, can be used carrying out natural language understanding to problem information
Major Natural language processing techniques have Text Classification and information extraction technique.In order to accomplish accurate user semantic solution
Analysis, it is necessary first to which information is manually marked aiming at the problem that historical user's input, inputs information for each user
Mark out corresponding attribute-name: Domain (domain name), Intent (intention), Sub-Intent (sub- intention) and Question-
Type (problem types) and Slot (slot).After completing data mark, need to start the training of semantic analytic modell analytical model, the mould
The input of type is originally inputted for user, exports as corresponding Domain, Intent, Sub-Intent and Question-
Type and Slot.For example, the problem of user inputs information is " Does Z2 play support 5G? ", managed by natural language
Xie Hou can obtain structured result as shown in Table 1.
Semantic analytic modell analytical model can be with textual classification model be used, such as using SVM (Support Vector Machine, branch
Hold vector machine), CNN (Convolutional Neural Networks, convolutional neural networks) or LSTM (Long
Short-Term Memory, shot and long term memory network) etc.；Information Extraction Model can also be selected, such as CRF (conditional
Random field algorithm, condition random field), LSTM+CRF etc..In the disclosure, in order to obtain preferably semantic solution
Effect is analysed, textual classification model and Information Extraction Model can be subjected to effective fusion.For customer problem Q, by certainly
Right language understanding module can obtain a corresponding structuring parsing result U.
Processor 403 is also used to carry out KnowledgeBase-query based on structuring parsing result, obtain corresponding with problem information
Answer information；
The structuring that the problem of user exports information can obtain after natural language understanding understands as a result, based on the knot
Fruit can obtain the answer of customer problem by KnowledgeBase-query.As customer problem " Does Z2play support 5G? " through
After crossing the parsing of natural language understanding module, the result that search knowledge base obtains is Yes.It should be noted that due to knowledge inventory
The answer of storage be it is fine-grained, can be used as the component part for generating and replying, but reply customer problem cannot be directly used in
's.
Processor 403 is also used to handle problem information；
Then, the problem of further inputting to user information is handled.Specifically, the problem of being inputted to user
Information carries out coded representation, obtains the corresponding coding result of problem information.
Processor 403 is also used to handle structuring parsing result；
Meanwhile further obtained structuring parsing result is handled.It is tied specifically, can be parsed to structuring
Fruit carries out coded representation, obtains the corresponding coding result of structuring parsing result.
Processor 403 is also used to handle answer information；
Meanwhile it further handling to obtaining answer information.Specifically, coded representation can be carried out to answer information,
Obtain the corresponding coding result of answer information.
Processor 403 is also used to the processing result based on problem information, structuring parsing result and answer information,
Export feedback information, wherein feedback information includes at least the both keyword in problem information, and keyword in feedback information
Between sequence it is identical as sequence of the keyword in described problem information.
Finally, by the corresponding coding result of problem information, the corresponding coding result of structuring parsing result and answer
The corresponding coding result of information exports feedback information as input.
By taking above-mentioned example as an example, when user input the problem of information be " Does z2work with mods? " when, response
The problem information, the feedback information of output are " Yes.Z2can work with mods. ", key in the feedback information of output
Word " work with mods " is included in problem information, and sequence between keyword and keyword are in problem information
It is sequentially identical；
When user input the problem of information be " Does z2support mods? " when, the problem information is responded, output
Feedback information is " Yes.Z2support mods. ", and keyword " Z2 support mods " is included in the feedback information of output
Sequence in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is z2mods supported? " when, the problem information is responded, output
Feedback information is " Yes.Z2is mods supported. ", keyword " mods supported " in the feedback information of output
Sequence included in problem information, and between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " Is mods compatible with z2? " when, the feedback information of output
For " Yes.Mods is compatible with Z2. ", keyword " compatible in the feedback information of output
Z2 " is included in problem information, and the sequence between keyword is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 is suitable for mods? " when, the feedback information of output is " yes.Z2
It can be adapted for mods.", keyword " suitable for mods " is included in problem information in the feedback information of output, and keyword
Between sequence it is identical as sequence of the keyword in problem information；
When user input the problem of information be " z2 supports mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " supporting mods " is included in problem information in the feedback information of output, and the sequence between keyword
It is identical as sequence of the keyword in problem information；
When user input the problem of information be " supporting z2mods? " when, the feedback information of output is " yes.Z2 branch
Hold mods.", keyword " support " is included in problem information in the feedback information of output, and the sequence between keyword and pass
Sequence of the key word in problem information is identical；
When user input the problem of information be " mods is compatible with z2? " when, the feedback information of output is " yes.Mods
It is compatible with Z2 ", keyword " Mods is compatible with Z2 " is included in problem information in the feedback information of output, and between keyword
Sequence it is identical as sequence of the keyword in problem information.
And existing information processing manner is, when user input the problem of information be " Does z2work with
Mods? ", " Does z2support mods? ", " Is z2mods supported? " or " Is mods compatible
With z2? " when, the feedback information of output is unified for " Yes, Z2supports mods. ".
In conclusion the reply different from the past based on template of technical solution disclosed in above-described embodiment generates scheme, it is first
It first passes through the parsing of natural language understanding module to be parsed to obtain the parsing result of structuring for customer problem, then pass through
Search knowledge base obtains the corresponding answer of customer problem, customer problem is finally combined, to generate the reply of customer problem.It should
Scheme can effectively avoid based on template generation reply mode it is single, excessively mechanization the shortcomings that.The program is asked in correct answer
On the basis of topic, reply mode can greatly be enriched, brings user's hommization, personalized experience.
In addition, it is three parts: customer problem, the structure of natural language understanding that the reply that the program proposes, which generates model,
Change parsing result and KnowledgeBase-query result.Various pieces play a crucial role the generation of reply, first
It is customer problem, directly using customer problem as one of the input generated is replied, has can establish customer problem and reply content
Direct correlation, it is ensured that reply mode and problem consistency；Structuring parsing result and KnowledgeBase-query result are made
The correctness for generating and replying can be effectively ensured to input.And reply generate model can by the accumulation of data, constantly into
Row learning improvement, therefore the program has good practicability and validity.
Professional further appreciates that, list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, hard in order to clearly demonstrate
The interchangeability of part and software generally describes each exemplary composition and step according to function in the above description.
These functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this reality
Now it is not considered that exceeding the scope of the present disclosure.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly use hardware, processor
The combination of the software module or the two of execution is implemented.Software module can be placed in random access memory (RAM), memory, only
Read memory (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or
In any other form of storage medium well known in technical field.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the disclosure.
Various modifications to these embodiments will be readily apparent to those skilled in the art, defined herein
General Principle can be realized in other embodiments without departing from the spirit or the scope of the present disclosure.Therefore, originally
It is open to be not intended to be limited to the embodiments shown herein, and be to fit to special with principles disclosed herein and novelty
The consistent widest scope of point.
Claims (10)
1. a kind of information processing method is applied in intelligent conversational system, the intelligence conversational system can be defeated to what is received
Enter information response and feedback information is provided；The described method includes:
The problem of obtaining user's input information；
Described problem information is responded, feedback information is exported；Wherein, the feedback information includes at least two in described problem information
A keyword, and sequence phase of the sequence in the feedback information between keyword with the keyword in described problem information
Together.
2. according to the method described in claim 1, the response described problem information, output feedback information include:
Natural language understanding is carried out to described problem information, obtains corresponding structuring parsing result；
KnowledgeBase-query is carried out based on the structuring parsing result, obtains answer information corresponding with described problem information；
Described problem information is handled；
The structuring parsing result is handled；
The answer information is handled；
Based on the processing result of described problem information, structuring parsing result and the answer information, the feedback letter is exported
Breath.
3. according to the method described in claim 2, it is described to described problem information carry out processing include:
Coded representation is carried out to described problem information, obtains the corresponding coding result of described problem information.
4. according to the method described in claim 2, it is described to the structuring parsing result carry out processing include:
Coded representation is carried out to the structuring parsing result, obtains the corresponding coding result of structuring parsing result.
5. according to the method described in claim 2, it is described to the answer information carry out processing include:
Coded representation is carried out to the answer information, obtains the corresponding coding result of the answer information.
6. being tied accordingly according to the method described in claim 2, described carry out natural language understanding to described problem information
Structure parsing result includes:
The semantic analytic modell analytical model obtained based on preparatory training carries out natural language understanding to described problem information, is tied accordingly
Structure parsing result.
7. a kind of intelligence conversational system, comprising: memory, for storing number caused by application program and application program operation
According to；
Input unit, for obtaining the problem of user inputs information；
Processor exports feedback information for running the application program to respond described problem information；Wherein, the feedback
Information include at least described problem information in both keyword, and the sequence in the feedback information between keyword with it is described
Sequence of the keyword in described problem information is identical.
8. system according to claim 7, the processor is in response described problem information, when exporting feedback information, tool
Body is used for:
Natural language understanding is carried out to described problem information, obtains corresponding structuring parsing result；
KnowledgeBase-query is carried out based on the structuring parsing result, obtains answer information corresponding with described problem information；
Described problem information is handled；
The structuring parsing result is handled；
The answer information is handled；
Based on the processing result of described problem information, structuring parsing result and the answer information, the feedback letter is exported
Breath.
9. system according to claim 8, the processor is specifically used for when handling described problem information:
Coded representation is carried out to described problem information, obtains the corresponding coding result of described problem information.
10. system according to claim 8, the processor is when handling the structuring parsing result, tool
Body is used for:
Coded representation is carried out to the structuring parsing result, obtains the corresponding coding result of structuring parsing result；
Or
The processor is specifically used for when handling the answer information:
Coded representation is carried out to the answer information, obtains the corresponding coding result of the answer information；
Or
The processor is carrying out natural language understanding to described problem information, when obtaining corresponding structuring parsing result, tool
Body is used for:
The semantic analytic modell analytical model obtained based on preparatory training carries out natural language understanding to described problem information, is tied accordingly
Structure parsing result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811601011.XA CN109684453A (en) | 2018-12-26 | 2018-12-26 | A kind of information processing method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811601011.XA CN109684453A (en) | 2018-12-26 | 2018-12-26 | A kind of information processing method and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109684453A true CN109684453A (en) | 2019-04-26 |
Family
ID=66189642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811601011.XA Pending CN109684453A (en) | 2018-12-26 | 2018-12-26 | A kind of information processing method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109684453A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07261653A (en) * | 1994-03-25 | 1995-10-13 | Hitachi Ltd | Practice question forming and assisting system |
| CN1794233A (en) * | 2005-12-28 | 2006-06-28 | 刘文印 | Network user interactive asking answering method and its system |
| CN102866990A (en) * | 2012-08-20 | 2013-01-09 | 北京搜狗信息服务有限公司 | Thematic conversation method and device |
| CN103425714A (en) * | 2012-05-25 | 2013-12-04 | 北京搜狗信息服务有限公司 | Query method and system |
| CN106469212A (en) * | 2016-09-05 | 2017-03-01 | 北京百度网讯科技有限公司 | Man-machine interaction method based on artificial intelligence and device |
| CN107918634A (en) * | 2017-06-27 | 2018-04-17 | 上海壹账通金融科技有限公司 | Intelligent answer method, apparatus and computer-readable recording medium |
| CN107967302A (en) * | 2017-11-08 | 2018-04-27 | 江苏名通信息科技有限公司 | Game customer service conversational system based on deep neural network |
-
2018
- 2018-12-26 CN CN201811601011.XA patent/CN109684453A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07261653A (en) * | 1994-03-25 | 1995-10-13 | Hitachi Ltd | Practice question forming and assisting system |
| CN1794233A (en) * | 2005-12-28 | 2006-06-28 | 刘文印 | Network user interactive asking answering method and its system |
| CN103425714A (en) * | 2012-05-25 | 2013-12-04 | 北京搜狗信息服务有限公司 | Query method and system |
| CN102866990A (en) * | 2012-08-20 | 2013-01-09 | 北京搜狗信息服务有限公司 | Thematic conversation method and device |
| CN106469212A (en) * | 2016-09-05 | 2017-03-01 | 北京百度网讯科技有限公司 | Man-machine interaction method based on artificial intelligence and device |
| CN107918634A (en) * | 2017-06-27 | 2018-04-17 | 上海壹账通金融科技有限公司 | Intelligent answer method, apparatus and computer-readable recording medium |
| CN107967302A (en) * | 2017-11-08 | 2018-04-27 | 江苏名通信息科技有限公司 | Game customer service conversational system based on deep neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200301954A1 (en) | Reply information obtaining method and apparatus | |
| CN107204184B (en) | Audio recognition method and system | |
| CN106776544B (en) | Character relation recognition method and device and word segmentation method | |
| CN104598579B (en) | A kind of automatic question-answering method and system | |
| CN109241330A (en) | The method, apparatus, equipment and medium of key phrase in audio for identification | |
| CN109977361A (en) | A kind of Chinese phonetic alphabet mask method, device and storage medium based on similar word | |
| CN111160041B (en) | Semantic understanding method and device, electronic equipment and storage medium | |
| CN112633003A (en) | Address recognition method and device, computer equipment and storage medium | |
| CN111694937A (en) | Interviewing method and device based on artificial intelligence, computer equipment and storage medium | |
| CN110019712A (en) | More intent query method and apparatus, computer equipment and computer readable storage medium | |
| CN113590810B (en) | Abstract generation model training method, abstract generation device and electronic equipment | |
| CN112328761A (en) | Intention label setting method and device, computer equipment and storage medium | |
| CN112463942A (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
| CN108304387B (en) | Method, device, server group and storage medium for recognizing noise words in text | |
| CN110246494A (en) | Service request method, device and computer equipment based on speech recognition | |
| CN110019714A (en) | More intent query method, apparatus, equipment and storage medium based on historical results | |
| Archer | Internationalisation, technology and translation | |
| CN113312924A (en) | Risk rule classification method and device based on NLP high-precision analysis label | |
| CN107688594B (en) | The identifying system and method for risk case based on social information | |
| WO2021211300A1 (en) | System and method for summerization of customer interaction | |
| CN112966076A (en) | Intelligent question and answer generating method and device, computer equipment and storage medium | |
| CN110489740A (en) | Semantic analytic method and Related product | |
| CN111046674A (en) | Semantic understanding method and device, electronic equipment and storage medium | |
| CN109684453A (en) | A kind of information processing method and electronic equipment | |
| CN110321129A (en) | Business based on developing instrument executes method, apparatus and computer 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 |