CN105653619B - The update method and device in correct log library in intelligent Answer System - Google Patents
The update method and device in correct log library in intelligent Answer System Download PDFInfo
- Publication number
- CN105653619B CN105653619B CN201510993059.XA CN201510993059A CN105653619B CN 105653619 B CN105653619 B CN 105653619B CN 201510993059 A CN201510993059 A CN 201510993059A CN 105653619 B CN105653619 B CN 105653619B
- Authority
- CN
- China
- Prior art keywords
- abstract semantics
- semantic
- customer problem
- expression formula
- word
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides the update methods and device in log library correct in a kind of intelligent Answer System.The intelligent Answer System includes abstract semantics database, the abstract semantics database includes the abstract semantics set of multiple classifications, the abstract semantics set of each classification includes multiple abstract semantics expression formulas, the correct log library includes customer problem and answer, the update method includes: to carry out abstract semantics recommendation process to the customer problem, and obtain the corresponding classification of abstract semantics expression formula recommended to;The abstract semantics set of acquired classification is extracted, to carry out expansion processing to the customer problem with extracted abstract semantics set, obtains one or more specific semantic formulas;The specific semantic formula is added in the correct log library.The present invention can greatly increase the amount in correct log library, and the log sum of manual examination and verification is needed when can be reduced second of iteration, i.e., correct log library increases the coverage rate of user journal.
Description
Technical field
The present invention relates to the update methods in log library correct in human-computer interaction technique field more particularly to intelligent Answer System
And device.
Background technique
Human-computer interaction is the science of the interactive relation between research system and user.System can be various machines
Device is also possible to the system and software of computerization.For example, various artificial intelligence systems, example may be implemented by human-computer interaction
Such as, intelligent customer service system, speech control system etc..Artificial intelligence semantics recognition is the basis of human-computer interaction, can be to people
Speech like sound identified, to be converted into machine it will be appreciated that language.
Intelligent Answer System is a kind of typical case of human-computer interaction, wherein after user's proposition problem, intelligent answer system
System provides the answer of the problem.For this purpose, there is a set of knowledge base in intelligent Answer System, there is a large amount of problems in the inside and asks with each
Inscribe corresponding answer.The problem of intelligent Answer System is proposed firstly the need of identification user is found and is somebody's turn to do from knowledge base
Then problem corresponding to customer problem finds out the answer to match with the problem.
The maintenance update of intelligent Answer System is a significant challenge.
Summary of the invention
A brief summary of one or more aspects is given below to provide to the basic comprehension in terms of these.This general introduction is not
The extensive overview of all aspects contemplated, and be both not intended to identify critical or decisive element in all aspects also non-
Attempt to define the range in terms of any or all.Its unique purpose is to provide the one of one or more aspects in simplified form
A little concepts are with the sequence for more detailed description given later.
According to an aspect of the present invention, a kind of update method in correct log library in intelligent Answer System, the intelligence are provided
Energy question answering system includes abstract semantics database, which includes the abstract semantics set of multiple classifications, each
The abstract semantics set of classification includes multiple abstract semantics expression formulas, which includes customer problem and answer, this is more
New method includes:
Abstract semantics recommendation process is carried out to the customer problem, and obtains the corresponding class of abstract semantics expression formula recommended to
Not;
The abstract semantics set of acquired classification is extracted, to be carried out with extracted abstract semantics set to the customer problem
Expansion processing obtains one or more specific semantic formulas;
The specific semantic formula is added in the correct log library.
In one example, which includes missing semantic component;The expansion is handled
The missing language with each abstract semantics expression formula of extracted abstract semantics set is extracted from the customer problem
The corresponding content of adopted ingredient, and by the fills of extraction into the corresponding missing semantic component of each abstract semantics expression formula with
Obtain specific semantic formula corresponding with the customer problem.
In one example, which includes:
Word segmentation processing is carried out to the customer problem, obtains several words, which is semantic rules word or non-semantic rule
Word;
Part-of-speech tagging processing is carried out to each non-semantic regular word respectively, obtains the part of speech letter of each non-semantic regular word
Breath;
Part of speech judgement processing is carried out to each semantic rules word respectively, obtains the grammatical category information of each semantic rules word;
Abstract semantics database is scanned for handling according to the part-of-speech information and grammatical category information, is obtained and the customer problem
Matched abstract semantics expression formula.
In one example, which further includes semantic rules word, with the matched abstract semantics table of the user
Meet the following conditions up to formula:
The corresponding part of speech of missing semantic component of abstract semantics expression formula includes the word of the corresponding filling content of customer problem
Property;
Abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
The sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
In one example, this method further include:
The data information of artificial maintenance knowledge point is received, and place is updated to correct log library according to the data information
Reason.
According to another aspect of the present invention, a kind of log analysis method of intelligent Answer System is provided, comprising:
Obtain user journal data;
At least from being filtered out in the user journal data and the log that content matches in correct log library, the correct log library
It is the correct log library updated by preceding method.
According to another aspect of the present invention, a kind of updating device in correct log library in intelligent Answer System is provided, it should
Intelligent Answer System includes abstract semantics database, which includes the abstract semantics set of multiple classifications, often
The abstract semantics set of a classification includes multiple abstract semantics expression formulas, which includes customer problem and answer, is somebody's turn to do
Updating device includes:
Abstract semantics recommending module for carrying out abstract semantics recommendation process to the customer problem, and is obtained and is recommended to
The corresponding classification of abstract semantics expression formula;
Enlargement module, for extracting the abstract semantics set of acquired classification, with extracted abstract semantics set pair
The customer problem carries out expansion processing, obtains one or more specific semantic formulas;And
Adding module, for the specific semantic formula to be added in the correct log library.
In one example, which includes missing semantic component;The enlargement module includes:
Extraction module, for extracting each abstract semantics table with extracted abstract semantics set from the customer problem
Up to the corresponding content of missing semantic component of formula;And
Module is filled, the fills for that will extract are into the corresponding missing semantic component of each abstract semantics expression formula
To obtain specific semantic formula corresponding with the customer problem.
In one example, which includes:
Word segmentation module obtains several words, which is semantic rules word for carrying out word segmentation processing to the customer problem
Or non-semantic regular word;
Part-of-speech tagging module obtains each non-language for carrying out part-of-speech tagging processing to each non-semantic regular word respectively
The part-of-speech information of adopted rule word;
Part of speech judgment module obtains each semantic rule for carrying out part of speech judgement processing to each semantic rules word respectively
The then grammatical category information of word;
Search module is obtained for scanning for handling to abstract semantics database according to the part-of-speech information and grammatical category information
To with the matched abstract semantics expression formula of the customer problem.
In one example, which further includes semantic rules word, with the matched abstract semantics table of the user
Meet the following conditions up to formula:
The corresponding part of speech of missing semantic component of abstract semantics expression formula includes the word of the corresponding filling content of customer problem
Property;
Abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
The sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
In one example, the device further include:
Maintenance module, for receiving the data information of artificial maintenance knowledge point, and according to the data information to correct log
Library is updated processing.
According to another aspect of the present invention, a kind of log analysis device of intelligent Answer System is provided, comprising:
Module is obtained, for obtaining user journal data;And
Filtering module, at least from the log filtered out in the daily record data with content matches in correct log library, being somebody's turn to do
Correct log library is the correct log library updated by aforementioned device.
Compared with prior art, the beneficial effect comprise that
The present invention can automatically be updated log library correct in intelligent Answer System, to greatly increase correct day
The amount in will library needs the log sum of manual examination and verification, i.e., covering of the correct log library to user journal when reducing by second of iteration
Rate increases.
Detailed description of the invention
After the detailed description for reading embodiment of the disclosure in conjunction with the following drawings, it better understood when of the invention
Features described above and advantage.In the accompanying drawings, each component is not necessarily drawn to scale, and has similar correlation properties or feature
Component may have same or similar appended drawing reference.
Fig. 1 is to show the stream of the update method in correct log library in intelligent Answer System according to an aspect of the present invention
Cheng Tu;And
Fig. 2 is to show the frame of the updating device in correct log library in intelligent Answer System according to an aspect of the present invention
Figure.
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.Note that below in conjunction with attached drawing and specifically real
The aspects for applying example description is merely exemplary, and is understood not to carry out any restrictions to protection scope of the present invention.
Basic knowledge point most original and simplest form in knowledge base are exactly usually common FAQ, general form
It is that " ask-answer " is right.In the present invention, " standard is asked " is used to indicate that the text of some knowledge point, and main target is that expression is clear,
It is convenient for safeguarding.For example, " rate of CRBT " are exactly that clearly standard asks description for expression.Here " asking " should not narrowly be understood
For " inquiry ", and should broadly understand one " input ", should " input " with corresponding " output ".For example, for for controlling
For the semantics recognition of system, the instruction of user, such as " opening radio " should also be understood to be one " asking ",
Corresponding at this time " answering " can be the calling for executing the control program accordingly controlled.
User to machine when inputting, the most ideal situation is that asked using standard, then the intelligent semantic identifying system of machine
At once it will be appreciated that the meaning of user.However, user often not uses standard to ask, but some deformations for asking of standard
Form.For example, if being " changing a radio station " for the standard form of asking of the radio station switching of radio, then what user may use
Order is " switching one radio station ", and what machine was also required to can to identify user's expression is the same meaning.
Therefore, for intelligent semantic identification, the extension that the standard that needs in knowledge base is asked is asked, which asks and standard
It asks that expression-form has slight difference, but expresses identical meaning.
Further, in order to it is more acurrate, efficiently identify customer problem, abstract semantics are also been developed in intelligent Answer System
Concept.Abstract semantics are being further abstracted to ontology generic attribute.The abstract semantics of one classification pass through one group of abstract semantics table
The different expression of a kind of abstract semantics are described up to the set of formula, to express more abstract semanteme, the expression of these abstract semantics
Formula is expanded on component.When the element that these expand once have been assigned corresponding value can express it is various respectively
The specific semanteme of sample.
Each abstract semantics expression formula mainly may include missing semantic component and semantic rules word.Lack semantic component by
Semantic component symbol indicates, can express after the semantic component of these missings is filled with corresponding value (i.e. content) all kinds of
Specific semanteme.
The semantic component of abstract semantics accords with can include:
[concept]: the word or phrase of main body or object composition are indicated.
Such as: " CRBT " in " how open-minded CRBT is "
[action]: the word or phrase of expression movement ingredient.
Such as: " handling " in " how credit card is handled "
[attribute]: the word or phrase of attribute composition are indicated.
Such as: " color " in " which color iphone has "
[adjective]: the word or phrase of ornamental equivalent are indicated.
Such as: " cheap " in " which brand of refrigerator is cheap "
Some main abstract semantics classification examples have:
What conceptual illustration [concept] is
Attribute constitutes which [attribute] [concept] has
Behavior [concept] how [action]
Behavior place [concept] is somewhere [action]
Why behavioral reasons [concept] can [action]
Behavior prediction [concept] can or can not [action]
Behavior judges [concept] either with or without [attribute]
[attribute] of attribute situation [concept] is [adjective]
Whether determined property [concept] has [attribute]
Why so [adjective] [attribute] of attribute reason [concept]
Where are the difference of proximate nutrition [concept1] and [concept2]
[attribute] that attribute compares [concept1] and [concept2] has any difference
Question sentence can do general judge in the composition judgement of abstract semantics level by part-of-speech tagging, and concept pairs
The part of speech answered is noun, and the corresponding part of speech of action is verb, the corresponding part of speech of attribute is noun, adjective correspondence
Be adjective.
By the abstract semantics [concept] that classification is " behavior " how for [action], the abstract language of the category
May include a plurality of abstract semantics expression formula under justice set:
Abstract semantics classification: behavior
Abstract semantics expression formula:
A. [concept] [need | should? ] [how]<[can with]?><carry out?>[action]
B. { [concept]~[action] }
C. [concept]<?>[action]<method | mode | step?>
D.<which has | what has | either with or without><pass through | use |>[concept] [action]<?>[method]
E. [how] [action]~[concept]
Tetra- abstract semantics expression formulas of above-mentioned a, b, c, d are all for describing " behavior " this abstract semantics classification
's.Symbol " | " expression "or" relationship, symbol "? " indicate that the ingredient is not essential.It, can by taking above-mentioned abstract semantics expression formula c as an example
Expand into abstract semantics expression below:
C1. [concept]<>[action]<method>
C2. [concept]<>[action]<mode>
C3. [concept]<>[action]<step>
C4. [concept]<>[action]
C5. [concept] [action]<method>
C6. [concept] [action]<mode>
C7. [concept] [action]<step>
c8.[concept][action]
In above-mentioned abstract semantics expression formula, other than the abstract semantic component symbol as missing semantic component,
The specific word that he occurs such as " how ", " should ", " method ", these words need to be used in abstract semantics are regular,
So can be collectively referred to as semantic rules word.
The basic concepts in intelligent Answer System are described above.
Intelligent Answer System can generate a large amount of user journal in use, and each log includes the user that user provides
Problem and the answer provided for the customer problem by intelligent Answer System.Manually need the sea generated to intelligent Answer System
The user journal of amount is analyzed to identify, the answer correctness provided with the system of checking, for intelligent Answer System
Optimization and maintenance.
Correct log library is in intelligent Answer System for storing the database of correct log.In the mass users to generation
When log is analyzed, if the content of a log matches with a log in correct log library, i.e. this log
Problem in correct log library the problem of this log it is identical, then the answer for defaulting the offer of intelligent Answer System is necessarily correct
, it is therefore not necessary to manual analysis again.
So correct log library can be used for before the mass users log caused by manual analysis institute intelligent Answer System,
First these massive logs are filtered, can directly be recognized with those of correct log matches in correct log library user journal
Be it is correct, no longer need to be analyzed to identify, and only to other user journals carry out manual analysis confirmation, this greatly reduces artificial
Workload.
As it can be seen that the amount for the correct log that correct log library is covered is the bigger the better, initial correct log library only includes one
How a little problems expand correct log library so that it includes that the more contents of covering are most important.
In the present invention, inventor recognizes if knowledge point reference in correct log library is abstract semantics
Words, then the problem of all abstract semantics Rule Extendeds come out all will be that correctly, can enter correct log library, this amount is non-
Normal is big, and expand come problem be all it is clear and coherent, it is significant.This will greatly increase the amount in correct log library, energy
The log sum of manual examination and verification is needed when reducing by second of iteration, i.e., correct log library increases the coverage rate of user journal.
Fig. 1 shows the update method 100 in correct log library in intelligent Answer System according to an aspect of the present invention
Flow chart.
The intelligent Answer System includes abstract semantics database, and the abstract semantics database includes the abstract of multiple classifications
Semanteme set, such as above-mentioned " conceptual illustration ", " attribute composition ", " behavior ".The abstract semantics set of each classification
Including multiple abstract semantics expression formulas.Still by taking above-mentioned " behavior " this kind of other abstract semantics set as an example, including it is abstract
Semantic formula a, b, c, d, e.
It include customer problem and answer in correct log library.For specific one customer problem, using abstract semantics
The problem of abstract semantics in database are extended to the customer problem, then obtain extension is added to correct log library again
To expand the amount in correct log library.
In step 102, abstract semantics recommendation process can be carried out to the customer problem, and obtains the abstract semantics table recommended to
Up to the corresponding classification of formula.
For example, a customer problem in correct log library are as follows: " how looking into violating the regulations ".
In this step, need to find abstract semantics classification corresponding with the customer problem in abstract semantics database.
For " how looking into violating the regulations " this customer problem, corresponding classification is the abstract semantics set of " behavior ".
In one example, which recommends to carry out word segmentation processing to the customer problem first, obtains several words, should
Word is semantic rules word or non-semantic regular word.
For example, " how looking into violating the regulations " can be divided into word " how ", " looking into ", " violating the regulations ".In these words, " how " it is semanteme
Regular word, " looking into " and " violating the regulations " are non-semantic rules words.
Then, part-of-speech tagging processing is carried out to each non-semantic regular word respectively, such as " looking into " is noted as verb, " disobeys
Chapter " is noted as noun.
Later, part of speech judgement processing is carried out to each semantic rules word, obtains the grammatical category information of each semantic rules word.Word
Class simply understands to be one group of word for having general character, these words can be similar or dissimilar semantically.
Finally, scanning for handling to abstract semantics database according to these part-of-speech informations and grammatical category information, obtains and use
Family problem " how looking into violating the regulations " matched abstract semantics expression formula.
In practice, meet the following conditions with the matched abstract semantics expression formula of user:
1) the corresponding part of speech of missing semantic component of abstract semantics expression formula includes the corresponding filling content of customer problem
Part of speech;
2) abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
3) sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
In above-mentioned abstract semantics classification " behavior ", the missing semantic component action's of abstract semantics expression formula e
Part of speech is verb, and the corresponding filling content " looking into " of customer problem " how looking into violating the regulations " is also verb, lacks semantic component concept
Part of speech be noun, the corresponding filling content " violating the regulations " of customer problem " how looking into violating the regulations " is also noun, therefore meets above-mentioned item
Part 1).
Secondly, in abstract semantics expression formula e semantic rules word " how " it is corresponding with customer problem " how looking into violating the regulations "
Semantic rules word " how " belong to same part of speech, therefore meet above-mentioned condition 2).
Finally, the sequence of abstract semantics expression formula e is also identical as the order of representation of customer problem, meet above-mentioned condition 3).
Therefore, it in abstract semantics database, finds and is expressed with customer problem " how looking into violating the regulations " matched abstract semantics
Formula e, i.e., [how] [action]~[concept].Since the abstract semantics expression formula belongs to " behavior " classification, it is
It recommends " behavior " this kind of other abstract semantics set.
It should be noted that matched abstract semantics expression formula can be not necessarily found, so some problems are looked for not
To the abstract semantics of recommendation, at this point, the customer problem can not be extended using method of the invention.
In step 104, extract the abstract semantics set of acquired classification, with extracted abstract semantics set to described
Customer problem carries out expansion processing, obtains one or more specific semantic formulas.
By taking above-mentioned customer problem " how looking into violating the regulations " as an example, acquired " behavior " this kind of other abstract languages are extracted
Justice set, obtains abstract semantics expression formula:
A. [concept] [need | should? ] [how]<[can with]?><carry out?>[action]
B. { [concept]~[action] }
C. [concept]<?>[action]<method | mode | step?>
D.<which has | what has | either with or without><pass through | use |>[concept] [action]<?>[method]
Expansion processing is carried out to customer problem " how looking into violating the regulations " with above-mentioned abstract semantics expression formula.
In one example, each abstract semantics expression formula with extracted abstract semantics set is extracted from customer problem
The corresponding content of missing semantic component, and it is the fills of extraction are semantic to the corresponding missing of each abstract semantics expression formula
To obtain specific semantic formula corresponding with the customer problem in ingredient.
With abstract semantics expression formula a:[concept] [need | should? ] [how]<[can with]?><carry out?>
For [action], from " how ", " looking into ", content corresponding with the missing semantic component of the expression formula is extracted in " violating the regulations ":
The corresponding content of concept: " violating the regulations "
The corresponding content of action: " looking into "
Therefore, it " will look into " and filling violating the regulations " violating the regulations " to corresponding missing semantic component obtains a specific semantic formula:
[violating the regulations] [need | should? ] [how]<[can with]?><carry out?>[inquiry].
By taking abstract semantics expression formula b. { [concept]~[action] } as an example, from " how ", " looking into ", mention in " violating the regulations "
Take content corresponding with the missing semantic component of the expression formula:
The corresponding content of concept: " violating the regulations "
The corresponding content of action: " looking into "
Therefore, it " will look into " and filling violating the regulations " violating the regulations " to corresponding missing semantic component obtains a specific semantic formula:
[violating the regulations] [inquiry].
With abstract semantics expression formula c. [concept]<?>[action]<method | mode | step?>for, from " why
", " looking into ", content corresponding with the missing semantic component of the expression formula is extracted in " violating the regulations ":
The corresponding content of concept: " violating the regulations "
The corresponding content of action: " looking into "
Therefore, it " will look into " and filling violating the regulations " violating the regulations " to corresponding missing semantic component obtains a specific semantic formula:
[violating the regulations]<?>[inquiry]<method | mode | step?>.
With abstract semantics expression formula d.<which has | what has | either with or without><pass through | use |>[concept] [action]
<? for>[method], from " how ", " looking into ", content corresponding with the missing semantic component of the expression formula is extracted in " violating the regulations ":
The corresponding content of concept: " violating the regulations "
The corresponding content of action: " looking into "
Therefore, it " will look into " and filling violating the regulations " violating the regulations " to corresponding missing semantic component obtains a specific semantic formula: <
Which has | what has | either with or without><pass through | use |>[violating the regulations] [inquiry]<?>[method].
In step 106, these obtained specific semantic formulas are added in correct log library.
Every words after these specific semantic formulas expansion be all it is clear and coherent, thus obtain dilatation in correct log library.
By carrying out similar procedure to all customer problems in correct log library, it can greatly expand correct log library
Amount.
Further, it is also possible to receive the data information of artificial maintenance knowledge point, and according to these data informations to correct log
Library is updated processing.It should be noted that update processing at this time can be increase, deletion or replacement.
Fig. 2 shows the updating devices 200 in log library correct in intelligent Answer System according to an aspect of the present invention
Block diagram.
As shown in Fig. 2, the updating device 200 may include abstract semantics recommending module 210, enlargement module 220 and addition
Module 230.
Abstract semantics recommending module 210 can carry out abstract semantics recommendation process to the customer problem in correct log library, and
Obtain the corresponding classification of abstract semantics expression formula recommended to.
In one example, abstract semantics recommending module 210 may include word segmentation module 211, with for the customer problem into
Row word segmentation processing, obtains several words, and the word is semantic rules word or non-semantic regular word.Abstract semantics recommending module
210 may also include part-of-speech tagging module 212, to obtain for carrying out part-of-speech tagging processing to each non-semantic regular word respectively
The part-of-speech information and part of speech judgment module 213 of each non-semantic regular word, for carrying out word to each semantic rules word respectively
Class judgement processing, obtains the grammatical category information of each semantic rules word.Finally, abstract semantics recommending module 210 can also include searching
Rope module, to obtain and be somebody's turn to do for scanning for handling to abstract semantics database according to these part-of-speech informations and grammatical category information
The matched abstract semantics expression formula of customer problem.
In practice, meet the following conditions with the matched abstract semantics expression formula of user:
The corresponding part of speech of missing semantic component of abstract semantics expression formula includes the word of the corresponding filling content of customer problem
Property;
Abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
The sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
Enlargement module 220 can be used for extracting the abstract semantics set of acquired classification, with extracted abstract semantics collection
It closes and expansion processing is carried out to the customer problem, obtain one or more specific semantic formulas.
In one example, enlargement module 220 may include extraction module 221 for extracting and being extracted from customer problem
Abstract semantics set each abstract semantics expression formula the corresponding content of missing semantic component.Enlargement module 220 can also wrap
Include filling module with the fills for that will extract into the corresponding missing semantic component of each abstract semantics expression formula with must
To specific semantic formula corresponding with customer problem.
Adding module 230 is responsible for obtained specific semantic formula being added in correct log library.
In some instances, updating device 200 may also include maintenance module with the number for receiving artificial maintenance knowledge point
It is believed that breath, and processing is updated to correct log library according to these data informations.It should be noted that update processing at this time
It can be increase, deletion or replacement.
According to another aspect of the present invention, the log analysis method and device for intelligent Answer System are additionally provided.
Above-mentioned updated correct log library is all utilized in the log analysis method and device.For example, log analysis method
It can be after obtaining user journal data, at least from the day filtered out in user journal data with content matches in correct log library
Will, the correct log library are the correct log libraries updated by the above method.
Before the mass users log caused by manual analysis institute intelligent Answer System, first to these massive logs
It is filtered, is fair to consider that correctly with those of correct log matches in correct log library user journal, no longer needed to point
Analysis confirmation, and manual analysis confirmation only is carried out to other user journals, this greatly reduces labor workloads.
Correspondingly, log analysis device may include filtering module, with obtain module obtain user journal data after, until
Less from the log filtered out in user journal data with content matches in correct log library, which is by above-mentioned side
The correct log library that method updates.
Although for simplify explain the above method is illustrated to and is described as a series of actions, it should be understood that and understand,
The order that these methods are not acted is limited, because according to one or more embodiments, some movements can occur in different order
And/or with from it is depicted and described herein or herein it is not shown and describe but it will be appreciated by those skilled in the art that other
Movement concomitantly occurs.
Those skilled in the art will further appreciate that, the various illustratives described in conjunction with the embodiments described herein
Logic plate, module, circuit and algorithm steps can be realized as electronic hardware, computer software or combination of the two.It is clear
Explain to Chu this interchangeability of hardware and software, various illustrative components, frame, module, circuit and step be above with
Its functional form makees generalization description.Such functionality be implemented as hardware or software depend on concrete application and
It is applied to the design constraint of total system.Technical staff can realize every kind of specific application described with different modes
Functionality, but such realization decision should not be interpreted to cause departing from the scope of the present invention.
Software should be broadly interpreted to mean instruction, instruction set, code, code segment, program code, program, son
Program, software module, application, software application, software package, routine, subroutine, object, executable item, the thread of execution, regulation,
Function etc., no matter it is all is to address with software, firmware, middleware, microcode, hardware description language or other terms
So.
General place can be used in conjunction with various illustrative logic plates, module and the circuit that presently disclosed embodiment describes
Reason device, digital signal processor (DSP), specific integrated circuit (ASIC), field programmable gate array (FPGA) other are compiled
Journey logical device, discrete door or transistor logic, discrete hardware component or its be designed to carry out function described herein
Any combination is realized or is executed.General processor can be microprocessor, but in alternative, which, which can be, appoints
What conventional processor, controller, microcontroller or state machine.Processor is also implemented as calculating the combination of equipment, example
As DSP and the combination of microprocessor, multi-microprocessor, the one or more microprocessors to cooperate with DSP core or it is any its
His such configuration.
The step of method or algorithm for describing in conjunction with embodiment disclosed herein, can be embodied directly in hardware, in by processor
It is embodied in the software module of execution or in combination of the two.Software module can reside in RAM memory, flash memory, ROM and deposit
Reservoir, eprom memory, eeprom memory, register, hard disk, removable disk, CD-ROM or known in the art appoint
In the storage medium of what other forms.Exemplary storage medium is coupled to processor so that the processor can be from/to the storage
Medium reads and writees information.In alternative, storage medium can be integrated into processor.
Offer is to make any person skilled in the art all and can make or use this public affairs to the previous description of the disclosure
It opens.The various modifications of the disclosure all will be apparent for a person skilled in the art, and as defined herein general
Suitable principle can be applied to other variants without departing from the spirit or scope of the disclosure.The disclosure is not intended to be limited as a result,
Due to example described herein and design, but should be awarded and principle disclosed herein and novel features phase one
The widest scope of cause.
Claims (10)
1. the update method in correct log library in a kind of intelligent Answer System, which is characterized in that the intelligent Answer System includes
Abstract semantics database, the abstract semantics database include the abstract semantics set of multiple classifications, the abstract language of each classification
Justice set includes multiple abstract semantics expression formulas, and the correct log library includes customer problem and answer, the update method packet
It includes:
Abstract semantics recommendation process is carried out to the customer problem, and obtains the corresponding class of abstract semantics expression formula recommended to
Not;
The abstract semantics set of acquired classification is extracted, to be expanded with extracted abstract semantics set the customer problem
Processing is filled, one or more specific semantic formulas are obtained;
The specific semantic formula is added in the correct log library;The abstract semantics expression formula includes that missing is semantic
Ingredient;The expansion is handled
It is extracted from the customer problem semantic with the missing of each abstract semantics expression formula of extracted abstract semantics set
The corresponding content of ingredient, and by the fills of extraction into the corresponding missing semantic component of each abstract semantics expression formula with
To specific semantic formula corresponding with the customer problem.
2. update method as described in claim 1, which is characterized in that the abstract semantics recommendation process includes:
Word segmentation processing is carried out to the customer problem, obtains several words, the word is semantic rules word or non-semantic rule
Word;
Part-of-speech tagging processing is carried out to each non-semantic regular word respectively, obtains the part-of-speech information of each non-semantic regular word;
Part of speech judgement processing is carried out to each semantic rules word respectively, obtains the grammatical category information of each semantic rules word;
Abstract semantics database is scanned for handling according to the part-of-speech information and grammatical category information, is obtained and the customer problem
Matched abstract semantics expression formula.
3. update method as claimed in claim 2, which is characterized in that the abstract semantics expression formula further includes semantic rules
Word meets the following conditions with the matched abstract semantics expression formula of the user:
The corresponding part of speech of missing semantic component of abstract semantics expression formula includes the part of speech of the corresponding filling content of customer problem;
Abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
The sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
4. update method as described in claim 1, which is characterized in that further include:
The data information of artificial maintenance knowledge point is received, and processing is updated to correct log library according to the data information.
5. a kind of log analysis method of intelligent Answer System characterized by comprising
Obtain user journal data;
At least from being filtered out in the user journal data and the log that content matches in correct log library, the correct log library
It is the correct log library updated by method according to any one of claims 1 to 4.
6. the updating device in correct log library in a kind of intelligent Answer System, which is characterized in that the intelligent Answer System includes
Abstract semantics database, the abstract semantics database include the abstract semantics set of multiple classifications, the abstract language of each classification
Justice set includes multiple abstract semantics expression formulas, and the correct log library includes customer problem and answer, the updating device packet
It includes:
Abstract semantics recommending module for carrying out abstract semantics recommendation process to the customer problem, and obtains the pumping recommended to
As the corresponding classification of semantic formula;
Enlargement module, for extracting the abstract semantics set of acquired classification, with extracted abstract semantics set to described
Customer problem carries out expansion processing, obtains one or more specific semantic formulas;And
Adding module, for the specific semantic formula to be added in the correct log library;The abstract semantics expression
Formula includes missing semantic component;The enlargement module includes:
Extraction module is expressed for extracting from the customer problem with each abstract semantics of extracted abstract semantics set
The corresponding content of missing semantic component of formula;And
Module is filled, the fills for that will extract are into the corresponding missing semantic component of each abstract semantics expression formula to obtain
To specific semantic formula corresponding with the customer problem.
7. updating device as claimed in claim 6, which is characterized in that the abstract semantics recommending module includes:
Word segmentation module obtains several words, the word is semantic rules word for carrying out word segmentation processing to the customer problem
Or non-semantic regular word;
Part-of-speech tagging module obtains each non-semantic rule for carrying out part-of-speech tagging processing to each non-semantic regular word respectively
The then part-of-speech information of word;
Part of speech judgment module obtains each semantic rules word for carrying out part of speech judgement processing to each semantic rules word respectively
Grammatical category information;
Search module is obtained for scanning for handling to abstract semantics database according to the part-of-speech information and grammatical category information
With the matched abstract semantics expression formula of the customer problem.
8. updating device as claimed in claim 7, which is characterized in that the abstract semantics expression formula further includes semantic rules
Word meets the following conditions with the matched abstract semantics expression formula of the user:
The corresponding part of speech of missing semantic component of abstract semantics expression formula includes the part of speech of the corresponding filling content of customer problem;
Abstract semantics expression formula is identical with semantic rules word corresponding in customer problem or belongs to same part of speech;
The sequence of abstract semantics expression formula and the order of representation of customer problem are identical.
9. updating device as claimed in claim 6, which is characterized in that further include:
Maintenance module, for receiving the data information of artificial maintenance knowledge point, and according to the data information to correct log library
It is updated processing.
10. a kind of log analysis device of intelligent Answer System, comprising:
Module is obtained, for obtaining user journal data;And
Filtering module, it is described at least from the log filtered out in the daily record data with content matches in correct log library
Correct log library is the correct log library updated by the device as described in any one of claim 6 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510993059.XA CN105653619B (en) | 2015-12-25 | 2015-12-25 | The update method and device in correct log library in intelligent Answer System |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510993059.XA CN105653619B (en) | 2015-12-25 | 2015-12-25 | The update method and device in correct log library in intelligent Answer System |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105653619A CN105653619A (en) | 2016-06-08 |
CN105653619B true CN105653619B (en) | 2019-01-25 |
Family
ID=56476784
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510993059.XA Active CN105653619B (en) | 2015-12-25 | 2015-12-25 | The update method and device in correct log library in intelligent Answer System |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105653619B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106326452A (en) * | 2016-08-26 | 2017-01-11 | 宁波薄言信息技术有限公司 | Method for human-machine dialogue based on contexts |
CN106485328B (en) * | 2016-10-31 | 2020-06-19 | 上海智臻智能网络科技股份有限公司 | Information processing system and method |
CN111382240B (en) * | 2018-12-27 | 2024-03-12 | 上海智臻智能网络科技股份有限公司 | Semantic reasoning method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006119697A (en) * | 2004-10-19 | 2006-05-11 | Fuji Xerox Co Ltd | Question answering system, question answering method, and question answering program |
CN102880645A (en) * | 2012-08-24 | 2013-01-16 | 上海云叟网络科技有限公司 | Semantic intelligent search method |
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104657346A (en) * | 2015-01-15 | 2015-05-27 | 深圳市前海安测信息技术有限公司 | Question matching system and question matching system in intelligent interaction system |
-
2015
- 2015-12-25 CN CN201510993059.XA patent/CN105653619B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006119697A (en) * | 2004-10-19 | 2006-05-11 | Fuji Xerox Co Ltd | Question answering system, question answering method, and question answering program |
CN102880645A (en) * | 2012-08-24 | 2013-01-16 | 上海云叟网络科技有限公司 | Semantic intelligent search method |
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104657346A (en) * | 2015-01-15 | 2015-05-27 | 深圳市前海安测信息技术有限公司 | Question matching system and question matching system in intelligent interaction system |
Also Published As
Publication number | Publication date |
---|---|
CN105653619A (en) | 2016-06-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20190005029A1 (en) | Systems and methods for natural language processing of structured documents | |
CN108875059B (en) | Method and device for generating document tag, electronic equipment and storage medium | |
EP2664997A2 (en) | System and method for resolving named entity coreference | |
CN106649742A (en) | Database maintenance method and device | |
KR20190133931A (en) | Method to response based on sentence paraphrase recognition for a dialog system | |
CN106601237A (en) | Interactive voice response system and voice recognition method thereof | |
CN109635288A (en) | A kind of resume abstracting method based on deep neural network | |
KR20080092337A (en) | Semantic processor for recognition of cause-effect relations in natural language documents | |
WO2021100902A1 (en) | Dialog system answering method based on sentence paraphrase recognition | |
CN104503998A (en) | Type identifying method and device aiming at query sentence of user | |
MXPA04011788A (en) | Learning and using generalized string patterns for information extraction. | |
CN110096599B (en) | Knowledge graph generation method and device | |
KR20200105057A (en) | Apparatus and method for extracting inquiry features for alalysis of inquery sentence | |
CN112380848B (en) | Text generation method, device, equipment and storage medium | |
CN105653619B (en) | The update method and device in correct log library in intelligent Answer System | |
CN105550360B (en) | Optimize the method and device in abstract semantics library | |
CN105677637A (en) | Method and device for updating abstract semantics database in intelligent question-answering system | |
CN109992651B (en) | Automatic identification and extraction method for problem target features | |
CN109800430B (en) | Semantic understanding method and system | |
CN116702736A (en) | Safe call generation method and device, electronic equipment and storage medium | |
CN110162615A (en) | A kind of intelligent answer method, apparatus, electronic equipment and storage medium | |
CN111625623B (en) | Text theme extraction method, text theme extraction device, computer equipment, medium and program product | |
CN111680493B (en) | English text analysis method and device, readable storage medium and computer equipment | |
CN114298048A (en) | Named entity identification method and device | |
Akhtar et al. | Unsupervised morphological expansion of small datasets for improving word embeddings |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Updating method and device of correct log base in Intelligent Question Answering System Effective date of registration: 20220211 Granted publication date: 20190125 Pledgee: Bank of Shanghai Limited by Share Ltd. Pudong branch Pledgor: SHANGHAI XIAOI ROBOT TECHNOLOGY Co.,Ltd. Registration number: Y2022310000021 |