CN105550360A - Method and apparatus for optimizing abstract semantic library - Google Patents
Method and apparatus for optimizing abstract semantic library Download PDFInfo
- Publication number
- CN105550360A CN105550360A CN201511030026.1A CN201511030026A CN105550360A CN 105550360 A CN105550360 A CN 105550360A CN 201511030026 A CN201511030026 A CN 201511030026A CN 105550360 A CN105550360 A CN 105550360A
- Authority
- CN
- China
- Prior art keywords
- abstract semantics
- semantic
- word
- expression formula
- speech
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/258—Heading extraction; Automatic titling; Numbering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a method for optimizing an abstract semantic library by a correct log library based on an intelligent ask-answer system. The correct log library comprises standard asks and similar asks related to all standard asks. The abstract semantic library comprises multiple types of abstract semantic sets. Each type of abstract semantic set comprises multiple abstract semantic expressions. The method comprises the steps of performing abstract semantic recommendation on the standard asks; performing abstract semantic recommendation on the similar asks of the standard asks; responding to failure of the abstract semantic recommendation of the similar asks of the standard asks, and automatically generating corresponding abstract semantic expressions according to the similar asks; and adding new abstract semantic expressions to the type of abstract semantic set corresponding to the abstract semantic expression to which the standard asks are recommended.
Description
Technical field
The present invention relates to human-computer interaction technique field, particularly relate to the method and apparatus in the optimization abstract semantics storehouse, correct daily record storehouse based on Intelligent Answer System.
Background technology
Man-machine interaction is the science of the interactive relation between Study system and user.System can be various machine, also can be computerized system and software.Such as, various artificial intelligence system can be realized by man-machine interaction, such as, intelligent customer service system, speech control system etc.Artificial intelligence semantics recognition is the basis of man-machine interaction, and it can identify human language, to convert the language that machine can be understood to.
Intelligent Answer System is a kind of typical apply of man-machine interaction, and wherein after user asks a question, Intelligent Answer System provides the answer of this problem.For this reason, have a set of knowledge base in Intelligent Answer System, there are a large amount of problems and the answer corresponding with each problem in the inside.First Intelligent Answer System needs the problem identifying that user proposes, and namely finds from knowledge base and the problem corresponding to this customer problem, then finds out the answer matched with this problem.
The maintenance update of Intelligent Answer System is a significant challenge.
Summary of the invention
Below provide the brief overview of one or more aspect to provide the basic comprehension to these aspects.Detailed the combining of this not all aspect contemplated of general introduction is look at, and both not intended to be pointed out out the scope of key or decisive any or all aspect of elements nor delineate of all aspects.Its unique object is the sequence that some concepts that will provide one or more aspect in simplified form think the more detailed description provided after a while.
According to an aspect of the present invention, provide a kind of method that abstract semantics storehouse is optimized in correct daily record storehouse based on Intelligent Answer System, this correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard, this abstract semantics storehouse comprises the abstract semantics set of multiple classification, the abstract semantics set of each classification comprises multiple abstract semantics expression formula, and the method comprises:
This standard is asked and carries out abstract semantics recommendation;
Abstract semantics recommendation is carried out to similar the asking that this standard is asked;
This similar abstract semantics of asking of asking in response to this standard is recommended unsuccessfully, automatically generates corresponding abstract semantics expression formula according to this similar asking;
Newly-generated abstract semantics expression formula is added into this standard ask recommended to classification corresponding to abstract semantics expression formula abstract semantics set in.
In one example, this abstract semantics is recommended to comprise:
Ask that similar the asking of asking with this standard carries out word segmentation processing to this standard, obtain some words respectively, this word is semantic rules word or non-semantic regular word;
Part-of-speech tagging process is carried out to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word;
Part of speech is carried out to each semantic rules word and judges process, obtain the grammatical category information of each semantic rules word;
According to this part-of-speech information and grammatical category information, search process is carried out to abstract semantics storehouse, obtain asking about to this standard it and similarly ask the abstract semantics matched respectively expression formula.
In one example, this abstract semantics expression formula comprises disappearance semantic component and semantic rules word, asks that the abstract semantics expression formula of mating meets the following conditions with this standard:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks middle corresponding content;
Abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of representation that order and the standard of abstract semantics expression formula are asked is identical, and
Similarly ask that the abstract semantics expression formula of mating meets the following conditions with this:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content;
With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of abstract semantics expression formula is identical with similar order of representation of asking.
In one example, this comprises according to this similar abstract semantics expression formula of asking that generation is corresponding automatically:
Carry out participle to obtain some words to this similar asking, each word is semantic rules word or non-semantic regular word;
Part-of-speech tagging is carried out to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word; And
Each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule by the part-of-speech information at least based on each non-semantic regular word.
In one example, each non-semantic regular word is at least replaced with corresponding semantic component symbol based on the part-of-speech information of each non-semantic regular word and also comprises by this:
Based on this similar context of asking, each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule.
In one example, this newly-generated abstract semantics expression formula is asked in the abstract semantics set of corresponding classification being added into this standard after manual confirmation is correct.
In one example, the method also comprises this similar abstract semantics of asking of asking in response to described standard and recommends successfully but ask corresponding different classes of abstract semantics set with described standard, to this similar ask recommended to abstract semantics expression formula carry out manual confirmation.
In one example, the method also comprises: ask that classification corresponding to the abstract semantics expression formula that recommends to carries out manual confirmation to this standard.
According to a further aspect in the invention, provide the device that abstract semantics storehouse is optimized in a kind of correct daily record storehouse based on Intelligent Answer System, this correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard, this abstract semantics storehouse comprises the abstract semantics set of multiple classification, the abstract semantics set of each classification comprises multiple abstract semantics expression formula, and this device comprises:
Abstract semantics recommending module, carries out abstract semantics recommendation for asking this standard, and carries out abstract semantics recommendation to similar the asking that this standard is asked;
Abstract semantics expression formula generation module, recommends unsuccessfully for this similar abstract semantics of asking of asking in response to this standard, automatically generates corresponding abstract semantics expression formula according to this similar asking; And
Abstract semantics storehouse editor module, for newly-generated abstract semantics expression formula is added into this standard ask recommended to classification corresponding to abstract semantics expression formula abstract semantics set in.
In one example, this abstract semantics pushes away module and comprises:
To this standard, word-dividing mode, asks that similar the asking of asking with this standard carries out word segmentation processing, obtains some words respectively, this word is semantic rules word or non-semantic regular word;
Part-of-speech tagging module, for carrying out part-of-speech tagging process to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word;
Part of speech judge module, judges process for carrying out part of speech to each semantic rules word, obtains the grammatical category information of each semantic rules word;
Search module, for carrying out search process according to this part-of-speech information and grammatical category information to abstract semantics storehouse, obtaining asking about to this standard it and similarly asking the abstract semantics matched respectively expression formula.
In one example, this abstract semantics expression formula comprises disappearance semantic component and semantic rules word, asks that the abstract semantics expression formula of mating meets the following conditions with this standard:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks middle corresponding content;
Abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of representation that order and the standard of abstract semantics expression formula are asked is identical, and
Similarly ask that the abstract semantics expression formula of mating meets the following conditions with this:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content;
With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of abstract semantics expression formula is identical with similar order of representation of asking.
In one example, this abstract semantics expression formula generation module comprises:
Word-dividing mode, for carrying out participle to obtain some words to this similar asking, each word is semantic rules word or non-semantic regular word;
Part-of-speech tagging module, for carrying out part-of-speech tagging to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word; And
Packing module, replaces with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule for the part-of-speech information at least based on each non-semantic regular word by each non-semantic regular word.
In one example, this packing module is further used for, based on this similar context of asking, each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule.
In one example, this similarly asks that the abstract semantics expression formula that automatically generates is asked in the abstract semantics set of corresponding classification being added into this standard after manual confirmation is correct according to this.
According to the solution of the present invention, similar process is carried out to all similar the asking that standard is asked, as long as similar asking does not navigate to a certain abstract semantics classification, just generate this similar abstract semantics expression formula of asking and add in corresponding abstract semantics classification, with this abstract semantics classification of improvement and optimization.Carrying out identical process by asking all standards in correct daily record storehouse, to optimize whole abstract semantics storehouse, thus knowledge O&M efficiency can be accelerated, reduce the cost of labor of knowledge O&M.
Accompanying drawing explanation
After the detailed description of reading embodiment of the present disclosure in conjunction with the following drawings, above-mentioned feature and advantage of the present invention can be understood better.In the accompanying drawings, each assembly is not necessarily drawn in proportion, and the assembly with similar correlation properties or feature may have identical or close Reference numeral.
Fig. 1 shows the process flow diagram of the method in the optimization abstract semantics storehouse, correct daily record storehouse based on Intelligent Answer System;
Fig. 2 shows the process flow diagram of the method for generating abstract semantics expression formula according to an aspect of the present invention; And
Fig. 3 shows the block diagram of the device in the optimization abstract semantics storehouse, correct daily record storehouse based on Intelligent Answer System.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.Note, the aspects described below in conjunction with the drawings and specific embodiments is only exemplary, and should not be understood to carry out any restriction to protection scope of the present invention.
The most original and the simplest form of basic knowledge point in knowledge base is exactly the FAQ commonly used at ordinary times, and general form is that " ask-answer " is right.In the present invention, " standard is asked " is used to the word representing certain knowledge point, and main target is that expression is clear, is convenient to safeguard.Such as, " rate of CRBT " are exactly express standard clearly to ask description.Here " asking " should be narrowly interpreted as " inquiry ", and broadly should understand one " input ", should " input " have corresponding " output ".Such as, for the semantics recognition for control system, an instruction of user, such as, " turn on radio " and also should be understood to be one " asking ", and now corresponding " answering " can be calling of control program for performing corresponding control.
User is when inputting to machine, and optimal situation is that use standard is asked, then the intelligent semantic recognition system of machine can understand the meaning of user at once.But user often not uses standard to ask, but some forms of being out of shape that standard is asked.Such as, if the standard form of asking switched for wireless radio station is " changing a radio station ", the order that so user may use is " switching a radio station ", and it is the same meaning that machine also needs to identify that user expresses.
Therefore, for intelligent semantic identification, the expansion that the standard that needs in knowledge base is asked is asked, this expansion is asked and asked that expression-form has difference slightly with standard, but expresses identical implication.
Further, in order to identify customer problem more accurately and efficiently, Intelligent Answer System also been developed the concept of abstract semantics.Abstract semantics is abstract further to body generic attribute.The difference that the abstract semantics of a classification describes a class abstract semantics by the set of one group of abstract semantics expression formula is expressed, and for expressing more abstract semanteme, these abstract semantics expression formulas expand on component.The element expanded when these just can express various concrete semanteme once be endowed corresponding value.
Each abstract semantics expression formula mainly can comprise disappearance semantic component and semantic rules word.Disappearance semantic component is accorded with by semantic component and representing, can express concrete semanteme miscellaneous after the semantic component of these disappearances has been filled corresponding value (i.e. content).
The semantic component symbol of abstract semantics can comprise:
[concept]: the word or the phrase that represent main body or object composition.
Such as: " CRBT " in " how open-minded CRBT is "
[action]: the word or the phrase that represent action composition.
Such as: " handling " in " how credit card is handled "
[attribute]: the word or the phrase that represent attribute composition.
Such as: " color " in " which color iphone has "
[adjective]: the word or the phrase that represent ornamental equivalent.
Such as: " cheaply " in " which brand of refrigerator is cheap "
Some main abstract semantics classification examples have:
What conceptual illustration [concept] is
Attribute forms [concept] for which [attribute]
How [action] behavior [concept]
Behavior place [concept] somewhere [action]
Behavioral reasons [concept] why can [action]
Behavior prediction [concept] can or can not [action]
Behavior judgement [concept] has and does not have [attribute]
[attribute] of attribute situation [concept] is [adjective]
Whether determined property [concept] has [attribute]
Why so [adjective] [attribute] of attribute reason [concept]
The difference of proximate nutrition [concept1] and [concept2] where
Attribute compares [concept1] and what difference [attribute] of [concept2] has
Question sentence judges to do general judge by part-of-speech tagging at the composition of abstract semantics aspect, and the part of speech that concept is corresponding is noun, and the part of speech that the part of speech that action is corresponding is verb, attribute is corresponding is noun, adjective is corresponding is adjective.
For classification be the abstract semantics [concept] of " behavior " how [action], many abstract semantics expression formulas can be comprised under such other abstract semantics set:
Abstract semantics classification: behavior
Abstract semantics expression formula:
A. [concept] [need | should? ] [how] < [can]? does >< carry out? > [action]
b.{[concept]～[action]}
C. [concept] <? > [action] < method | mode | step? >
Which d.< has | what has | have >< to pass through | use | > [concept] [action] <'s? > [method]
E. [how] [action] ~ [concept]
Above-mentioned a, b, c, d tetra-abstract semantics expression formulas are all used to description " behavior " this abstract semantics classification.Symbol " | " represents "or" relation, symbol "? " represent that this composition is not essential.For above-mentioned abstract semantics expression formula c, deployable is that following abstract semantics is expressed:
C1. > [action] the < method > of [concept] <
C2. > [action] the < mode > of [concept] <
C3. > [action] the < step > of [concept] <
C4. the > [action] of [concept] <
C5. [concept] [action] < method >
C6. [concept] [action] < mode >
C7. [concept] [action] < step >
c8.[concept][action]
Above abstract semantics expression formula c1-c8 all launches from abstract semantics expression formula c, therefore, is similar to this abstract semantics expression formula that can launch of abstract semantics expression formula c and also can be described as degeneracy abstract semantics expression formula.
In above-mentioned abstract semantics expression formula, except according with as the abstract semantic component of disappearance semantic component, other concrete words occurred are as " how ", " should ", " method " etc., and these words need to be used in abstract semantics rule, so can be referred to as semantic rules word.
As from the foregoing, for a specific class abstract semantics, more for the abstract semantics expression formula describing this abstract semantics, this abstract semantics is more perfect.For " behavior " this abstract semantics, abstract semantics expression formula is more, and this " behavior " abstract semantics is more perfect.
Therefore, how to optimize abstract semantics storehouse, to have made more abstract semantics expression formula extremely important to annotate abstract semantics.
Fig. 1 shows the process flow diagram of the method 100 in the optimization abstract semantics storehouse, correct daily record storehouse based on Intelligent Answer System.Correct daily record storehouse is for storing the database of correct daily record in Intelligent Answer System.Correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard.
Abstract semantics storehouse comprises the abstract semantics set of multiple classification, " conceptual illustration ", " attribute formation " described above, " behavior " etc.The abstract semantics set of each classification comprises multiple abstract semantics expression formula to explain such other abstract semantics.Still for the abstract semantics set of above-mentioned " behavior " this classification, comprise abstract semantics expression formula a, b, c, d, e.
In step 102, standard is asked and carries out abstract semantics recommendation.
A standard in correct daily record storehouse is asked, carries out abstract semantics recommendation.In this step, need to find in abstract semantics storehouse and ask corresponding abstract semantics classification with this standard.For " how looking into violating the regulations " this standard is asked, corresponding classification is the abstract semantics set of " behavior ".
In one example, this abstract semantics is recommended first to ask this standard and is carried out word segmentation processing, and obtain some words, this word is semantic rules word or non-semantic regular word.
How such as, " how to look into " can be divided into word " ", " looking into ", " breaking rules and regulations " violating the regulations.In these words, " how " be semantic rules word, " looking into " and " breaking rules and regulations " is non-semantic rules word.
Then, carry out part-of-speech tagging process to each non-semantic regular word respectively, such as " look into " and be noted as verb, " breaking rules and regulations " is noted as noun.
Afterwards, part of speech is carried out to each semantic rules word and judges process, obtain the grammatical category information of each semantic rules word.Part of speech is simply understood and is the word that a group has general character, these words semantically can similar also can be dissimilar.
Finally, according to these part-of-speech information and grammatical category information, search process is carried out to abstract semantics storehouse, obtain the abstract semantics expression formula of asking that with standard " how looking into violating the regulations " mates.
In practice, the abstract semantics expression formula of mating with user meets the following conditions:
1) part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks corresponding content;
2) abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
3) order of abstract semantics expression formula is identical with the order of representation that standard is asked.
In above-mentioned abstract semantics classification " behavior ", the part of speech of the disappearance semantic component action of abstract semantics expression formula e is verb, it is also verb that corresponding filling content that standard is asked " how looking into violating the regulations " " is looked into ", the part of speech of disappearance semantic component concept is noun, corresponding filling content that standard is asked " how looking into violating the regulations " " breaks rules and regulations " to be also noun, therefore meets above-mentioned condition 1).
How secondly, semantic rules word in abstract semantics expression formula e " how " asks that with standard " how looking into violating the regulations " middle corresponding semantic rules word " " belongs to same part of speech, therefore meets above-mentioned condition 2).
Finally, the order of representation that the order of abstract semantics expression formula e is also asked with standard is identical, meets above-mentioned condition 3).
Therefore, in abstract semantics storehouse, find the abstract semantics expression formula e asking that with standard " how looking into violating the regulations " mates, namely [how] [action] ~ [concept].Because this abstract semantics expression formula belongs to " behavior " classification, therefore for it recommends the abstract semantics set of " behavior " this classification.
It should be noted that not to be an abstract semantics expression formula finding coupling surely, so some standard asks the abstract semantics that can not find recommendation, now, this standard asks that corresponding similar asking cannot optimize abstract semantics storehouse.
Ask " which smoking has endanger " for standard again, can be recommended to abstract semantics classification: attribute is formed.
In order to ensure correctly, can ask that classification corresponding to the abstract semantics expression formula that recommends to carries out manual confirmation to standard.
In step 104, abstract semantics recommendation is carried out to similar the asking that this standard is asked.
The similar abstract semantics of asking that this standard is asked recommends the abstract semantics of asking with standard to recommend similar.First word segmentation processing is carried out to this similar asking, obtain some words respectively, word is semantic rules word or non-semantic regular word, then part-of-speech tagging process is carried out to each non-semantic regular word, obtain the part-of-speech information of each non-semantic regular word, part of speech is carried out to each semantic rules word and judges process, obtain the grammatical category information of each semantic rules word, according to part-of-speech information and grammatical category information, search process is carried out to abstract semantics storehouse again, obtain asking the abstract semantics matched expression formula to similar.
Ask that the abstract semantics expression formula of coupling meets the following conditions with similar: the part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content; With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech; The order of abstract semantics expression formula is identical with similar order of representation of asking.
In step 106, recommend unsuccessfully in response to this similar abstract semantics of asking, automatically generate corresponding abstract semantics expression formula according to this similar asking.
As mentioned above, not all similarly ask the abstract semantics expression formula that can search coupling, if fail to find the abstract semantics expression formula of coupling, be then considered as abstract semantics and recommend unsuccessfully.
Ask that the similar of " which smoking has endanger " is asked " what the harm of smoking has " for standard, according to reason, if standard asks the classification of corresponding some abstract semantics expression formulas, so this standard asks that all similar asking below all should corresponding same classification.But may to form the abstract semantics set of this classification perfect not enough due to attribute, ask that so similar " what the harm of smoking has " is not recommended to such other abstract semantics set, and the not recommended abstract semantics set to other classifications, then need to ask for this is similar and automatically generate corresponding abstract semantics expression formula, then joined in this abstract semantics set to improve such other abstract semantics set.
If this similar abstract semantics of asking is recommended successfully in addition, i.e. recommended abstract semantics set of having arrived certain classification, but this is similar asks and asks corresponding different classes of abstract semantics set with its this standard, be likely then that mistake appears in previous abstract semantics expression formula, therefore can carry out manual confirmation.
Fig. 2 shows the process flow diagram of the method 200 for generating abstract semantics expression formula according to an aspect of the present invention.
In step 202, participle is carried out, to obtain some words to this similar asking.
For following several language materials:
1) what the harm of smoking has;
2) what eye cream is for black-eyed effect;
3) to stay up late the harm brought to skin;
4) matter acid effect in skin.
For language material 1), participle is: what the harm of smoking has
For language material 2), participle is: what eye cream is for black-eyed effect
For language material 3), participle is: the harm brought to skin of staying up late
For language material 4), participle is: matter acid effect in skin.
Each word identification is semantic rules word or non-semantic regular word, and this semantic rules word belongs to the word that abstract semantics expression formula is used.Such as can search each word in semantic rules dictionary, semantic rules dictionary comprises the set belonging to all words that abstract semantics expression formula is used.If some words are present in semantic rules dictionary, are semantic rules word by this word identification, otherwise be identified as non-semantic regular word.And for example: can also judge with part of speech, as the part of speech such as preposition, auxiliary word.If some words are preposition or auxiliary word, then this word identification is semantic rules word, otherwise is identified as non-semantic regular word.
For language material 1), semantic rules word comprises: what has
For language material 2), semantic rules word comprises: for what is
For language material 3), semantic rules word comprises: to what bring
For language material 4), semantic rules word comprises: in effect
In step 204, respectively part-of-speech tagging is carried out to each non-semantic regular word, obtain the part-of-speech information of each non-semantic regular word.
For language material 1), non-semantic regular word: (verb) harm (noun) of smoking
For language material 2), non-semantic regular word: eye cream (noun) livid ring around eye (noun) effect (noun)
For language material 3), non-semantic regular word: (noun) skin (noun) of staying up late harm (noun)
For language material 4), non-semantic regular word: matter acid (noun) skin (noun) effect (noun)
In step 206, non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics expression formula by the part-of-speech information at least based on each non-semantic regular word.
Such as, the non-semantic regular word that part of speech is marked as noun can be replaced with the semantic component symbol concept of the word or phrase that represent main body or object composition, the non-semantic regular word that part of speech is marked as verb can be replaced with the semantic component symbol action of the word or phrase that represent action composition, part of speech is marked as adjectival non-semantic regular word and can replaces with the semantic component symbol adjective of the word or phrase that represent ornamental equivalent, and part of speech is marked as the non-semantic regular word also word of available expression attribute composition or the semantic component symbol attribute replacement of phrase of noun.
In addition, also based on the context of language material, non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics expression formula.
Above predicate material 1)-4) be example:
For language material 1), the part of speech of " smoking " is verb, but considers context of co-text, and smoking here represents subject composition, therefore smoking is replaced with concept; The part of speech of " harm " is noun, considers context of co-text, is replaced by " attribute ".
Thus, language material 1) abstract semantics expression formula be: the attribute of concept has anything.
For language material 2), the part of speech of " eye cream " is noun, and represents bulk composition, replaces with concept, the part of speech of " livid ring around eye " is noun, represents guest species, replaces with " concept ", the part of speech of " effect " is noun, represents attribute composition here, replaces with " attribute ".
Thus, language material 2) abstract semantics expression formula be: what concept1 is for the attribute of concept2.
Similarly, language material 3) abstract semantics expression formula be: the attribute that concept1 brings to concept2.
Language material 4) abstract semantics expression formula be: the attribute of concept1 in concept2.
In step 108, newly-generated abstract semantics expression formula is added into standard and asks in the abstract semantics set of corresponding classification.
Such as, can be added to based on similar abstract semantics expression formula " what attribute of concept has " of asking that " what harm of smoking has " generates in the set of the abstract semantics " which [attribute] [concept] has " that classification is " attribute formation ".
More preferably, newly-generated abstract semantics expression formula is asked in the abstract semantics set of corresponding classification being added into standard after manual confirmation is correct.
In this way, similar process is carried out to all similar the asking that standard is asked, as long as similar asking to be asked with standard and do not navigated to same abstract semantics classification, just generate this similar abstract semantics expression formula of asking and add in corresponding abstract semantics classification, with this abstract semantics classification of improvement and optimization.Again in this way, all standards in correct daily record storehouse are asked and carries out identical process, to optimize whole abstract semantics storehouse, thus knowledge O&M efficiency can be accelerated, reduce the cost of labor of knowledge O&M.
Said method illustrated although simplify for making explanation and is described as a series of actions, it should be understood that and understand, these methods not limit by the order of action, because according to one or more embodiment, some actions can occur by different order and/or with from illustrating herein and describe or not shown and to describe but other actions that it will be appreciated by those skilled in the art that occur concomitantly herein.
Fig. 3 shows the block diagram of the device 300 in the optimization abstract semantics storehouse, correct daily record storehouse based on Intelligent Answer System.Correct daily record storehouse is for storing the database of correct daily record in Intelligent Answer System.Correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard.
Abstract semantics storehouse comprises the abstract semantics set of multiple classification, " conceptual illustration ", " attribute formation " described above, " behavior " etc.The abstract semantics set of each classification comprises multiple abstract semantics expression formula to explain such other abstract semantics.Still for the abstract semantics set of above-mentioned " behavior " this classification, comprise abstract semantics expression formula a, b, c, d, e.
Device 300 can comprise abstract semantics recommending module 302, abstract semantics expression formula generation module 304 and abstract semantics storehouse editor module 306.
Abstract semantics recommending module 302 can be asked standard and be carried out abstract semantics recommendation.
A standard in correct daily record storehouse is asked, carries out abstract semantics recommendation, that is, need to find in abstract semantics storehouse and ask corresponding abstract semantics classification with this standard.For " how looking into violating the regulations " this standard is asked, corresponding classification is the abstract semantics set of " behavior ".
In one example, this abstract semantics recommending module 302 can comprise word-dividing mode 3022, part-of-speech tagging module 3024, part of speech judge module 3026, search module 3028.
First word-dividing mode 3022 can ask this standard and carry out word segmentation processing, obtains some words, and this word is semantic rules word or non-semantic regular word.
How such as, " how to look into " can be divided into word " ", " looking into ", " breaking rules and regulations " violating the regulations.In these words, " how " be semantic rules word, " looking into " and " breaking rules and regulations " is non-semantic rules word.
Then, part-of-speech tagging module 3024 can carry out part-of-speech tagging process to each non-semantic regular word respectively, and such as " look into " and be noted as verb, " breaking rules and regulations " is noted as noun.
Afterwards, part of speech judge module 3026 can carry out part of speech to each semantic rules word and judge process, obtains the grammatical category information of each semantic rules word.Part of speech is simply understood and is the word that a group has general character, these words semantically can similar also can be dissimilar.
Finally, search module 3028 can carry out search process according to these part-of-speech information and grammatical category information to abstract semantics storehouse, obtains the abstract semantics expression formula of asking that with standard " how looking into violating the regulations " mates.
In practice, the abstract semantics expression formula of mating with user meets the following conditions:
1) part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks corresponding content;
2) abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
3) order of abstract semantics expression formula is identical with the order of representation that standard is asked.
In above-mentioned abstract semantics classification " behavior ", the part of speech of the disappearance semantic component action of abstract semantics expression formula e is verb, it is also verb that corresponding filling content that standard is asked " how looking into violating the regulations " " is looked into ", the part of speech of disappearance semantic component concept is noun, corresponding filling content that standard is asked " how looking into violating the regulations " " breaks rules and regulations " to be also noun, therefore meets above-mentioned condition 1).
How secondly, semantic rules word in abstract semantics expression formula e " how " asks that with standard " how looking into violating the regulations " middle corresponding semantic rules word " " belongs to same part of speech, therefore meets above-mentioned condition 2).
Finally, the order of representation that the order of abstract semantics expression formula e is also asked with standard is identical, meets above-mentioned condition 3).
Therefore, in abstract semantics storehouse, find the abstract semantics expression formula e asking that with standard " how looking into violating the regulations " mates, namely [how] [action] ~ [concept].Because this abstract semantics expression formula belongs to " behavior " classification, therefore for it recommends the abstract semantics set of " behavior " this classification.
It should be noted that not to be an abstract semantics expression formula finding coupling surely, so some standard asks the abstract semantics that can not find recommendation, now, this standard asks that corresponding similar asking cannot optimize abstract semantics storehouse.
Ask " which harm (its abstract semantics expression formula is: which [attribute] [concept] has) smoking has " for standard again, can be recommended to abstract semantics classification: attribute is formed.
In order to ensure correctly, can ask that classification corresponding to the abstract semantics expression formula that recommends to carries out manual confirmation to standard.
Abstract semantics recommending module 302 also can carry out abstract semantics recommendation to similar the asking that this standard is asked.
The similar abstract semantics of asking that this standard is asked recommends the abstract semantics of asking with standard to recommend similar.First by word-dividing mode 3022, word segmentation processing is carried out to this similar asking, obtain some words respectively, word is semantic rules word or non-semantic regular word, then by part-of-speech tagging module 3024, part-of-speech tagging process is carried out to each non-semantic regular word, obtain the part-of-speech information of each non-semantic regular word, by part of speech judge module 3026, part of speech is carried out to each semantic rules word again and judge process, obtain the grammatical category information of each semantic rules word, finally carry out search according to part-of-speech information and grammatical category information to abstract semantics storehouse by search module 3028 to process, obtain asking the abstract semantics matched expression formula to similar.
Ask that the abstract semantics expression formula of coupling meets the following conditions with similar: the part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content; With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech; The order of abstract semantics expression formula is identical with similar order of representation of asking.
Abstract semantics expression formula generation module 304 can be recommended unsuccessfully in response to this similar abstract semantics of asking, automatically generates corresponding abstract semantics expression formula according to this similar asking.
As mentioned above, not all similarly ask the abstract semantics expression formula that can search coupling, if fail to find the abstract semantics expression formula of coupling, be then considered as abstract semantics and recommend unsuccessfully.
Ask that the similar of " which smoking has endanger " is asked " what the harm of smoking has " for standard, according to reason, if standard asks corresponding some abstract semantics, so this standard asks that all similar asking below all should corresponding same abstract semantics.But may to form the abstract semantics set of this classification perfect not enough due to attribute, ask that so similar " what the harm of smoking has " is not recommended to such other abstract semantics set, and the not recommended abstract semantics set to other classifications, then need to ask for this is similar and automatically generate corresponding abstract semantics expression formula, then joined in this abstract semantics set to improve such other abstract semantics set.
Such as, similarly ask that " what the harm of smoking has " not corresponding standard may ask that " which smoking has endanger " institute is recommended to abstract semantics classification: attribute is formed.Therefore need to generate this similar abstract semantics expression formula of asking.
If this similar abstract semantics of asking is recommended successfully in addition, i.e. recommended abstract semantics set of having arrived certain classification, but this is similar asks and asks corresponding different classes of abstract semantics set with its this standard, be likely then that mistake appears in previous abstract semantics expression formula, therefore can carry out manual confirmation.
In one example, abstract semantics expression formula generation module 304 can comprise word-dividing mode 3042, part-of-speech tagging module 3044 and packing module 3046.
Word-dividing mode 3042 can carry out participle to this similar asking, to obtain some words.
For following several language materials:
1) what the harm of smoking has;
2) what eye cream is for black-eyed effect;
3) to stay up late the harm brought to skin;
4) matter acid effect in skin.
For language material 1), participle is: what the harm of smoking has
For language material 2), participle is: what eye cream is for black-eyed effect
For language material 3), participle is: the harm brought to skin of staying up late
For language material 4), participle is: matter acid effect in skin.
Each word is semantic rules word or non-semantic regular word, and this semantic rules word belongs to the word that abstract semantics expression formula is used.Such as can search each word in semantic rules dictionary, semantic rules dictionary comprises the set belonging to all words that abstract semantics expression formula is used.If some words are present in semantic rules dictionary, are semantic rules word by this word identification, otherwise be identified as non-semantic regular word.And for example: can also judge with part of speech, as the part of speech such as preposition, auxiliary word.If some words are preposition or auxiliary word, then this word identification is semantic rules word, otherwise is identified as non-semantic regular word.
For language material 1), semantic rules word comprises: what has
For language material 2), semantic rules word comprises: for what is
For language material 3), semantic rules word comprises: to what bring
For language material 4), semantic rules word comprises: in effect
Part-of-speech tagging module 3044 can carry out part-of-speech tagging to each non-semantic regular word respectively, obtains the part-of-speech information of each non-semantic regular word.
For language material 1), non-semantic regular word: (verb) harm (noun) of smoking
For language material 2), non-semantic regular word: eye cream (noun) livid ring around eye (noun) effect (noun)
For language material 3), non-semantic regular word: (noun) skin (noun) of staying up late harm (noun)
For language material 4), non-semantic regular word: matter acid (noun) skin (noun) effect (noun)
Non-semantic regular word at least can be replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics expression formula based on the part-of-speech information of each non-semantic regular word by packing module 3046.
Such as, the non-semantic regular word that part of speech is marked as noun can be replaced with the semantic component symbol concept of the word or phrase that represent main body or object composition, the non-semantic regular word that part of speech is marked as verb can be replaced with the semantic component symbol action of the word or phrase that represent action composition, part of speech is marked as adjectival non-semantic regular word and can replaces with the semantic component symbol adjective of the word or phrase that represent ornamental equivalent, and part of speech is marked as the non-semantic regular word also word of available expression attribute composition or the semantic component symbol attribute replacement of phrase of noun.
In addition, non-semantic regular word also can be replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics expression formula based on the context of language material by packing module 3046.
Above predicate material 1)-4) be example:
For language material 1), the part of speech of " smoking " is verb, but considers context of co-text, and smoking here represents subject composition, therefore smoking is replaced with concept; The part of speech of " harm " is noun, considers context of co-text, is replaced by " attribute ".
Thus, language material 1) abstract semantics expression formula be: the attribute of concept has anything.
For language material 2), the part of speech of " eye cream " is noun, and represents bulk composition, replaces with concept, the part of speech of " livid ring around eye " is noun, represents guest species, replaces with " concept ", the part of speech of " effect " is noun, represents attribute composition here, replaces with " attribute ".
Thus, language material 2) abstract semantics expression formula be: what concept1 is for the attribute of concept2.
Similarly, language material 3) abstract semantics expression formula be: the attribute that concept1 brings to concept2.
Language material 4) abstract semantics expression formula be: the attribute of concept1 in concept2.
Newly-generated abstract semantics expression formula can be added into standard and ask in the abstract semantics set of corresponding classification by abstract semantics storehouse editor module 306.
Such as, can be added to based on similar abstract semantics expression formula " what attribute of concept has " of asking that " what harm of smoking has " generates in the set of the abstract semantics " which [attribute] [concept] has " that classification is " attribute formation ".
More preferably, newly-generated abstract semantics expression formula is asked in the abstract semantics set of corresponding classification being added into standard after manual confirmation is correct.
According to the solution of the present invention, similar process is carried out to all similar the asking that standard is asked, as long as similar asking does not navigate to a certain abstract semantics classification, just generate this similar abstract semantics expression formula of asking and add in corresponding abstract semantics classification, with this abstract semantics classification of improvement and optimization.Carrying out identical process by asking all standards in correct daily record storehouse, to optimize whole abstract semantics storehouse, thus knowledge O&M efficiency can be accelerated, reduce the cost of labor of knowledge O&M.
Those skilled in the art will understand further, and the various illustrative logic plates, module, circuit and the algorithm steps that describe in conjunction with embodiment disclosed herein can be embodied as electronic hardware, computer software or the combination of both.For clearly explaining orally this interchangeability of hardware and software, various illustrative components, frame, module, circuit and step are done vague generalization above with its functional form and are described.This type of is functional is implemented as hardware or software depends on embody rule and puts on the design constraint of total system.Technician can realize described functional by different modes for often kind of application-specific, but such realize decision-making and should not be interpreted to and cause having departed from scope of the present invention.
Software should be construed broadly into mean instruction, instruction set, code, code segment, program code, program, subroutine, software module, application, software application, software package, routine, subroutine, object, can executive item, execution thread, code, function etc., no matter it is that to address with software, firmware, middleware, microcode, hardware description language or other term be all like this.
The various illustrative logic plates, module and the circuit that describe in conjunction with embodiment disclosed herein can realize with general processor, digital signal processor (DSP), special IC (ASIC), field programmable gate array (FPGA) or other programmable logic device (PLD), discrete door or transistor logic, discrete nextport hardware component NextPort or its any combination being designed to perform function described herein or perform.General processor can be microprocessor, but in alternative, and this processor can be the processor of any routine, controller, microcontroller or state machine.Processor can also be implemented as the combination of computing equipment, the combination of such as DSP and microprocessor, multi-microprocessor, with one or more microprocessor of DSP central cooperation or any other this type of configure.
The method described in conjunction with embodiment disclosed herein or the step of algorithm can be embodied directly in hardware, in the software module performed by processor or in the combination of both and embody.Software module can reside in the storage medium of RAM storer, flash memory, ROM storer, eprom memory, eeprom memory, register, hard disk, removable dish, CD-ROM or any other form known in the art.Exemplary storage medium is coupled to processor and can reads and written information from/to this storage medium to make this processor.In alternative, storage medium can be integrated into processor.
Thering is provided previous description of the present disclosure is for making any person skilled in the art all can make or use the disclosure.To be all apparent for a person skilled in the art to various amendment of the present disclosure, and generic principles as defined herein can be applied to other variants and can not depart from spirit or scope of the present disclosure.Thus, the disclosure not intended to be is defined to example described herein and design, but the widest scope consistent with principle disclosed herein and novel features should be awarded.
Claims (14)
1. the method based on the optimization abstract semantics storehouse, correct daily record storehouse of Intelligent Answer System, described correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard, described abstract semantics storehouse comprises the abstract semantics set of multiple classification, the abstract semantics set of each classification comprises multiple abstract semantics expression formula, and described method comprises:
Described standard is asked and carries out abstract semantics recommendation;
Abstract semantics recommendation is carried out to similar the asking that described standard is asked;
This similar abstract semantics of asking of asking in response to described standard is recommended unsuccessfully, automatically generates corresponding abstract semantics expression formula according to this similar asking;
Newly-generated abstract semantics expression formula is added into described standard ask recommended to classification corresponding to abstract semantics expression formula abstract semantics set in.
2. the method for claim 1, is characterized in that, described abstract semantics is recommended to comprise:
Ask that similar the asking of asking with described standard carries out word segmentation processing to described standard, obtain some words respectively, described word is semantic rules word or non-semantic regular word;
Part-of-speech tagging process is carried out to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word;
Part of speech is carried out to each semantic rules word and judges process, obtain the grammatical category information of each semantic rules word;
According to described part-of-speech information and grammatical category information, search process is carried out to abstract semantics storehouse, obtain asking about to described standard it and similarly ask the abstract semantics matched respectively expression formula.
3. method as claimed in claim 2, is characterized in that, described abstract semantics expression formula comprises disappearance semantic component and semantic rules word, asks that the abstract semantics expression formula of mating meets the following conditions with described standard:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks middle corresponding content;
Abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of representation that order and the standard of abstract semantics expression formula are asked is identical, and
Similarly ask that the abstract semantics expression formula of mating meets the following conditions with described:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content;
With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of abstract semantics expression formula is identical with similar order of representation of asking.
4. the method for claim 1, is characterized in that, describedly similarly asks that automatically generating corresponding abstract semantics expression formula comprises according to this:
Carry out participle to obtain some words to described similar asking, each word is semantic rules word or non-semantic regular word;
Part-of-speech tagging is carried out to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word; And
Each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule by the part-of-speech information at least based on each non-semantic regular word.
5. method as claimed in claim 4, is characterized in that, describedly at least based on the part-of-speech information of each non-semantic regular word, each non-semantic regular word is replaced with corresponding semantic component symbol and also comprises:
Based on described similar context of asking, each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule.
6. the method for claim 1, is characterized in that, described newly-generated abstract semantics expression formula is asked in the abstract semantics set of corresponding classification being added into described standard after manual confirmation is correct.
7. the method for claim 1, it is characterized in that, this similar abstract semantics of asking of asking in response to described standard is recommended successfully but asks corresponding different classes of abstract semantics set with described standard, to this similar ask recommended to abstract semantics expression formula carry out manual confirmation.
8. the method for claim 1, is characterized in that, also comprises:
Ask that classification corresponding to the abstract semantics expression formula that recommends to carries out manual confirmation to described standard.
9. the device based on the optimization abstract semantics storehouse, correct daily record storehouse of Intelligent Answer System, described correct daily record storehouse comprises standard and asks and ask that be associated similar is asked with each standard, described abstract semantics storehouse comprises the abstract semantics set of multiple classification, the abstract semantics set of each classification comprises multiple abstract semantics expression formula, and described device comprises:
Abstract semantics recommending module, carries out abstract semantics recommendation for asking described standard, and carries out abstract semantics recommendation to similar the asking that described standard is asked;
Abstract semantics expression formula generation module, recommends unsuccessfully for this similar abstract semantics of asking of asking in response to described standard, automatically generates corresponding abstract semantics expression formula according to this similar asking; And
Abstract semantics storehouse editor module, for newly-generated abstract semantics expression formula is added into described standard ask recommended to classification corresponding to abstract semantics expression formula abstract semantics set in.
10. device as claimed in claim 9, it is characterized in that, described abstract semantics pushes away module and comprises:
To described standard, word-dividing mode, asks that similar the asking of asking with described standard carries out word segmentation processing, obtains some words respectively, described word is semantic rules word or non-semantic regular word;
Part-of-speech tagging module, for carrying out part-of-speech tagging process to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word;
Part of speech judge module, judges process for carrying out part of speech to each semantic rules word, obtains the grammatical category information of each semantic rules word;
Search module, for carrying out search process according to described part-of-speech information and grammatical category information to abstract semantics storehouse, obtaining asking about to described standard it and similarly asking the abstract semantics matched respectively expression formula.
11. devices as claimed in claim 10, is characterized in that, described abstract semantics expression formula comprises disappearance semantic component and semantic rules word, asks that the abstract semantics expression formula of mating meets the following conditions with described standard:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises the part of speech that standard asks middle corresponding content;
Abstract semantics expression formula and standard ask that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of representation that order and the standard of abstract semantics expression formula are asked is identical, and
Similarly ask that the abstract semantics expression formula of mating meets the following conditions with described:
The part of speech that the disappearance semantic component of abstract semantics expression formula is corresponding comprises similar part of speech of asking middle corresponding content;
With similar, abstract semantics expression formula asks that the semantic rules word of middle correspondence is identical or belong to same part of speech;
The order of abstract semantics expression formula is identical with similar order of representation of asking.
12. devices as claimed in claim 9, is characterized in that, described abstract semantics expression formula generation module comprises:
Word-dividing mode, for carrying out participle to obtain some words to described similar asking, each word is semantic rules word or non-semantic regular word;
Part-of-speech tagging module, for carrying out part-of-speech tagging to each non-semantic regular word, obtains the part-of-speech information of each non-semantic regular word; And
Packing module, replaces with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule for the part-of-speech information at least based on each non-semantic regular word by each non-semantic regular word.
13. devices as claimed in claim 12, is characterized in that, described packing module is further used for, based on described similar context of asking, each non-semantic regular word is replaced with corresponding semantic component symbol using the disappearance semantic component as newly-generated abstract semantics rule.
14. devices as claimed in claim 8, is characterized in that, describedly similarly ask that the abstract semantics expression formula that automatically generates is asked in the abstract semantics set of corresponding classification being added into described standard after manual confirmation is correct according to this.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511030026.1A CN105550360B (en) | 2015-12-31 | 2015-12-31 | Optimize the method and device in abstract semantics library |
| CN201810819571.6A CN109063075A (en) | 2015-12-31 | 2015-12-31 | Optimize the method, apparatus and storage medium in abstract semantics library |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511030026.1A CN105550360B (en) | 2015-12-31 | 2015-12-31 | Optimize the method and device in abstract semantics library |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810819571.6A Division CN109063075A (en) | 2015-12-31 | 2015-12-31 | Optimize the method, apparatus and storage medium in abstract semantics library |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105550360A true CN105550360A (en) | 2016-05-04 |
| CN105550360B CN105550360B (en) | 2018-09-04 |
Family
ID=55829549
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511030026.1A Active CN105550360B (en) | 2015-12-31 | 2015-12-31 | Optimize the method and device in abstract semantics library |
| CN201810819571.6A Pending CN109063075A (en) | 2015-12-31 | 2015-12-31 | Optimize the method, apparatus and storage medium in abstract semantics library |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810819571.6A Pending CN109063075A (en) | 2015-12-31 | 2015-12-31 | Optimize the method, apparatus and storage medium in abstract semantics library |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN105550360B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485328A (en) * | 2016-10-31 | 2017-03-08 | 上海智臻智能网络科技股份有限公司 | Information processing system and method |
| CN106601237A (en) * | 2016-12-29 | 2017-04-26 | 上海智臻智能网络科技股份有限公司 | Interactive voice response system and voice recognition method thereof |
| WO2020135462A1 (en) * | 2018-12-27 | 2020-07-02 | 上海智臻智能网络科技股份有限公司 | Automatic generalization method and apparatus |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339551A (en) * | 2007-07-05 | 2009-01-07 | 日电（中国）有限公司 | Natural language query demand extension equipment and its method |
| CN101799822A (en) * | 2010-02-23 | 2010-08-11 | 李祯 | Method for modeling layered network knowledge model and method for establishing knowledge warehouse system |
| CN102968409A (en) * | 2012-11-23 | 2013-03-13 | 海信集团有限公司 | Intelligent human-machine interaction semantic analysis method and interaction system |
| CN103309886A (en) * | 2012-03-13 | 2013-09-18 | 阿里巴巴集团控股有限公司 | Trading-platform-based structural information searching method and device |
| US20130253913A1 (en) * | 2000-11-28 | 2013-09-26 | Evi Technologies Limited | Knowledge storage and retrieval system and method |
| CN104516949A (en) * | 2014-12-10 | 2015-04-15 | 北京百度网讯科技有限公司 | Webpage data processing method and apparatus, query processing method and question-answering system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010105214A2 (en) * | 2009-03-13 | 2010-09-16 | Invention Machine Corporation | Question-answering system and method based on semantic labeling of text documents and user questions |
| US9189742B2 (en) * | 2013-11-20 | 2015-11-17 | Justin London | Adaptive virtual intelligent agent |
| CN104991955B (en) * | 2015-07-17 | 2018-06-12 | 安徽科大讯飞医疗信息技术有限公司 | Method and system for automatically constructing template library |
| CN105183716B (en) * | 2015-09-21 | 2017-12-15 | 上海智臻智能网络科技股份有限公司 | A kind of intelligent interactive method based on abstract semantics |
-
2015
- 2015-12-31 CN CN201511030026.1A patent/CN105550360B/en active Active
- 2015-12-31 CN CN201810819571.6A patent/CN109063075A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130253913A1 (en) * | 2000-11-28 | 2013-09-26 | Evi Technologies Limited | Knowledge storage and retrieval system and method |
| CN101339551A (en) * | 2007-07-05 | 2009-01-07 | 日电（中国）有限公司 | Natural language query demand extension equipment and its method |
| CN101799822A (en) * | 2010-02-23 | 2010-08-11 | 李祯 | Method for modeling layered network knowledge model and method for establishing knowledge warehouse system |
| CN103309886A (en) * | 2012-03-13 | 2013-09-18 | 阿里巴巴集团控股有限公司 | Trading-platform-based structural information searching method and device |
| CN102968409A (en) * | 2012-11-23 | 2013-03-13 | 海信集团有限公司 | Intelligent human-machine interaction semantic analysis method and interaction system |
| CN104516949A (en) * | 2014-12-10 | 2015-04-15 | 北京百度网讯科技有限公司 | Webpage data processing method and apparatus, query processing method and question-answering system |
Non-Patent Citations (1)
| Title |
|---|
| 梁正平等: "基于语义模板的问答系统研究", 《深圳大学学报理工版》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485328A (en) * | 2016-10-31 | 2017-03-08 | 上海智臻智能网络科技股份有限公司 | Information processing system and method |
| CN106601237A (en) * | 2016-12-29 | 2017-04-26 | 上海智臻智能网络科技股份有限公司 | Interactive voice response system and voice recognition method thereof |
| CN106601237B (en) * | 2016-12-29 | 2020-02-07 | 上海智臻智能网络科技股份有限公司 | Interactive voice response system and voice recognition method thereof |
| WO2020135462A1 (en) * | 2018-12-27 | 2020-07-02 | 上海智臻智能网络科技股份有限公司 | Automatic generalization method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109063075A (en) | 2018-12-21 |
| CN105550360B (en) | 2018-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106649742B (en) | Database maintenance method and device | |
| Hardeniya et al. | Natural language processing: python and NLTK | |
| US9785887B2 (en) | Extraction of semantic relations using distributional relation detection | |
| US9152623B2 (en) | Natural language processing system and method | |
| Sawyer et al. | Shallow knowledge as an aid to deep understanding in early phase requirements engineering | |
| EP3804915A1 (en) | Conversational systems and methods for robotic task identification using natural language | |
| CN104049755A (en) | Information processing method and device | |
| Francesconi et al. | Integrating a bottom–up and top–down methodology for building semantic resources for the multilingual legal domain | |
| CN102439590A (en) | System and method for automatic semantic labeling of natural language texts | |
| US9734238B2 (en) | Context based passage retreival and scoring in a question answering system | |
| CN105677637A (en) | Method and device for updating abstract semantics database in intelligent question-answering system | |
| CN106601237A (en) | Interactive voice response system and voice recognition method thereof | |
| CN105550360A (en) | Method and apparatus for optimizing abstract semantic library | |
| KR20210106372A (en) | New category tag mining method and device, electronic device and computer-readable medium | |
| CN105653620A (en) | Log analysis method and device of intelligent question answering system | |
| CN114139560B (en) | Translation system based on artificial intelligence | |
| Zhong et al. | Natural language processing for systems engineering: Automatic generation of systems modelling language diagrams | |
| JP5317061B2 (en) | A simultaneous classifier in multiple languages for the presence or absence of a semantic relationship between words and a computer program therefor. | |
| CN105653619A (en) | Update method and device of correct log library in intelligent question-answering system | |
| CN110413737B (en) | Synonym determination method, synonym determination device, server and readable storage medium | |
| Malhar et al. | Deep learning based Answering Questions using T5 and Structured Question Generation System’ | |
| CN106970905A (en) | A kind of semantic analysis | |
| CN111382240B (en) | Semantic reasoning method and device | |
| CN114691820A (en) | Question-answering implementation method and device based on knowledge graph | |
| CN110197192B (en) | Natural language processing, query construction and classification |
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 |