CN101930428A - System and method for ensuring computer to understand natural languages - Google Patents
System and method for ensuring computer to understand natural languages Download PDFInfo
- Publication number
- CN101930428A CN101930428A CN2009101079753A CN200910107975A CN101930428A CN 101930428 A CN101930428 A CN 101930428A CN 2009101079753 A CN2009101079753 A CN 2009101079753A CN 200910107975 A CN200910107975 A CN 200910107975A CN 101930428 A CN101930428 A CN 101930428A
- Authority
- CN
- China
- Prior art keywords
- program
- logic
- subject
- predicate
- natural language
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Machine Translation (AREA)
Abstract
The invention relates to a system for ensuring a computer to understand natural languages, comprising a database, a related command set, a string creating unit, a subject-predicate recognizing unit, a key word recognizing unit and a logic processing unit, wherein the database is used for storing the categories of language material resources and application programs; the related command set is used for logical operation and program control of input natural languages; the string creating unit is used for generating input strings according to the input natural languages; the subject-predicate recognizing unit is used for recognizing the subject and predicate character fields in the input strings and respectively acquiring the subject and predicate object code segments matched with the subject and predicate character fields by retrieving the database; the key word recognizing unit is used for recognizing the key words/phrases in the input strings and acquiring corresponding logical operation or program control commands from the command set according to the key words/phrases; and the logic processing unit is used for combining the subject and predicate object code segments into an integral program file. The invention also provides a method for ensuring the computer to understand natural languages. The method realizes integral recognition and understanding of the natural languages by recognizing the subjects and predicates and the logical connectives thereof in the natural languages.
Description
Technical field
The present invention relates to human natural language understanding field, particularly a kind of system and method for ensuring computer to understand natural languages.
Background technology
Being used for transmission information and the language that exchanges in the human daily life, is natural language, comprises two kinds of citation forms of spoken word and written language.Natural language is that the mankind produce in the process of transforming the objective world, and the constantly development along with the variation of human society and objective world.Each related in people's natural language notion is always corresponding with a certain concrete things or incident in the objective world, because the arbitrary concrete things in the objective world or the existence form of incident be a specific unique object always, so arbitrary specific matters or the incident in one's power of the notion in people's natural language also must be unique object.The inventor is in its " the body logic theory and application " of writing (in October, 2008, science and technology publishing house in Guangdong publishes) book, and the logical proof method that utilization is strict has been founded the body logical theory based on objective uniqueness.
The body logic itself is defined as body to all and/or arbitrary things in the objective world, describing or the natural language concept definition of expression body is the title of body.So, between the body title (natural language notion) of the concrete things (body) of outwardness and expression objective things is exactly one to one, because the existence form of objective body object is unique, so human can be ideational form in the natural language by the body title, unique corresponding body object is objective things with it for mapping and coupling in the brain knowledge base.Mapping and coupling are penetrated process between this body title and the body, are exactly the process of human intelligible natural language---and the people has known what this title or notion refer to.For example when we see or hear " tree is grown on the mountain " the words, will map out immediately in the brain and " tree ", " growth ", the corresponding objective things of " on the mountain " these titles, by unique corresponding mapping with it, we have just understood the notion in the statement, and have also just understood this statement by the connection of the logic between the subject and predicate notion.If there are two or more objects to be called as " tree " in the objective world, when we see or hear the title of " tree ", just can not understand its meaning exactly so---do not know indication to as if which.This shows that objective things are the objective uniqueness of body object, is that the mankind can use natural language to come transmission information and logical foundations that exchanges and nature of philosophy on the basis of mutual understanding.This unique and have the method that notion is not understood on ambiguity ground, disclosed the essential laws and the secret of human intelligible natural language.The natural language of also having illustrated each languages can both be realized the objective base of its understanding by translation.
Obviously, this process and the method for understanding natural language by unique corresponding mapping and matching relationship, be in computer system, to use easily and realize, and now in the scheme of Computer Processing natural language, generally also be by setting up database, and realize its coupling by mapping and searching database.Yet, active computer is handled in the scheme of natural language, has only considered the independent implication of the independent morpheme in the natural language, and does not integrally consider the logical communication link between each notion part in the natural language, thereby the processing to natural language is imperfect, also system not.The inventor is in " body logic theory and application " book, the logical theory and the design philosophy of machine instruction have been proposed people's natural language is converted to, but because the technology that is confined to is at that time understood, also only considered the section processes method relevant, and systemic and complete implementation is not provided with existing natural language processing scheme.
Summary of the invention
The technical problem to be solved in the present invention is, existing problem at natural language processing set forth above and in understanding provides a kind of system and method for new ensuring computer to understand natural languages.Realize that this system and method will make computing machine can mechanically understand people's natural language, and make that natural language can become a kind of machine can Direct Recognition and the programming language of execution.System and method of the present invention is a kind of fully based on body and pure object-oriented systems and method.
The scheme that the present invention solves the problems of the technologies described above is, a kind of system of ensuring computer to understand natural languages is provided, and comprising:
Database is used to store the class of language material resource (as being the class of object with independent morpheme) and various application programs.Class in the described database is identical with class in the existing object based programming method.Object in the described database has been expanded the object range in the existing object based programming method, and Object oriented programming of the present invention is a kind of fully based on body and pure Object oriented programming system and method.Object wherein comprises arbitrary type and the subclass in arbitrary type and all the relevant data and function modules in the database.The data mode of promptly representing arbitrary objective things all is the object in the described database, and is this to liking body object of the present invention.Body object in the described database all is the body function, and the body function is to be the function module or the code segment of arbitrary or a certain specific matters or incident in the expression real world of unit with the grammar concept in the natural language;
Associated instruction set is used for the locial join character/word of input natural language is converted to machine instruction, and support will be imported the natural language character field and be converted into the machine program language automatically.The instruction that wherein is used for logical operation comprises: logic NOT operational order, logically equivalent operational order, logic and operation instruction, logic are with exclusive disjunction instruction, the instruction of logical difference exclusive disjunction, the abundant subsume instruction of logic, the necessary subsume instruction of logic; The instruction that is used for wherein that program connects and shifts control comprises: connect the assignment of subject and predicate and call and the sequential transfer instruction of link order, connection logical and relation, connect adequate condition transfer instruction that logic fully contains relation, connect the necessary necessary condition transfer instruction that contains relation of logic, connect logic with or the same or selection instruction of relation, the XOR that connection logic XOR concerns select an instruction.
The character string creating unit is used for generating input of character string according to the natural language of input;
The subject and predicate recognition unit is used for discerning the subject and the predicate character field of described input of character string, and obtains respectively and described subject and the subject of predicate character field coupling and the code segment of predicate object by searching database;
The keyword recognition unit is used for discerning the locial join character/word of described input of character string presentation logic connection relation.For the ease of identification, system will represent that the connection character/word of various logic connection relation is defined as crucial character/word.And the logical operation and the program control instruction of from instruction set, obtaining correspondence according to this key character/word.Especially, also be used for discerning the crucial character/word of subject and predicate field, crucial character/word such as " non-", " no ", " nothing ", " not having " are arranged in subject and/or the predicate field, before field, add 0, otherwise add 1.
Logical processing unit, be used to use described logical operation and program control instruction, with described subject and predicate and carry out locial join by the statement sequence that subject and predicate are formed and handle---the locial join character/word of importing in the natural language character field is converted to logical operation or program control instruction, and will imports the subject of natural language and predicate character field and the statement sequence be made up of subject and predicate character field is combined as complete program file.
Especially, also be used for discerning with each logical block (subject and predicate character field) of handler file with expression certainly and/or negative semantic nature 0 and/or 1) logical symbol; By reading or working procedure, understand 0 and/or 1 logical symbol of the expression semantic nature that runs through wherein.
In described subject and predicate recognition unit, described subject character field or predicate character field can comprise one or more attributes, the adverbial modifier and/or complement respectively, comprise attribute object, adverbial modifier's object and complement object in the described database.Therefore, the described system of understanding natural language also comprises:
The parameter recognition unit is used for discerning attribute, the adverbial modifier and/or the complement of described subject and predicate character field;
Parameter processing unit, be used for obtaining attribute, the adverbial modifier and/or complement object with described attribute, the adverbial modifier and/or complement coupling, and to use described attribute, the adverbial modifier and/or complement be corresponding parameter assignment in described subject and/or the predicate object by searching database.
In the system of ensuring computer to understand natural languages of the present invention, described logic instruction comprises that logically equivalent operational order, logic NOT operational order, logic and operation instruction, logic are with exclusive disjunction instruction, the instruction of logical difference exclusive disjunction, the abundant subsume instruction of logic, the necessary subsume instruction of logic, the negative keyword/speech in the corresponding natural language of wherein said described logic NOT operational order; The subject in the corresponding natural language of logically equivalent operational order and the junction symbol and the equivalent crucial character/word of predicate; Described logic and operation instructs the arranged side by side crucial character/word in the corresponding natural language; Described logic is instructed the crucial character/word of selection in the corresponding natural language with exclusive disjunction; Described logical difference exclusive disjunction instructs the crucial character/word of selecting in the corresponding natural language; The crucial character/word of the adequate condition in the corresponding natural language is instructed in the abundant subsume of described logic; The crucial character/word of necessary condition in the corresponding natural language is instructed in the necessary subsume of described logic.
In the system of ensuring computer to understand natural languages of the present invention, described program control instruction is meant program connection, selection, the transfer instruction corresponding with logic instruction, wherein corresponding assignment and the call instruction that connects subject and predicate of logically equivalent operational order; Logic and operation instructs the sequential transfer between the corresponding atomic sentence to instruct; Logic is instructed selection instruction between the corresponding atomic sentence with exclusive disjunction; The logical difference exclusive disjunction instructs and selects an instruction between the corresponding atomic sentence; The adequate condition transfer instruction between the corresponding atomic sentence is instructed in the abundant subsume of logic; The necessary condition transfer instruction between the corresponding atomic sentence is instructed in the necessary subsume of logic.
Comprise in the input of character string of the present invention: be used to separate first separator of subject and predicate character field, second separator that is used to separate two atomic sentences, the 3rd separator that is used to separate two molecusar statements, the 4th separator that is used to separate part statement before and after the reasoning and the 5th separator that is used to separate the paragraph statement, wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator increases progressively successively.
The present invention also provides a kind of method of ensuring computer to understand natural languages, and at understanding in the step of described method, the natural language character field of being imported is converted into the program file that computing machine is carried out, and said method comprising the steps of:
(a) natural language according to input generates input of character string;
(b) identification in the described input of character string subject and predicate character field and obtain respectively and described subject and the subject of predicate character field coupling and the code segment of predicate object by searching database, the code segment of described subject and predicate object is the code segment that computing machine is carried out, comprise the parameter of one or more attributes, the adverbial modifier and/or complement form in the code segment of each subject and/or predicate object, and to use described attribute, the adverbial modifier and/or complement object be that described subject carries out assignment with the corresponding parameter in the predicate object or is connected (parameter of by reference need connect);
(c) identification in the described input of character string subject and the predicate character field in be used to represent to negate semantic crucial character/word, crucial character/word such as " non-", " no ", " nothing ", " not having " are arranged in subject and/or the predicate field, before field, add 0, otherwise add 1;
(d) the crucial character/word of presentation logic connection relation in the described input natural language of identification, and the logical operation and/or the program control instruction of from instruction set, obtaining correspondence according to this key character/word;
(e) use described logical operation or program control instruction will import the subject of natural language and predicate character field and the statement sequence be made up of subject and predicate character field is combined as complete program file.
Above-mentioned steps is by searching database, computer understanding the object of subject and predicate notion indication; By being used to represent in identification subject and the predicate character field negates semantic crucial character/word, computer understanding the logical semantics of subject and predicate notion; By being used for the crucial character/word of locial join in the input of character string, computer understanding the logical relation between subject and predicate notion and each statement of being formed by subject and predicate notion.
Crucial character/word in the described step (c) comprises: the logic NOT in presentation logic computing and the programmed control, logical and, logic with or, logic XOR, logic fully contain, logic necessaryly contains, logical connective and reading and writing, repetition, the deposit data of logically equivalent, fetch data, the keyword of multiple functional instructions such as output.Negative character/word in the corresponding natural language of the crucial character/word of wherein said logic NOT computing, for example non-, or not do not have, do not wait; Connection character/word arranged side by side in the corresponding natural language of the crucial character/word of logic and operation program, for example and, again, also and and etc.; Logic connects character/word with the selection in the corresponding natural language of crucial character/word of exclusive disjunction program, for example or, may wait; Selecting in the corresponding natural language of the crucial character/word of logical difference exclusive disjunction program one connects character/word, for example or; Adequate condition in the corresponding natural language of the crucial character/word of the abundant subsume program of logic connects character/word, if for example ... so, because ... so so, must wait; Necessary condition in the corresponding natural language of the crucial character/word of the necessary subsume program of logic connects character/word, for example has only ... ability, ability etc.; Junction symbol and the equivalence in the natural language between corresponding subject of the crucial character/word of logically equivalent operation program and the predicate connect character/word, for example _ ,=,, equal etc.The pairing instruction of described crucial character/word, comprise the pairing functional instructions of mandatory character in the natural language, for example the reading and writing deposit data, fetch data, pairing instruction code and the pairing program control instruction of logical operation form such as output, wherein corresponding assignment and the call instruction that connects subject and predicate of logically equivalent operation program; The corresponding sequential transfer instruction of logic and operation program; Logic is instructed with exclusive disjunction program corresponding selection; Logical difference exclusive disjunction program correspondence is selected an instruction; The corresponding adequate condition transfer instruction of the abundant subsume program of logic; The corresponding necessary condition transfer instruction of the necessary subsume program of logic.Be computing machine carry out connection, selection and/or a transfer instruction synchronously corresponding logical operation program of realization (wherein code segment with 0 and 1 truth symbol of expression semantic nature can also be the drive signal of described program corresponding to the level signal in the digital logic circuit design).
In the method for ensuring computer to understand natural languages of the present invention, comprise a plurality of separators in the natural language character string of described input with different priorities, described a plurality of separators are used for the execution sequence of the code segment of definite program file.The described separator that is used for definite program file order is, be used to separate subject and predicate character field first separator, be used to separate two atomic sentences second separator, be used to separate two molecusar statements the 3rd separator, be used to separate the 4th separator between the part statement before and after the reasoning and be used to separate the 5th separator between the paragraph statement, wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator is successively decreased successively.
The system and method for ensuring computer to understand natural languages of the present invention by the locial join character/word that the subject in the identification natural language, predicate and expression natural language connect, has been realized the complete identification and the understanding of natural language.Method of the present invention can be widely used in the every field of computing machine, Logic Circuit Design in automated programming, mechanical translation, text-processing, artificial intelligence and the hardware or the like for example, make the user who does not possess computer major technical ability also can realize programming, perhaps make the user who does not have relevant foreign language technical ability can realize the translation of relational language.Especially, because system of the present invention, be to control by 0 and 1 semantic form, this is that the semantic control system of form will make the logical circuit in the computer hardware and the structure of device thereof become simple and directly perceived with 0 and 1.
Description of drawings
The invention will be further described below in conjunction with drawings and Examples, in the accompanying drawing:
Fig. 1 is the system embodiment synoptic diagram of ensuring computer to understand natural languages of the present invention;
Fig. 2 is the method embodiment process flow diagram of ensuring computer to understand natural languages of the present invention.
Embodiment
The present invention is on existing natural language independence morpheme base of recognition, increased the identification of the crucial character/word that subject in the natural language and predicate and presentation logic connect, realized the identification and the understanding of the logical relation between each morpheme in the natural language, thereby made computing machine for the identification of natural language complete sum system more.
Shown in the accompanying drawing 1, be the embodiment synoptic diagram of ensuring computer to understand natural languages of the present invention system.In the present embodiment, this system comprises database 11, associated instruction set 12, character string creating unit 13, subject-predicate recognition unit 14, keyword recognition unit 15 and Logical processing unit 16.
Character string creating unit 13 is used for generating input of character string according to the natural language of input, and this input of character string comprises literal and separates the separator of literal.Separator in the input of character string of above-mentioned generation comprises: be used to separate subject and predicate character field first separator (for example " _ "), be used to separate two atomic sentences (this atom language inclusion subject default statement) second separator (for example "; "), be used to separate two molecusar statements (this molecusar statement is meant the sentence with full sense, i.e. the statement that must be made up of two atomic sentences) the 3rd separator (for example ".; "), be used to separate the 4th separator (for example ". ") between the part statement before and after the reasoning and be used to separate the 5th separator between the paragraph statement (for example ".”)。Wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator is successively decreased successively.The 3rd above-mentioned separator can be nested, and other separators then can not be nested.
Subject and predicate recognition unit 14 is used for discerning the subject of input of character string and predicate character field and obtains respectively and subject and the subject of predicate character field coupling and the function code section of predicate object by searching database 11.
In the present embodiment, the subject object for the some or a certain class specific matters of expression in the objective world (for example, Zhang San, people etc.) code segment, and wherein can comprise one or more parameters that are used to describe this specific matters and/or functions (for example, (tall) Zhang San).In the same manner, the predicate object for the code segment of the attribute of some or a certain class specific matters and/or behavior in the expression objective world (for example, be the man, driving etc.), and wherein can comprise one or more parameter and/or functions (for example, be (black) man, drive) that are used to describe this particular community and/or behavior at (on 107 highways).
In the present embodiment, keyword recognition 15 is at first discerned the crucial character/word that expression is negated in subject and the predicate field, crucial character/word such as " non-", " no ", " nothing ", " not having " are arranged in subject and/or the predicate field, before field, add 0, otherwise add 1, by the crucial character/word in knowledge subject and the predicate field each subject and predicate field are all had and represent certainly and/or the logical symbol of negative semantic nature.
Above-mentioned logical operation program is the operation program based on the logical semantics symbol in subject and the predicate field.Particularly, if with
Presentation logic inverse symbol then has
If with " " presentation logic Equivalence Operation symbol, then has 0_0=1,0_1=0,1_0=0,1_1=1; If with “ ﹠amp; " presentation logic AND operator number, then have 0﹠amp; 0=0,0﹠amp; 1=0,1﹠amp; 0=0,1﹠amp; 1=1; If with the exclusive disjunction symbol, then have 0+0=0,0+1=1,1+0=1,1+1=1 with "+" presentation logic; If with
Presentation logic XOR symbol then has
If with the abundant subsume symbol of " → " presentation logic, then has 0 → 0=1,0 → 1=1,1 → 0=0,1 → 1=1; If, then have 0 ← 0=1,0 ← 1=0,1 ← 0=1,1 ← 1=1 with the necessary subsume symbol of " ← " presentation logic.
Logic control instruction described in the embodiment is used to control connection, selection and transfer instruction between the input natural language statement.Described logic control instruction and logical operation program have corresponding relation.Wherein logically equivalent operation program correspondence connects the assignment and the call instruction of subject and predicate; The corresponding sequential transfer instruction of logic and operation program; Logic is instructed with exclusive disjunction program corresponding selection; Logical difference exclusive disjunction program correspondence is selected an instruction; The corresponding adequate condition transfer instruction of the abundant subsume program of logic; The corresponding necessary condition transfer instruction of the necessary subsume program of logic.
Especially, computing machine has comprised the logical operation corresponding with it in the process of connection, selection and the transfer instruction of actuating logic control, the level signal of the truth symbol of resultant expression semantic nature in can corresponding control program in the computing.It makes that rreturn value type in the program process all is 0 or 1 the semantic type that is used for logical operation.
There is modification property grammer lexical item in the natural language, for example attribute, the adverbial modifier and/or complement etc.Correspondingly, in the subject character field or predicate character field that character string is converged, include one or more attributes, the adverbial modifier and/or complement respectively, attribute corresponding with it, complement and/or adverbial modifier's object are arranged in the database.Correspondingly, should increase a parameter recognition unit and parameter processing unit (not shown) in the system.Wherein the parameter recognition unit is used for discerning attribute, the adverbial modifier and/or the complement of subject character field and/or predicate character field, and parameter processing unit is used for obtaining and the attribute of attribute, the adverbial modifier and/or complement coupling, the adverbial modifier/or the code of complement object and to use the code of attribute, the adverbial modifier and/or complement object be the corresponding parameter assignment of described subject and/or predicate object by searching database 11.
Shown in the accompanying drawing 2, be the method embodiment schematic flow sheet of ensuring computer to understand natural languages of the present invention.May further comprise the steps:
Step S21: the natural language according to input generates input of character string.
Comprise a plurality of separators with different priorities in the character string of above-mentioned generation, these a plurality of separators are used for the execution sequence of the code segment of definite program file.Above-mentioned separator comprises first separator, second separator that is used to separate two atomic sentences, the 3rd separator that is used to separate two molecusar statements that are used to separate subject and predicate character field, be used to separate the 4th separator of part statement before and after the reasoning and the 5th separator that is used to separate the paragraph statement, and wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator is successively decreased successively.
Step S22: discern subject in the described input of character string and predicate character field and obtain respectively and described subject and the subject of predicate character field coupling and the code segment of predicate object, comprise the subject in the natural language of being imported and the code segment (body function) of predicate object in the database by searching database.The code segment of subject and predicate object is code segment or the modular program that computing machine is carried out, and comprises one or more parameters in the code segment of each subject/predicate object.
Step S23: logical operation and/or the program control instruction discerning the crucial character/word in the described input of character string and from instruction set, obtain correspondence according to this key character/word, described logical operation and/or program control instruction, comprise logically equivalent computing and/or assignment and call instruction, logic and operation and/or sequential transfer instruction, logic is with exclusive disjunction and/or selection instruction, logical difference exclusive disjunction and/or select an instruction, abundant subsume of logic and/or adequate condition transfer instruction, necessary subsume of logic and/or necessary condition transfer instruction, subject in the corresponding natural language of wherein said logically equivalent computing and assignment and the junction symbol of predicate and equivalent crucial character/word with call instruction, described logic and operation and sequential transfer are instructed the arranged side by side crucial character/word in the corresponding natural language, described logic is with the crucial character/word of selection in exclusive disjunction and the corresponding natural language of selection instruction, described logical difference exclusive disjunction and the crucial character/word of selecting in the corresponding natural language of an instruction of selecting, the crucial character/word of adequate condition in the corresponding natural language of abundant subsume program of described logic and adequate condition transfer instruction, the crucial character/word of necessary condition in necessary subsume of described logic and the corresponding natural language of necessary condition transfer instruction.
Step S24: use described logical operation and program control instruction be combined as complete program file corresponding to the code segment of subject and predicate object.Program file rreturn value type in the process of implementation is 0 or 1 the semantic type that is used for logical operation.
The aforementioned calculation machine is handled in the method for natural language, include one or more attributes respectively in subject character field and/or the predicate character field, the adverbial modifier and/or complement, comprise corresponding attribute in the described database, the adverbial modifier and/or complement object, comprise the attribute in the described subject character field of identification among the step S22, the adverbial modifier and/or complement, and obtain and described attribute by searching database, the attribute of the adverbial modifier and/or complement coupling, the adverbial modifier and/or complement object also use described attribute, the code of the adverbial modifier and/or complement object or function are the corresponding parameter assignment in described subject object and/or the predicate object.
The above only be the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto.Anyly be familiar with those skilled in the art in the technical scope that the present invention discloses, the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claim.
Claims (11)
1. the system of a Computer Processing natural language is characterized in that, comprising:
Database is used to store the class of language material resource (as being the class of object with independent morpheme) and various application programs.Class in the described database is identical with class in the existing object based programming method.Object in the described database has been expanded the object range in the existing object based programming method, and Object oriented programming of the present invention is a kind of fully based on body and pure OO automatic programming system and method.Object wherein comprises arbitrary type and the subclass in arbitrary type and all the relevant data and function modules in the database.The data mode of promptly representing arbitrary objective things all is the object in the described database, and is this to liking body object of the present invention.Body object in the described database all is the body function, and the body function is to be the function module or the code segment of arbitrary or a certain specific matters or incident in the expression real world of unit with the grammar concept in the natural language;
Associated instruction set is used to realize the logical operation and the programmed control of natural language.The instruction that wherein is used for logical operation comprises that logic NOT operational order, logically equivalent operational order, logic and operation instruction, logic are with exclusive disjunction instruction, the instruction of logical difference exclusive disjunction, the abundant subsume instruction of logic, the necessary subsume instruction of logic; Being used for program connects, selects to comprise the assignment that is connected subject and predicate with the instruction of shifting control and calls (returning) instruction, sequential transfer instruction, selection instruction, selects an instruction, adequate condition transfer instruction, necessary condition transfer instruction;
The character string creating unit is used for generating input of character string according to the natural language of input;
The subject and predicate recognition unit, be used for discerning the subject of described input of character string and predicate character field and obtain subject and predicate object with described subject and predicate character field coupling respectively by searching database, wherein each subject/predicate object is the code segment that computing machine is carried out, and comprise one or more parameters in each master/meaning object, each logical operation program is the code segment that computing machine is carried out;
The keyword recognition unit is used for discerning the crucial character/word of described input of character string and obtains corresponding logical operation or program control instruction according to this key character/word from instruction set.Especially, be used for subject and predicate field are carried out code identification, crucial character/word such as " non-", " no ", " nothing ", " not having " are arranged in subject and/or the predicate field, before field, add 0, otherwise add 1;
Logical processing unit, being used to use described logical operation and program control instruction is complete program file with the code combination of described subject and predicate object.
2. the system of Computer Processing natural language according to claim 1, it is characterized in that, include one or more attributes, the adverbial modifier and/or complement respectively in described subject character field or the predicate character field, comprise attribute, the adverbial modifier and/or complement object in the described database, also comprise:
The parameter recognition unit is used for discerning attribute, the adverbial modifier and/or the complement of described subject character field;
Parameter processing unit, being used for obtaining attribute, the adverbial modifier and/or the complement object that mates with described attribute, the adverbial modifier and/or complement and using described attribute, the adverbial modifier and/or complement object by searching database is the corresponding parameter assignment of described subject and/or predicate object.
3. the system of Computer Processing natural language according to claim 1, it is characterized in that, described operation program comprises the logically equivalent operation program, the logic NOT operation program, the logic and operation program, logic is with the exclusive disjunction program, logical difference exclusive disjunction program, the abundant subsume program of logic, the necessary subsume program of logic, the subject in the corresponding natural language of wherein said logically equivalent operation program and the connector and the equivalent crucial character/word of predicate, negative keyword/speech in the corresponding natural language of described logic NOT operation program, arranged side by side crucial character/word in the corresponding natural language of described logic and operation program, described logic is with the crucial character/word of selection in the corresponding natural language of exclusive disjunction program, select a crucial character/word in the corresponding natural language of described logical difference exclusive disjunction program, the crucial character/word of adequate condition in the corresponding natural language of the abundant subsume program of described logic, the crucial character/word of necessary condition in the corresponding natural language of the necessary subsume program of described logic.
4. the system of Computer Processing natural language according to claim 1, it is characterized in that the program control instruction in the described instruction set comprises equivalent program control instruction, sequential programme steering order, option program steering order, selects a program control instruction, adequate condition program control instruction, necessary condition program control instruction.Wherein logically equivalent operation program correspondence connects the assignment and the call instruction of subject and predicate; The corresponding sequential transfer instruction of logic and operation program; Logic with or and logical difference exclusive disjunction program corresponding selection steering order; Logic fully contains and the necessary subsume program of logic respective conditions transfer instruction.Computing machine has comprised logical operation corresponding with it in the process of connection, selection and the transfer instruction of actuating logic control.It makes and all to have i.e. 0 and 1 the logical symbol of expression affirmation and negation semantic nature in the program file and each code segment in the program process that is connect by instruction.Described 0 and 1 logical symbol is corresponding to the high and low level in the driven by program.
5. the system of Computer Processing natural language according to claim 1, it is characterized in that, comprise in the described character field: be used to separate subject and predicate character field first separator, be used to separate two atomic sentences second separator, be used to separate two molecusar statements the 3rd separator, be used to separate the 4th separator between the statement before and after the reasoning and be used to separate the 5th separator between the paragraph statement, wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator increases progressively successively.
6. the method for a Computer Processing natural language is characterized in that, may further comprise the steps:
(a) natural language according to input generates input of character string;
(b) identification in the described input of character string subject and predicate character field and obtain subject and predicate object with described subject and predicate character field coupling respectively by searching database, comprise subject and predicate object in the wherein said database, each subject/predicate object is to comprise one or more parameters in the code segment carried out of computing machine and each object;
(c) logical operation and the program control instruction of discerning the crucial character/word in the described input of character string and obtaining correspondence according to this key character/word from database;
(d) use described logical operation and program control instruction that the code segment of described subject and predicate object is combined as complete program file.
7. the method for Computer Processing natural language according to claim 6, it is characterized in that, include one or more attributes respectively in described subject character field or the predicate character field, the adverbial modifier and/or complement, comprise attribute in the described database, the adverbial modifier and/or complement object, comprise the attribute in the described subject character field of identification in the described step (a), the adverbial modifier and/or complement, and obtain and described attribute by searching database, the attribute of the adverbial modifier and/or complement coupling, the adverbial modifier and/or complement object also use described attribute, the adverbial modifier and/or complement object are the corresponding parameter assignment in described subject and/or the predicate object.
8. according to the method for claim 6 or 7 described Computer Processing natural languages, it is characterized in that, logical operation program in the described database comprises the logically equivalent operation program, the logic NOT operation program, the logic and operation program, logic is with the exclusive disjunction program, logical difference exclusive disjunction program, the abundant subsume program of logic, the necessary subsume program of logic, subject in the corresponding natural language of wherein said logically equivalent operation program is connected with predicate and/or equivalent crucial character/word, negative keyword/speech in the corresponding natural language of described logic NOT operation program, arranged side by side crucial character/word in the corresponding natural language of described logic and operation program, described logic is with the crucial character/word of selection in the corresponding natural language of exclusive disjunction program, select a crucial character/word in the corresponding natural language of described logical difference exclusive disjunction program, the crucial character/word of adequate condition in the corresponding natural language of the abundant subsume program of described logic, the crucial character/word of necessary condition in the corresponding natural language of the necessary subsume program of described logic.Wherein logically equivalent operation program correspondence connects the assignment and the call instruction of subject and predicate; The corresponding sequential transfer instruction of logic and operation program; Logic with or and logical difference exclusive disjunction program corresponding selection steering order; Logic fully contains and the necessary subsume program of logic respective conditions transfer instruction.Comprised logical operation corresponding in the described logic control instruction with it.
9. the method for Computer Processing natural language according to claim 6 is characterized in that, comprises a plurality of separators with different priorities in the described character string, and described a plurality of separators are used for determining the execution sequence of program file code.
10. the method for Computer Processing natural language according to claim 9, it is characterized in that, be used to separate subject and predicate character field first separator, be used to separate two atomic sentences second separator, be used to separate two molecusar statements the 3rd separator, be used to separate the 4th separator between the statement before and after the reasoning and be used to separate the 5th separator between the paragraph statement, wherein the priority of first separator, second separator, the 3rd separator, the 4th separator, the 5th separator is successively decreased successively.
11. the method for Computer Processing natural language according to claim 10 is characterized in that, comprises first separator, second separator, the 3rd separator, the 4th separator, the 5th separator in the natural language of described input.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101079753A CN101930428A (en) | 2009-06-18 | 2009-06-18 | System and method for ensuring computer to understand natural languages |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101079753A CN101930428A (en) | 2009-06-18 | 2009-06-18 | System and method for ensuring computer to understand natural languages |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101930428A true CN101930428A (en) | 2010-12-29 |
Family
ID=43369610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009101079753A Pending CN101930428A (en) | 2009-06-18 | 2009-06-18 | System and method for ensuring computer to understand natural languages |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101930428A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103123620A (en) * | 2012-12-11 | 2013-05-29 | 中国互联网新闻中心 | Web text sentiment analysis method based on propositional logic |
| CN103226606A (en) * | 2013-04-28 | 2013-07-31 | 浙江核新同花顺网络信息股份有限公司 | Inquiry selection method and system |
| CN103294470A (en) * | 2013-05-27 | 2013-09-11 | 上海华兴数字科技有限公司 | Programming language system with built-in natural language |
| CN103617159A (en) * | 2012-12-07 | 2014-03-05 | 万继华 | Method for translating natural languages into computer language, semantic analyzer and human-machine conversation system |
| CN104020942A (en) * | 2013-03-03 | 2014-09-03 | 上海能感物联网有限公司 | Method for calling computer program to operate by Chinese text |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106354036A (en) * | 2016-08-30 | 2017-01-25 | 广东美的制冷设备有限公司 | Household appliance control method and device |
| CN106649212A (en) * | 2016-09-12 | 2017-05-10 | 济南浪潮高新科技投资发展有限公司 | Business data processing method and device |
| WO2017084185A1 (en) * | 2015-11-19 | 2017-05-26 | 深圳创想未来机器人有限公司 | Intelligent terminal control method and system based on semantic analysis, and intelligent terminal |
| CN110287487A (en) * | 2019-06-17 | 2019-09-27 | 北京百度网讯科技有限公司 | The recognition methods of subject-predicate language, device, equipment and computer readable storage medium |
| CN112328780A (en) * | 2020-11-13 | 2021-02-05 | 北京明略软件系统有限公司 | Natural language conversion processing method and device, electronic equipment and storage medium |
| WO2023206267A1 (en) * | 2022-04-28 | 2023-11-02 | 西门子股份公司 | Method and apparatus for adjusting natural language statement, and storage medium |
-
2009
- 2009-06-18 CN CN2009101079753A patent/CN101930428A/en active Pending
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617159A (en) * | 2012-12-07 | 2014-03-05 | 万继华 | Method for translating natural languages into computer language, semantic analyzer and human-machine conversation system |
| WO2014086101A1 (en) * | 2012-12-07 | 2014-06-12 | Wan Jihua | Method for translating natural language into computer language, semantic analyzer and human-computer dialogue system |
| CN103617159B (en) * | 2012-12-07 | 2016-10-12 | 万继华 | Natural language translation is become the method for computer language, semantic analyzer and interactive system |
| CN103123620A (en) * | 2012-12-11 | 2013-05-29 | 中国互联网新闻中心 | Web text sentiment analysis method based on propositional logic |
| CN104020942A (en) * | 2013-03-03 | 2014-09-03 | 上海能感物联网有限公司 | Method for calling computer program to operate by Chinese text |
| US10922371B2 (en) | 2013-04-28 | 2021-02-16 | Hithink Royalflush Information Network Co., Ltd. | Query selection method and system |
| CN103226606A (en) * | 2013-04-28 | 2013-07-31 | 浙江核新同花顺网络信息股份有限公司 | Inquiry selection method and system |
| US11714861B2 (en) | 2013-04-28 | 2023-08-01 | Hithink Royalflush Information Network Co., Ltd. | Query selection method and system |
| US10185772B2 (en) | 2013-04-28 | 2019-01-22 | Hithink Royalflush Information Network Co., Ltd. | Query selection method and system |
| CN103294470A (en) * | 2013-05-27 | 2013-09-11 | 上海华兴数字科技有限公司 | Programming language system with built-in natural language |
| WO2017084185A1 (en) * | 2015-11-19 | 2017-05-26 | 深圳创想未来机器人有限公司 | Intelligent terminal control method and system based on semantic analysis, and intelligent terminal |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106354036A (en) * | 2016-08-30 | 2017-01-25 | 广东美的制冷设备有限公司 | Household appliance control method and device |
| CN106354036B (en) * | 2016-08-30 | 2019-04-30 | 广东美的制冷设备有限公司 | Household electric appliance control method and device |
| CN106649212A (en) * | 2016-09-12 | 2017-05-10 | 济南浪潮高新科技投资发展有限公司 | Business data processing method and device |
| CN110287487A (en) * | 2019-06-17 | 2019-09-27 | 北京百度网讯科技有限公司 | The recognition methods of subject-predicate language, device, equipment and computer readable storage medium |
| CN110287487B (en) * | 2019-06-17 | 2023-08-11 | 北京百度网讯科技有限公司 | Master predicate identification method, apparatus, device, and computer-readable storage medium |
| CN112328780A (en) * | 2020-11-13 | 2021-02-05 | 北京明略软件系统有限公司 | Natural language conversion processing method and device, electronic equipment and storage medium |
| WO2023206267A1 (en) * | 2022-04-28 | 2023-11-02 | 西门子股份公司 | Method and apparatus for adjusting natural language statement, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101930428A (en) | System and method for ensuring computer to understand natural languages | |
| Deursen et al. | Language Prototyping: an algebraic specification approach | |
| CN101840334B (en) | Software component service packaging method | |
| Engelen et al. | Integrating textual and graphical modelling languages | |
| CN101763212B (en) | Man-machine interactive system and relevant system, device and method thereof | |
| CN102708161B (en) | Modeling method of data logic model utilizing public conceptual sets | |
| CN103123590A (en) | Compiling method from intermediate language (IL) program to C language program of instruction list | |
| CN103617159A (en) | Method for translating natural languages into computer language, semantic analyzer and human-machine conversation system | |
| KR20110062084A (en) | Hybrid translation apparatus and its method | |
| CN101185116A (en) | Using strong data types to express speech recognition grammars in software programs | |
| EP2021918A1 (en) | A generic interface for deep embedding of expression trees in programming languages | |
| CN103440234A (en) | Natural language understanding system and method | |
| CN106951305A (en) | It is a kind of based on method of the model conversion by QVT R language generation midCore scripts | |
| CN103902677A (en) | Cross-platform database access method | |
| Abdelnabi et al. | Generating uml class diagram from natural language requirements: A survey of approaches and techniques | |
| CN111209211A (en) | Cross-project software defect prediction method based on long-term and short-term memory neural network | |
| CN108153522B (en) | Method for generating Spark and Hadoop program codes by midcore based on model conversion | |
| CN114528846A (en) | Concept network for artificial intelligence and generation method thereof | |
| Ballance et al. | Grammatical abstraction and incremental syntax analysis in a language-based editor | |
| CN101968785A (en) | Logical machine model capable of understanding human natural language | |
| Kumar et al. | Deep learning driven natural languages text to sql query conversion: A survey | |
| CN110705316A (en) | Method and device for generating linear time sequence logic protocol of smart home | |
| Al-Azzoni et al. | A utility to transform CSV data into EMF | |
| CN114860869A (en) | Controllable universal dialogue model with generalized intentions | |
| Bajwa et al. | UCD-generator-A LESSA application for use case design |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20101229 |