CN108549662A - The supplement digestion procedure and device of semantic analysis result in more wheel sessions - Google Patents
The supplement digestion procedure and device of semantic analysis result in more wheel sessions Download PDFInfo
- Publication number
- CN108549662A CN108549662A CN201810219417.5A CN201810219417A CN108549662A CN 108549662 A CN108549662 A CN 108549662A CN 201810219417 A CN201810219417 A CN 201810219417A CN 108549662 A CN108549662 A CN 108549662A
- Authority
- CN
- China
- Prior art keywords
- analysis result
- supplement
- semantic analysis
- resolution
- content
- 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
- Machine Translation (AREA)
Abstract
The present invention relates to the supplement digestion procedures and device of semantic analysis result in a kind of more wheel sessions, wherein method includes：Semantic parsing is carried out to the first problem input by user got, and the target domain belonging to first problem is determined according to the obtain first semantic analysis result；The knowledge mapping for obtaining target domain determines the corresponding entity link in the knowledge mapping of target domain of the entity in the first semantic analysis result；Entity and corresponding entity link are recorded as conversation history information；When getting the Second Problem of user's new round input, determines to whether there is in the corresponding second semantic analysis result of Second Problem and wait for supplement resolution content；When it is present, supplement resolution content is treated according to conversation history information and carries out supplement resolution.With this solution, the accuracy rate that can improve the supplement resolution of the corresponding semantic analysis result of the problem of user proposes under specific area, improves the accuracy for the problem of system answer user proposes, promotes the usage experience of user.
Description
Technical field
The present invention relates to a kind of supplements of semantic analysis result in technical field of voice recognition more particularly to more wheel sessions to disappear
Solve method and device.
Background technology
In interactive question and answer, people can establish corresponding context while puing question to, following in this context
The problem of, it often will appear reference and omission, thereby result in single the problem of puing question to semantic missing, brought to problem retrieval
A large amount of trouble.
In the related technology, in the supplement resolution of the problem of progress, the size of accuracy rate and corpus dependent on participle, this
Sample, the accuracy rate that resolution is supplemented under specific area is inadequate, this may result in the response message of the problem next found not
The problems such as accurate.
Invention content
The embodiment of the present invention provides a kind of supplement digestion procedure and device for taking turns semantic analysis result in session more, to reality
The accuracy rate of the supplement resolution of the corresponding semantic analysis result of the problem of user proposes under specific area is now improved, and then improves system
System answers the accuracy for the problem of user proposes, promotes the usage experience of user.
According to a first aspect of the embodiments of the present invention, a kind of supplement resolution side for taking turns semantic analysis result in session is provided more
Method, which is characterized in that including：
Semantic parsing is carried out to the first problem input by user got, and according to the obtain first semantic analysis result
Determine the target domain belonging to the first problem；
The knowledge mapping of the target domain is obtained, and determines the entity in the described first semantic analysis result in the mesh
Corresponding entity link in the knowledge mapping in mark field；
The entity and corresponding entity link are recorded as conversation history information；
When getting the Second Problem of user's new round input, the corresponding second semantic parsing of the Second Problem is determined
As a result it whether there is in and wait for supplement resolution content；
Exist when supplement clears up content, according to the conversation history information when determining in the described second semantic analysis result
Wait for that supplement resolution content carries out supplement resolution to described.
In this embodiment, the target domain belonging to it is determined according to the corresponding first semantic analysis result of first problem,
And then the corresponding entity link of entity in the first semantic analysis result is determined according to the knowledge mapping of the target domain, and record
In being recorded to conversation history, in this way, when subsequently having the Second Problem of a new round again, it can be recorded according to conversation history and be asked second
Wait for that supplement resolution content carries out supplement resolution in topic.In this way, it is contemplated that the logic in the specific and domain knowledge of domain knowledge
Relationship is promoted independent of basis NLP (Natural Language Processing, natural language processing) technologies such as participles
The accuracy rate of supplement resolution.Since the accuracy rate of semantic analysis result improves, so subsequently according to semantic analysis result
When searching for corresponding response message, the accuracy rate of obtained response message also improves, and improves the usage experience of user.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement resolution content, including：
It whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
When determining in the described second semantic analysis result comprising word is referred to, the reference is identified using first
The reference word is determined as waiting for supplement resolution content by word.
In this embodiment it is possible to according to CRF (conditional random field algorithm, condition random
Algorithm) the reference word in the semantic analysis result of mark first carries out reference discovery, and to referring to word in semantic analysis result
It is identified, to convenient follow-up progress reference resolution.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement clears up content, further includes：
Whether the semanteme that the described second semantic analysis result is detected according to the knowledge mapping of the target domain is complete；
When detecting that the semanteme of the described second semantic analysis result is imperfect, missing portion is identified using second
Point, the lack part is determined as to wait for supplement resolution content.
In this embodiment, in addition to referring to word, semantic incomplete situation may be there is also in semantic analysis result.
It when semantic imperfect, is identified using the notation methods different from word is referred to, is subsequently corresponded to according to mark to convenient
Supplement resolution.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the first label, the references word corresponding target entity link is determined according to the conversation history information；
Wait for that supplement clears up content using described in target entity link replacement, it is corresponding most to obtain the Second Problem
Semanteme analysis result eventually.
In this embodiment, if waiting for that supplement resolution content refers to pronoun, it can be determined and be referred to according to conversation history information
The corresponding target entity link of pronoun, replaces thereby using target entity link and waits for supplement resolution content, in this way, just having obtained complete
Whole final semantic analysis result subsequently can search for corresponding response message, in turn according to this final semantic analysis result
Accurate response message is exported for user.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the second label, using all entity links for being recorded in the conversation history information replace described in wait supplementing
Content is cleared up, to obtain at least one semantic analysis result；
The target semanteme analysis result from being screened at least one semantic analysis result, the target semanteme is parsed
As a result it is determined as the corresponding final semantic analysis result of the Second Problem.
In this embodiment, if waiting for that supplement resolution content is semantic missing, the reality that can be above recorded with direct splicing
Body links, and filters out most rational semantic results as final semantic analysis result from multiple results spliced and obtained.This
Sample has just obtained complete final semantic analysis result, subsequently can be corresponding according to this final semantic analysis result search
Response message, and then export accurate response message for user.
According to a second aspect of the embodiments of the present invention, a kind of supplement resolution dress for taking turns semantic analysis result in session is provided more
It sets, including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
Semantic parsing is carried out to the first problem input by user got, and according to the obtain first semantic analysis result
Determine the target domain belonging to the first problem；
The knowledge mapping of the target domain is obtained, and determines the entity in the described first semantic analysis result in the mesh
Corresponding entity link in the knowledge mapping in mark field；
The entity and corresponding entity link are recorded as conversation history information；
When getting the Second Problem of user's new round input, the corresponding second semantic parsing of the Second Problem is determined
As a result it whether there is in and wait for supplement resolution content；
Exist when supplement clears up content, according to the conversation history information when determining in the described second semantic analysis result
Wait for that supplement resolution content carries out supplement resolution to described.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement resolution content, including：
It whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
When determining in the described second semantic analysis result comprising word is referred to, the reference is identified using first
The reference word is determined as waiting for supplement resolution content by word.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement clears up content, further includes：
Whether the semanteme that the described second semantic analysis result is detected according to the knowledge mapping of the target domain is complete；
When detecting that the semanteme of the described second semantic analysis result is imperfect, missing portion is identified using second
Point, the lack part is determined as to wait for supplement resolution content.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the first label, the references word corresponding target entity link is determined according to the conversation history information；
Wait for that supplement clears up content using described in target entity link replacement, it is corresponding most to obtain the Second Problem
Semanteme analysis result eventually.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the second label, using all entity links for being recorded in the conversation history information replace described in wait supplementing
Content is cleared up, to obtain at least one semantic analysis result；
The target semanteme analysis result from being screened at least one semantic analysis result, the target semanteme is parsed
As a result it is determined as the corresponding final semantic analysis result of the Second Problem.
It should be understood that above general description and following detailed description is only exemplary and explanatory, not
It can the limitation present invention.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that understand through the implementation of the invention.The purpose of the present invention and other advantages can be by the explanations write
Specifically noted structure is realized and is obtained in book, claims and attached drawing.
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
Description of the drawings
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the present invention
Example, and be used to explain the principle of the present invention together with specification.
Fig. 1 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
Flow chart.
Fig. 2 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
The flow chart of middle step S104.
Fig. 3 is the supplement resolution side according to semantic analysis result in the more wheel sessions of another kind shown in an exemplary embodiment
The flow chart of step S104 in method.
Fig. 4 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
The flow chart of middle step S105.
Fig. 5 is the supplement resolution side according to semantic analysis result in the more wheel sessions of another kind shown in an exemplary embodiment
The flow chart of step S105 in method.
Specific implementation mode
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects being described in detail in claims, of the invention.
Fig. 1 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
Flow chart.The supplement digestion procedure of semantic analysis result can be applied in terminal device in more wheel sessions, the terminal device
Can be mobile phone, computer, digital broadcast terminal, messaging devices, game console, tablet device, Medical Devices,
Body-building equipment, any equipment with phonetic function such as personal digital assistant.As shown in Figure 1, the method comprising the steps of S101-
S105：
In step S101, semantic parsing carried out to the first problem input by user that gets, and according to the obtained
One semantic analysis result determines the target domain belonging to the first problem；Each problem can have affiliated field, such as music
Field, sports field, information security field etc..
In step s 102, the knowledge mapping of the target domain is obtained, and is determined in the described first semantic analysis result
Entity in the knowledge mapping of the target domain corresponding entity link；
The knowledge mapping of target domain belonging to problem determines the entity link of semantic analysis result, and then is recorded as
Conversation history information, convenient for subsequently carrying out supplement resolution.
In step s 103, the entity and corresponding entity link are recorded as conversation history information；
In step S104, when getting the Second Problem of user's new round input, determine that the Second Problem corresponds to
The second semantic analysis result in the presence or absence of waiting for supplement resolution content；
In step S105, exist when supplement clears up content, according to institute when determining in the described second semantic analysis result
It states conversation history information and waits for that supplement resolution content carries out supplement resolution to described.
In this embodiment, the target domain belonging to it is determined according to the corresponding first semantic analysis result of first problem,
And then the corresponding entity link of entity in the first semantic analysis result is determined according to the knowledge mapping of the target domain, and record
In being recorded to conversation history, in this way, when subsequently having the Second Problem of a new round again, it can be recorded according to conversation history and be asked second
Wait for that supplement resolution content carries out supplement resolution in topic.In this way, it is contemplated that the logic in the specific and domain knowledge of domain knowledge
Relationship, independent of participle etc. basis NLP (Neuro-Linguistic Programming, neural LISP program LISP) technology,
Improve the accuracy rate of supplement resolution.Since the accuracy rate of semantic analysis result improves, so subsequently being parsed according to semanteme
When as a result searching for corresponding response message, the accuracy rate of obtained response message also improves, and improves the usage experience of user.
Fig. 2 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
The flow chart of middle step S104.
As shown in Fig. 2, in one embodiment, above-mentioned steps S104 includes step S201-S202：
In step s 201, it whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
Refer to word can be him, she, we, I, you, it, they, the case, here, in this way, this, this thing, these, those etc..
In step S202, when determining in the described second semantic analysis result comprising word is referred to, marked using the first label
Know and the reference word, the reference word is determined as to wait for supplement resolution content.
In this embodiment it is possible to according to CRF (conditional random field algorithm, condition random
Algorithm) the reference word in the semantic analysis result of mark first carries out reference discovery, and to referring to word in semantic analysis result
It is identified, to convenient follow-up progress reference resolution.
Fig. 3 is the supplement resolution side according to semantic analysis result in the more wheel sessions of another kind shown in an exemplary embodiment
The flow chart of step S104 in method.
As shown in figure 3, in one embodiment, above-mentioned steps S104 further includes step S301-S302：
In step S301, the semanteme of the described second semantic analysis result is detected according to the knowledge mapping of the target domain
It is whether complete；
In step s 302, when detecting that the semanteme of the described second semantic analysis result is imperfect, the second label is used
Lack part is identified, the lack part is determined as to wait for supplement resolution content.
In this embodiment, in addition to referring to word, semantic incomplete situation may be there is also in semantic analysis result.
It when semantic imperfect, is identified using the notation methods different from word is referred to, is subsequently corresponded to according to mark to convenient
Supplement resolution.
Fig. 4 is a kind of supplement digestion procedures for taking turns semantic analysis result in sessions shown according to an exemplary embodiment more
The flow chart of middle step S105.
As shown in figure 4, in one embodiment, above-mentioned steps S105 includes step S401-S402：
In step S401, supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and described the
When mark in two semantic analysis results includes the first label, determine that the reference word is corresponding according to the conversation history information
Target entity links；
In step S402, using target entity link replace described in wait for supplement resolution content, to obtain described the
The corresponding final semantic analysis result of two problems.
In this embodiment, if waiting for that supplement resolution content refers to pronoun, it can be determined and be referred to according to conversation history information
The corresponding target entity link of pronoun, replaces thereby using target entity link and waits for supplement resolution content, in this way, just having obtained complete
Whole final semantic analysis result subsequently can search for corresponding response message, in turn according to this final semantic analysis result
Accurate response message is exported for user.
Fig. 5 is the supplement resolution side according to semantic analysis result in the more wheel sessions of another kind shown in an exemplary embodiment
The flow chart of step S105 in method.
As shown in figure 5, in one embodiment, above-mentioned steps S105 further includes step S501-S502：：
In step S501, supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and described the
When mark in two semantic analysis results includes the second label, all entity links recorded in the conversation history information are used
Supplement resolution content is waited for described in replacement, to obtain at least one semantic analysis result；
In step S502, the target semanteme analysis result from being screened at least one semantic analysis result, by institute
It states target semanteme analysis result and is determined as the corresponding final semantic analysis result of the Second Problem.
In this embodiment, if waiting for that supplement resolution content is semantic missing, the reality that can be above recorded with direct splicing
Body links, and filters out most rational semantic results as final semantic analysis result from multiple results spliced and obtained.This
Sample has just obtained complete final semantic analysis result, subsequently can be corresponding according to this final semantic analysis result search
Response message, and then export accurate response message for user.
Following is apparatus of the present invention embodiment, can be used for executing the method for the present invention embodiment.
According to a second aspect of the embodiments of the present invention, a kind of supplement resolution dress for taking turns semantic analysis result in session is provided more
It sets, including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
Semantic parsing is carried out to the first problem input by user got, and according to the obtain first semantic analysis result
Determine the target domain belonging to the first problem；
The knowledge mapping of the target domain is obtained, and determines the entity in the described first semantic analysis result in the mesh
Corresponding entity link in the knowledge mapping in mark field；
The entity and corresponding entity link are recorded as conversation history information；
When getting the Second Problem of user's new round input, the corresponding second semantic parsing of the Second Problem is determined
As a result it whether there is in and wait for supplement resolution content；
Exist when supplement clears up content, according to the conversation history information when determining in the described second semantic analysis result
Wait for that supplement resolution content carries out supplement resolution to described.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement resolution content, including：
It whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
When determining in the described second semantic analysis result comprising word is referred to, the reference is identified using first
The reference word is determined as waiting for supplement resolution content by word.
In one embodiment, it whether there is in the corresponding second semantic analysis result of the determination Second Problem and wait for
Supplement clears up content, further includes：
Whether the semanteme that the described second semantic analysis result is detected according to the knowledge mapping of the target domain is complete；
When detecting that the semanteme of the described second semantic analysis result is imperfect, missing portion is identified using second
Point, the lack part is determined as to wait for supplement resolution content.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the first label, the references word corresponding target entity link is determined according to the conversation history information；
Wait for that supplement clears up content using described in target entity link replacement, it is corresponding most to obtain the Second Problem
Semanteme analysis result eventually.
It is in one embodiment, described when determining in the described second semantic analysis result exist when supplement clears up content,
Wait for that supplement resolution content carries out supplement resolution to described according to the conversation history information, including：
Supplement resolution content, and the described second semantic analysis result are waited for when determining in the described second semantic analysis result to exist
In mark when including the second label, using all entity links for being recorded in the conversation history information replace described in wait supplementing
Content is cleared up, to obtain at least one semantic analysis result；
The target semanteme analysis result from being screened at least one semantic analysis result, the target semanteme is parsed
As a result it is determined as the corresponding final semantic analysis result of the Second Problem.
It should be understood by those skilled in the art that, the embodiment of the present invention can be provided as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, the present invention can be used in one or more wherein include computer usable program code computer
The shape for the computer program product implemented in usable storage medium (including but not limited to magnetic disk storage and optical memory etc.)
Formula.
The present invention be with reference to according to the method for the embodiment of the present invention, the flow of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided
Instruct the processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine so that the instruction executed by computer or the processor of other programmable data processing devices is generated for real
The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that instruction generation stored in the computer readable memory includes referring to
Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device so that count
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, in computer or
The instruction executed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
God and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. the supplement digestion procedure of semantic analysis result in a kind of more wheel sessions, which is characterized in that including：
Semantic parsing is carried out to the first problem input by user got, and is determined according to the obtain first semantic analysis result
Target domain belonging to the first problem；
The knowledge mapping of the target domain is obtained, and determines that the entity in the described first semantic analysis result is led in the target
Corresponding entity link in the knowledge mapping in domain；
The entity and corresponding entity link are recorded as conversation history information；
When getting the Second Problem of user's new round input, the corresponding second semantic analysis result of the Second Problem is determined
In with the presence or absence of wait for supplement resolution content；
Exist when supplement clears up content, according to the conversation history information to institute when determining in the described second semantic analysis result
It states and waits for that supplement resolution content carries out supplement resolution.
2. the supplement digestion procedure of semantic analysis result in more wheel sessions according to claim 1, which is characterized in that described
It determines to whether there is in the corresponding second semantic analysis result of the Second Problem and waits for supplement resolution content, including：
It whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
When determining in the described second semantic analysis result comprising word is referred to, the reference word is identified using first, it will
The reference word is determined as waiting for supplement resolution content.
3. the supplement digestion procedure of semantic analysis result in more wheel sessions according to claim 2, which is characterized in that described
It determines in the corresponding second semantic analysis result of the Second Problem with the presence or absence of supplement resolution content is waited for, further includes：
Whether the semanteme that the described second semantic analysis result is detected according to the knowledge mapping of the target domain is complete；
When detecting that the semanteme of the described second semantic analysis result is imperfect, lack part is identified using second, it will
The lack part is determined as waiting for supplement resolution content.
4. the supplement digestion procedure of semantic analysis result in more wheel sessions according to claim 2, which is characterized in that described
Exist when supplement clears up content when determining in the described second semantic analysis result, is waited for described according to the conversation history information
Supplement resolution content carries out supplement resolution, including：
Supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and in the described second semantic analysis result
When mark is comprising the first label, determine that the corresponding target entity of the reference word links according to the conversation history information；
Using supplement resolution content is waited for described in target entity link replacement, to obtain the corresponding final language of the Second Problem
Adopted analysis result.
5. the supplement digestion procedure of semantic analysis result in more wheel sessions according to claim 3, which is characterized in that described
Exist when supplement clears up content when determining in the described second semantic analysis result, is waited for described according to the conversation history information
Supplement resolution content carries out supplement resolution, including：
Supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and in the described second semantic analysis result
When mark is comprising the second label, supplement resolution is waited for using described in all entity links replacement recorded in the conversation history information
Content, to obtain at least one semantic analysis result；
The target semanteme analysis result from being screened at least one semantic analysis result, by the target semanteme analysis result
It is determined as the corresponding final semantic analysis result of the Second Problem.
6. the supplement digestion instrument of semantic analysis result in a kind of more wheel sessions, which is characterized in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
Semantic parsing is carried out to the first problem input by user got, and is determined according to the obtain first semantic analysis result
Target domain belonging to the first problem；
The knowledge mapping of the target domain is obtained, and determines that the entity in the described first semantic analysis result is led in the target
Corresponding entity link in the knowledge mapping in domain；
The entity and corresponding entity link are recorded as conversation history information；
When getting the Second Problem of user's new round input, the corresponding second semantic analysis result of the Second Problem is determined
In with the presence or absence of wait for supplement resolution content；
Exist when supplement clears up content, according to the conversation history information to institute when determining in the described second semantic analysis result
It states and waits for that supplement resolution content carries out supplement resolution.
7. the supplement digestion instrument of semantic analysis result in more wheel sessions according to claim 6, which is characterized in that described
It determines to whether there is in the corresponding second semantic analysis result of the Second Problem and waits for supplement resolution content, including：
It whether determines in the corresponding second semantic analysis result of the Second Problem comprising reference word；
When determining in the described second semantic analysis result comprising word is referred to, the reference word is identified using first, it will
The reference word is determined as waiting for supplement resolution content.
8. the supplement digestion instrument of semantic analysis result in more wheel sessions according to claim 7, which is characterized in that described
It determines in the corresponding second semantic analysis result of the Second Problem with the presence or absence of supplement resolution content is waited for, further includes：
Whether the semanteme that the described second semantic analysis result is detected according to the knowledge mapping of the target domain is complete；
When detecting that the semanteme of the described second semantic analysis result is imperfect, lack part is identified using second, it will
The lack part is determined as waiting for supplement resolution content.
9. the supplement digestion instrument of semantic analysis result in more wheel sessions according to claim 7, which is characterized in that described
Exist when supplement clears up content when determining in the described second semantic analysis result, is waited for described according to the conversation history information
Supplement resolution content carries out supplement resolution, including：
Supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and in the described second semantic analysis result
When mark is comprising the first label, determine that the corresponding target entity of the reference word links according to the conversation history information；
Using supplement resolution content is waited for described in target entity link replacement, to obtain the corresponding final language of the Second Problem
Adopted analysis result.
10. the supplement digestion instrument of semantic analysis result in more wheel sessions according to claim 8, which is characterized in that institute
It states when determining in the described second semantic analysis result exist when supplement clears up content, according to the conversation history information to described
Wait for that supplement resolution content carries out supplement resolution, including：
Supplement resolution content is waited for when determining in the described second semantic analysis result to exist, and in the described second semantic analysis result
When mark is comprising the second label, supplement resolution is waited for using described in all entity links replacement recorded in the conversation history information
Content, to obtain at least one semantic analysis result；
The target semanteme analysis result from being screened at least one semantic analysis result, by the target semanteme analysis result
It is determined as the corresponding final semantic analysis result of the Second Problem.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810219417.5A CN108549662B (en) | 2018-03-16 | 2018-03-16 | Complementary digestion method and device for semantic analysis results in multi-round conversation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810219417.5A CN108549662B (en) | 2018-03-16 | 2018-03-16 | Complementary digestion method and device for semantic analysis results in multi-round conversation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108549662A true CN108549662A (en) | 2018-09-18 |
| CN108549662B CN108549662B (en) | 2021-08-06 |
Family
ID=63516538
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810219417.5A Active CN108549662B (en) | 2018-03-16 | 2018-03-16 | Complementary digestion method and device for semantic analysis results in multi-round conversation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108549662B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109753557A (en) * | 2018-12-26 | 2019-05-14 | 出门问问信息科技有限公司 | Answer output method, device, equipment and the storage medium of question answering system |
| CN109800425A (en) * | 2018-12-25 | 2019-05-24 | 远光软件股份有限公司 | A kind of semantic missing automatic error-correcting processing method and system of cross-platform information exchange |
| CN109830231A (en) * | 2018-12-17 | 2019-05-31 | 北京百度网讯科技有限公司 | Conversation processing method, device and storage medium |
| CN110222162A (en) * | 2019-05-10 | 2019-09-10 | 天津中科智能识别产业技术研究院有限公司 | A kind of intelligent answer method based on natural language processing and knowledge mapping |
| CN110377715A (en) * | 2019-07-23 | 2019-10-25 | 天津汇智星源信息技术有限公司 | Reasoning type accurate intelligent answering method based on legal knowledge map |
| CN110489538A (en) * | 2019-08-27 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Sentence answer method, device and electronic equipment based on artificial intelligence |
| CN111401034A (en) * | 2018-12-28 | 2020-07-10 | 深圳市优必选科技有限公司 | Text semantic analysis method, semantic analysis device and terminal |
| CN112183060A (en) * | 2020-09-28 | 2021-01-05 | 重庆工商大学 | Reference resolution method of multi-round dialogue system |
| CN112199473A (en) * | 2020-10-16 | 2021-01-08 | 上海明略人工智能(集团)有限公司 | Multi-turn dialogue method and device in knowledge question-answering system |
| CN112800189A (en) * | 2019-11-14 | 2021-05-14 | 科沃斯商用机器人有限公司 | Human-computer interaction method and device, intelligent robot and storage medium |
| WO2023087935A1 (en) * | 2021-11-22 | 2023-05-25 | 京东方科技集团股份有限公司 | Coreference resolution method, and training method and apparatus for coreference resolution model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070033221A1 (en) * | 1999-06-15 | 2007-02-08 | Knova Software Inc. | System and method for implementing a knowledge management system |
| CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
| CN105589844A (en) * | 2015-12-18 | 2016-05-18 | 北京中科汇联科技股份有限公司 | Missing semantic supplementing method for multi-round question-answering system |
| US20160203256A1 (en) * | 2010-06-29 | 2016-07-14 | Medeolinx, LLC | Inter-class molecular association connectivity mapping |
| CN107358315A (en) * | 2017-06-26 | 2017-11-17 | 深圳市金立通信设备有限公司 | A kind of information forecasting method and terminal |
| CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
-
2018
- 2018-03-16 CN CN201810219417.5A patent/CN108549662B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070033221A1 (en) * | 1999-06-15 | 2007-02-08 | Knova Software Inc. | System and method for implementing a knowledge management system |
| US20160203256A1 (en) * | 2010-06-29 | 2016-07-14 | Medeolinx, LLC | Inter-class molecular association connectivity mapping |
| CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
| CN105589844A (en) * | 2015-12-18 | 2016-05-18 | 北京中科汇联科技股份有限公司 | Missing semantic supplementing method for multi-round question-answering system |
| CN107358315A (en) * | 2017-06-26 | 2017-11-17 | 深圳市金立通信设备有限公司 | A kind of information forecasting method and terminal |
| CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109830231A (en) * | 2018-12-17 | 2019-05-31 | 北京百度网讯科技有限公司 | Conversation processing method, device and storage medium |
| CN109830231B (en) * | 2018-12-17 | 2021-03-16 | 北京百度网讯科技有限公司 | Session processing method, device and storage medium |
| CN109800425B (en) * | 2018-12-25 | 2023-05-09 | 远光软件股份有限公司 | Semantic miss automatic error correction processing method and system for cross-platform information exchange |
| CN109800425A (en) * | 2018-12-25 | 2019-05-24 | 远光软件股份有限公司 | A kind of semantic missing automatic error-correcting processing method and system of cross-platform information exchange |
| CN109753557A (en) * | 2018-12-26 | 2019-05-14 | 出门问问信息科技有限公司 | Answer output method, device, equipment and the storage medium of question answering system |
| CN111401034A (en) * | 2018-12-28 | 2020-07-10 | 深圳市优必选科技有限公司 | Text semantic analysis method, semantic analysis device and terminal |
| CN111401034B (en) * | 2018-12-28 | 2023-10-10 | 深圳市优必选科技有限公司 | Semantic analysis method, semantic analysis device and terminal for text |
| CN110222162A (en) * | 2019-05-10 | 2019-09-10 | 天津中科智能识别产业技术研究院有限公司 | A kind of intelligent answer method based on natural language processing and knowledge mapping |
| CN110377715A (en) * | 2019-07-23 | 2019-10-25 | 天津汇智星源信息技术有限公司 | Reasoning type accurate intelligent answering method based on legal knowledge map |
| CN110489538A (en) * | 2019-08-27 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Sentence answer method, device and electronic equipment based on artificial intelligence |
| CN110489538B (en) * | 2019-08-27 | 2020-12-25 | 腾讯科技（深圳）有限公司 | Statement response method and device based on artificial intelligence and electronic equipment |
| CN112800189A (en) * | 2019-11-14 | 2021-05-14 | 科沃斯商用机器人有限公司 | Human-computer interaction method and device, intelligent robot and storage medium |
| CN112183060A (en) * | 2020-09-28 | 2021-01-05 | 重庆工商大学 | Reference resolution method of multi-round dialogue system |
| CN112183060B (en) * | 2020-09-28 | 2022-05-10 | 重庆工商大学 | Reference resolution method of multi-round dialogue system |
| CN112199473A (en) * | 2020-10-16 | 2021-01-08 | 上海明略人工智能(集团)有限公司 | Multi-turn dialogue method and device in knowledge question-answering system |
| WO2023087935A1 (en) * | 2021-11-22 | 2023-05-25 | 京东方科技集团股份有限公司 | Coreference resolution method, and training method and apparatus for coreference resolution model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108549662B (en) | 2021-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108549662A (en) | The supplement digestion procedure and device of semantic analysis result in more wheel sessions | |
| Adamopoulou et al. | Chatbots: History, technology, and applications | |
| EP3792732A1 (en) | A system and method for visual analysis of emotional coherence in videos | |
| CN103324604B (en) | For the standardized method and system of the specific natural language in territory | |
| CN106649694A (en) | Method and device for identifying user's intention in voice interaction | |
| CN106992001A (en) | Processing method, the device and system of phonetic order | |
| CN104718569A (en) | Improving phonetic pronunciation | |
| CN108519998B (en) | Problem guiding method and device based on knowledge graph | |
| CN107945802A (en) | Voice recognition result processing method and processing device | |
| CN110444229A (en) | Communication service method, device, computer equipment and storage medium based on speech recognition | |
| CN112199473A (en) | Multi-turn dialogue method and device in knowledge question-answering system | |
| US20150186354A1 (en) | Method for analyzing emotion based on messenger conversation | |
| CN109635305A (en) | Voice translation method and device, equipment and storage medium | |
| CN109659009A (en) | Motion management method, apparatus and electronic equipment | |
| CN114140814A (en) | Emotion recognition capability training method and device and electronic equipment | |
| CN116561265A (en) | Personalized dialogue generation method, model training method and device | |
| CN114969352A (en) | Text processing method, system, storage medium and electronic equipment | |
| CN105373527B (en) | Omission recovery method and question-answering system | |
| Motta et al. | Users’ error recovery strategies in the interaction with voice assistants (VAs) | |
| CN111625629B (en) | Task type dialogue robot response method and device, robot and storage medium | |
| Inupakutika et al. | Integration of NLP and Speech-to-text Applications with Chatbots | |
| CN106847273A (en) | The wake-up selected ci poem selection method and device of speech recognition | |
| US10559298B2 (en) | Discussion model generation system and method | |
| CN108854072A (en) | A kind of voice prompt method and device | |
| CN106782516A (en) | Language material sorting technique and device |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 101, 1st floor, building 1, Xisanqi building materials City, Haidian District, Beijing 100096 Applicant after: Yunzhisheng Intelligent Technology Co.,Ltd. Address before: 12 / F, Guanjie building, building 1, No. 16, Taiyanggong Middle Road, Chaoyang District, Beijing Applicant before: BEIJING UNISOUND INFORMATION TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |