EP3750051A1 - Procédé de transformation d'une séquence pour la rendre exécutable par une machine - Google Patents
Procédé de transformation d'une séquence pour la rendre exécutable par une machineInfo
- Publication number
- EP3750051A1 EP3750051A1 EP19710040.7A EP19710040A EP3750051A1 EP 3750051 A1 EP3750051 A1 EP 3750051A1 EP 19710040 A EP19710040 A EP 19710040A EP 3750051 A1 EP3750051 A1 EP 3750051A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- class
- function
- upstream
- sequence
- classes
- 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.)
- Ceased
Links
- 238000000034 method Methods 0.000 title claims abstract description 125
- 230000001131 transforming effect Effects 0.000 title claims abstract description 6
- 230000006870 function Effects 0.000 claims abstract description 592
- 238000011144 upstream manufacturing Methods 0.000 claims abstract description 224
- 230000014509 gene expression Effects 0.000 claims abstract description 78
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 58
- 241000167880 Hirundinidae Species 0.000 claims description 18
- 238000003780 insertion Methods 0.000 claims description 6
- 230000037431 insertion Effects 0.000 claims description 6
- 239000003550 marker Substances 0.000 claims description 2
- 239000010755 BS 2869 Class G Substances 0.000 description 23
- 238000011156 evaluation Methods 0.000 description 6
- 239000010749 BS 2869 Class C1 Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 238000010276 construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000011282 treatment Methods 0.000 description 3
- 239000010750 BS 2869 Class C2 Substances 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004883 computer application Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000003362 replicative effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0426—Programming the control sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2452—Query translation
- G06F16/24522—Translation of natural language queries to structured queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/282—Hierarchical databases, e.g. IMS, LDAP data stores or Lotus Notes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/23—Pc programming
- G05B2219/23272—Natural language, use simple words like move, rotate
Definitions
- the present invention relates to a method of transforming a sequence comprising several words from a natural language to make said sequence executable in real time to control a machine.
- Some known methods use classes but are very limited in interpreting a natural language given the complexity of the language and the differences from one language to another.
- Some documents of the prior art also provide methods for interpreting a natural language, for example document US6243670.
- This document describes a process that recognizes a list of type objects characteristic of languages, verbs, adverbs, nouns, pronouns ... This process is also limited because the number of languages and differences in structure from one language to another thus render the process described non-universal.
- the described method also requires a pre-knowledge of a dictionary of the words of the language in question and their semantic status.
- WO 00/23889 which describes a method for compiling a source code comprising declarations in natural language
- US2017239576 which describes a method for automatically generating a program from a natural language for specific computer applications to generate an application, such as a video game.
- the machine is a device or set of devices capable of performing some work or performing a certain function, either under the guidance of an operator, or in an autonomous manner.
- Machines are more and more complex. It becomes ineffective to use an array of buttons or levers. That is why it is necessary to find a new technical solution to improve the control of the machines by providing a recognition of the commands formulated in natural language and making it possible to control the machine.
- the present invention aims to remedy these disadvantages.
- the present invention is directed to a method implemented by computer for transforming a sequence comprising several words from a natural language to make said sequence executable in real time in order to control a machine, remarkable in that said method comprises the following steps:
- said class if said class also includes a class expression and a sense, then said class is called Generalized class, if the meaning is upstream said class is called upstream Generalized class, if the meaning is downstream, said class is called Generalized class down, if it is an upstream Generalized class then the class in the first position of its class expression is an upstream function class and the second class of its class expression is a class.parameter,
- a function class is at least one of the following: an upstream function class, and a downstream function class,
- a function class Dep is either a function class Dep up, or a function class Dep down
- said classes function Dep present in the table have the characteristic that any class generalized of one of said function classes Dep is known from the method as a function class of direction inverted to said function class Dep, called function class ParDeo, and
- an upstream ParDeo function class is an upstream function class and is a generalized class of a function class Dep avale, and
- a function class ParDeo downstream is a downstream function class and is a generalized class of an upstream function class Dep,
- a hierarchical relationship table is also defined for pairs consisting of an upstream function class or a function class Dep and a downstream function class or a function class Dep, said hierarchical report indicates for each of the pairs present in the table which of the two classes functions of the pair is hierarchically dominant over the other,
- a downstream function class is subsequent to an upstream function class or a hierarchically dominant Dep function class, and said function classes are separated by a non-empty parameter, then said function class swallows non-hierarchically dominant precedence in taking said parameter on the upstream function class or the function class Dep hierarchically dominant,
- said database also includes reserved classes, each reserved class includes a number and a simple expression, each simple expression of which is preceded by a marking,
- said database comprises the following predefined elements: first class reserved, second class reserved, third class reserved, fourth class reserved, fifth class reserved,
- the first reserved class, the second reserved class, the third reserved class and the fourth reserved class are hereinafter referred to as GI classes.
- the first three GI classes correspond to the first GI class, the second GI class and the third GI class.
- the last three GI classes correspond to the second class IG, third class IG and fourth class IG.
- the first reserved class and the fourth reserved class have the function of indicating in a sequence the beginning and the end of a generalized class.
- the second reserved class always positioned within a portion of the sequence delimited by the first reserved class and the fourth reserved class, separates said sequence part into two sub-parts, the first of which represents an upstream function class and the second represents a parameter class, and indicates that said sequence portion represents an upstream Generalized class.
- the third reserved class always positioned within a portion of the sequence delimited by the first reserved class and the fourth reserved class, separates said sequence part into two sub-parts, the first of which represents a parameter class and the second represents a downstream function class, and indicates that said sequence part represents a generalized class downstream,
- IG expression a sequence or part of a sequence comprising IG classes.
- the fifth reserved class is a function class and designates any class as executable by a machine.
- the preprocessing step includes a comparison of the sequence with the data of the database:
- dichotomy in which the simple expressions and IG expressions known in the process are searched for in the sequence as an upstream function class or a downstream function class, the sequence is dichotomized until all the function classes contained in the sequence likely to give rise to dichotomies has been
- the dichotomy step consists of:
- said upstream function class if it is an upstream function class, and said upstream function class if No Dep is at the beginning of the sequence or immediately preceded by one of the first three IG classes and said upstream function is not at the end of sequence and not immediately followed by one of the last three classes IG, then it is identified a part of the sequence playing the role of a parameter class for said upstream function class, said parameter class, contiguous to said upstream function class, starts at the position following said upstream function class and whose end position is identified by continuing on the positions of the sequence from the start position of the parameter class and is:
- step b) of dichotomy allows nested Generalized classes.
- sequence part is stored in the database with at least the following two information: a new number and its simple expression
- the resulting sequence is called the LIG expression of the sequence.
- the proposed method allows a system to learn as and when, and build its knowledge from the previous knowledge.
- the method allows:
- the method makes it possible to understand an unknown sequence in order to apply it to control a machine.
- Said method is light in resource, can be local (without internet connection) and carried by a computer.
- a computer is a programmable information processing system or device that operates by sequentially reading a set of instructions organized in programs.
- computers a digital tablet, a mobile phone, especially of the "smartphone" type, a connected watch, connected glasses, a remote control, a headset, a virtual reality headset, a connected television, a game console, or an internet box.
- characters are transmitted or collected by means of treatment without waiting for the end of the process treatment.
- characters are transmitted in writing or by the dictation of a voice transmitted in real time.
- the method is capable of driving a physical machine at a speed adapted to the evolution of the machine.
- Database is defined as physical memory The database is used to store and retrieve all raw data or information related to the character or characters of a sequence.
- IG classes are visible and easily readable by the user when inserted into a text (a sequence / a class). And this means for the user an extremely reduced learning.
- a method made extremely light by this method allowing it to operate with very limited resources, allowing it to be shipped directly into the machine without external means of communication necessary.
- this method will require only about thirty instances in the database.
- the update of the table of the functions Dep in step b) is determined according to the following conditions: if an upstream function class applies to a first upstream number corresponding to the recurrence number of said upstream function class of the upstream Generalized classes of the database, and if said upstream function class applies to a second upstream number corresponding to the number of recurrence of said generalized classes which are aval function classes, and of which said second upstream number is at least greater than half of the first number, then the upstream function class is an upstream function class Dep,
- downstream function class if a downstream function class applies to a first downstream number corresponding to the number of recurrences of said downstream function class of the generalized classes downstream of the database, and that said downstream function class applies to a second downstream number corresponding to the number of recurrence of said generalized classes which are upstream function classes, and of which said second downstream number is at least greater than half of the first number, then the downstream function class is a function class Dep downstream.
- the method comprises the following elements: if an upstream function class is earlier in a sequence than a downstream function class, the hierarchical relationship between the upstream and the upstream function class is evaluated the function class swallows as:
- said upstream function class applies to an upstream number corresponds to the number of recurrences of said upstream function class of the upstream generalized classes of the database in which the parameter classes are generalized downstream classes whose function is said downstream function class, and that said downstream function class applies to a downstream number corresponds to the recurrence number of said downstream function class of the downstream Generalized classes of the database in which the parameter classes are upstream generalized classes whose function is said upstream function class , and that: if the upstream number is greater than the downstream number then said upstream function class is a hierarchically dominant on said downstream function class, if the downstream number is greater than the upstream number then said downstream function class is a hierarchically dominant on said upstream function class,
- Said evaluation allows the method: either updating the table of hierarchical relations between the functions, or in step b) of dichotomy and if realized in real time, to replace the table of hierarchical relations, and thus of operate without prior knowledge of said hierarchical relationship table.
- step b) of dichotomy the method comprises the following elements:
- an upstream list number then being obtained by accumulating the upstream numbers for each of the pairs consisting of an upstream function class belonging to the list an upstream function class and a downstream function class belonging to the list of the downstream function class
- the downstream list number then being obtained by cumulating the downstream numbers for each of the pairs consisting of an upstream function class belonging to the list upstream function class and a downstream function class belonging to the downstream function class list
- said upstream function class list is a hierarchically dominant on said downstream function class list
- downstream function class list is hierarchically dominant on said upstream function class list, otherwise there is no hierarchical relationship between said upstream function class list and said function class list swallows, if an upstream function class is earlier in a sequence than a downstream function class and there is no hierarchical relationship directly between said upstream function class and said downstream function class and that one or both are class ParDep functions, then it is evaluated the hierarchical relationship of the list of function classes ParDep upstream of the same class function Dep down from said upstream function class if it is a function class ParDep and of the list of function classes ParDep downstream of the same function class Dep upstream of said downstream function class if it is a function class ParDep,
- an upstream function class if an upstream function class is earlier in a sequence than a function class Dep also upstream, it evaluates the hierarchical relationship of the upstream function class and the list of all the function classes ParDep downstream of the function class Dep upstream,
- step b) it is added if it is an upstream function the following option in determining the end position of the parameter of said upstream function:
- step b) it is added if it is a function downstream the following option in determining the starting position of the parameter of said downstream function:
- the application of this provision anticipates the appearance of a function ParDep posterior downstream in the sequence to an upstream function and authorizes the application of the hierarchy of said upstream function with the downstream Pardep functions of the upstream function Dep, and accordingly increases the ability of the process to predict how to be dichotomized thus including a new sequence.
- step c) if the generalized class is known, its class expression and its meaning are updated.
- the update will allow on the one hand to correct the understanding of a sequence by using the IG classes directly in the sequence, but also to apply the most recent hierarchies and thus to correct errors. understanding of previously learned sequences. The system will be able to evolve constantly.
- the second upstream number is at least greater than two-thirds of the first number and the second downstream number is at least greater than two-thirds of the first number.
- said step b) when searching for function classes in the sequence, further comprises that if a function class is found in the sequence and the IG expression of said function class contains other classes function then said function class is treated in priority on said other function classes.
- the database also comprises a sixth reserved class
- said sixth reserved class is a downstream function class and designates any function class Dep as an equivalence function since said function class Dep expects to be parameter in a generalized class whose function is the sixth reserved class, thus any parameter class of a generalized class whose function is the function class ParDep of a function class Dep and equivalence has for equivalent class the parameter class of the function class Dep in the class generalized function ParDep, in the execution step d), if a class is encountered in the sequence and a generalized class is known whose parameter is said class and the function is a function class ParDep of a function class Dep and d equivalence, then the LIG expression of said class is substituted in the sequence by the LIG expression of its equivalent class, the iteration taking over at the beginning of said class,
- the LIG expression of said equivalent class starts with the number of said class and is obtained by iteration by following all the numbers of the generalized classes of the LIG expression of said class by the number of the first class IG, then the number of the class the first class of the class expression of the generalized class, then the number of the second class IG if the class is a generalized class upstream or the number of the third class IG if the class is a generalized class swallows, then the number of the class the second class of the class expression of the generalized class, then the number of the fourth class IG.
- the set of these provisions makes it possible to transform any class into another class and in particular a sequence representing an order written in natural language into a class executable by a machine.
- step b) of dichotomy if there is no hierarchical relationship between two function classes, it is searched for each of the two class functions the list of classes related functions in the database then the recursion of each function class is compared to evaluate the hierarchy between these two lists, the list of classes related to a class being defined as the list of classes which are a parameter class in generalized classes and whose function classes are also apply to said class in a generalized class.
- This arrangement makes it possible to name and categorize classes sharing various characteristics and thus by naming a new class of the same category, to make it benefit from the hierarchies already learned for the other classes in the category.
- pretreatment step a) if the reserved class marker is followed by a number, then that expression is substituted in the sequence by the number of the class of the same number.
- FIG. 1 in the form of a logic diagram, the steps implemented in a particular embodiment of the method that is the subject of the present invention,
- FIGS. 2 to 5 in the form of a logic diagram, represent the details of the steps implemented in a particular embodiment of the method that is the subject of the present invention.
- FIG. 6 represents a diagram of reconstitution of a LIG sequence of a class.
- a sequence is defined as a series of elements put and processed one after the other.
- a sequence will be understood here as a sequence of numbers.
- An order such as ⁇ turn right ⁇ which is a sequence of characters is actually a sequence of numbers - encoded via the Ascii table (acronym for American Standard Code for Information Interchange in English terminology for US Standard Code for Information Exchange ) or Uni code (computer standard that allows exchanges of texts in different languages) and so ⁇ turn right ⁇ is a sequence.
- the essential function of the method is to recognize and process sequences.
- a database will keep the sequences processed by the method.
- the process will assign a number (No) to each of these sequences, this No is called No Class or more, simply "class"
- a class can also be written as other classes.
- a sequence can be written from already known sequences.
- ⁇ C1 C2 ⁇ being the sequence consisting of class C1 and class C2.
- classes can play a function or parameter role.
- CRO the first reserved class ⁇ [ ⁇ , CR1: the second class reserved ⁇ » ⁇ , CR2: the third reserved class ⁇ " ⁇ , CR3: the fourth class reserved ⁇ j ⁇ .
- the method recognizes in the sequence the sequences of the Reserved Words used and substitutes them by the numbers (Nos) of corresponding classes.
- the sequence will then be ⁇ CROturn CR1 leftCR3 ⁇ (step 0)
- the process identifies on the base (or will create if unknown) the classes ⁇ turn ⁇ and ⁇ right ⁇ is C1 and C2, ⁇ turn ⁇ being the sequence between CRO and CR1 and ⁇ right ⁇ being the sequence between CR1 and CR3 . Then the process identifies on the base (or will create if unknown) the class ⁇ turn right ⁇ either C3, ⁇ turn right ⁇ being the sequence of C1 + the sequence of C2.
- step 2 This step is ensured by the step identification-creation, (step 2) whose mission is to identify on the database, or to create if unknown, all the component classes delimited by the classes CRO to CR3 in the sequence.
- step 2 This step is ensured by the step of Dichotomy (step 1) whose mission is to insert in the sequence that it has to treat classes CRO to CR3 by relying on the known functions of the method. Then the process will follow on learning this sequence exactly as above (see step 2).
- Step 1 pretreatment
- step 1 dichotomy
- step 2 identification-creation.
- Classes C3 and C5 are designated as "Generalized Classes" or Classes G.
- the notion of generalization consists in replicating in the generality a behavior found in a case taken as an example, in this case the role of function that a class can have. .
- the G classes have in the database in addition to their simple expression, an expression in terms of class, always consisting of two classes, and a meaning. Since classes G always consist of a function and a parameter, the meaning will indicate where the function is in the expression in terms of class. If the sense is Am the function will be at the beginning and the parameter at the end and vice versa.
- a class G can therefore be written as ⁇ [Function class ⁇ »Parameter class ⁇ for a class G Amont Or as ⁇ [Parameter class ⁇ " Function class ⁇ ] ⁇ for a class G Swallow.
- IG Classes G indication classes
- function and parameter classes can themselves be classes G, it follows that the IG classes make it possible to establish the functional dichotomy of a sequence. This functional dichotomy reflects how the process understood the sequence.
- the method knows how to recognize certain functions that make it possible to connect two classes.
- Cx and Cy are any two classes acting as parameters. These two classes are connected in the sequence by the function F0.
- F0 is an Upstream function in the class G ⁇ [F0 ⁇ »Cy ⁇ j ⁇ that the class G ⁇ [F0 ⁇ » Cy ⁇ j ⁇ is itself a function, Swallow this time in our sequence (1)
- a class being a function in one direction and whose classes G using this class as a function are also functions but in the opposite direction is the characteristic of the so-called dependent functions or "Dep functions".
- the dichotomy step does not modify the sequence (no function is known).
- ⁇ turn left means ⁇ SETWHEEL-1 ⁇
- step of dichotomy recognizes in the sequence the function ⁇ means ⁇ (step 1/1) learned previously and will insert in this sequence the Reserved Classes CRO to CR3 corresponding to a situation of Function Amount Dep. Which will give: (steps 1/2 and 1/3)
- These executable classes G have a parameter (or parameter class).
- any class GDep ⁇ [Cx ⁇ " ⁇ [Fx ⁇ » Cy ⁇ ] ⁇ ] ⁇ or reciprocal such that ⁇ [Fx ⁇ " ⁇ EXE ⁇ ] ⁇ has been learned.
- These executable GDep classes have two parameters.
- the method has an "Execution” step (step 3/3) that will recognize these classes and ensure their execution.
- ⁇ SETWHEEL is assumed to have been initialized in the Reserved Words and linked in the Execution algorithm to the vehicle wheel control program.
- Cx is a class G whose function ⁇ SETWHEEL is
- ⁇ SETWHEEL0 ⁇ sets the wheels straight and ⁇ SETWHEEL-1 ⁇ sets the wheels to the left.
- step 0 preprocessing
- step 1 dichotomy
- step 2 identification-creation
- step 3 application 3
- the G Dep and Equivalence classes will be of the form: ⁇ [C ⁇ " ⁇ [Feqv ⁇ " Ceqv ⁇ ] ⁇ with ⁇ [Feqv ⁇ " ⁇ EQV ⁇ ] ⁇ known / learned and in which
- Feqv is the function here Am and Dep and equivalence
- Ceqv is the class equivalent to C
- the method has an equivalence step which will search on the basis for the classes C which are submitted to it, their equivalent class Ceqv. (step 3/2)
- step 0 pretreatment
- step 1 dichotomy
- step 2 identification-creation
- step 3 application 3, equivalent 3/2, execution 3/3
- step 1 The dichotomy step has no effect here (there is no function in the sequence) (step 1)
- the identification-creation step finds the class because created in (1) above either Cy (step 2)
- the Application step starts the iteration for Cx (step 3/1)
- ⁇ turn left means ⁇ SETWHEEL-1 ⁇ and ⁇ go straight means ⁇ SETWHEEL0 ⁇ so that the ⁇ turn left ⁇ and ⁇ go straight ⁇ commands are now correctly executed.
- the received sequence is of type "String" (string of characters) each character is coded is:
- the first classes are reserved for characters.
- the non-character classes coming next. And this in order to make coexist in the same sequence characters and classes (Rmots for example) whose No will be beyond 255 (ascii case), the sequence is classically binarized on 32 bits for current computers (step 0 / 1).
- the maximum number of classes then being (2 ⁇ 31 -1) (Class nos are always positive).
- sequences processed by the method are at the end of the Sequence Reception (step 0) and for the continuation of the "Large Integer” chains in English terminology for "large integers” in French, During preprocessing, the sequences corresponding to reserved words are replaced by their class number
- the mission of the Dichotomizer is to insert in the sequence to be treated, the classes IG (CRO to CR3) relying on the functions
- the Dichotomizer does not have to process this function and therefore does not modify the sequence.
- the first function found will be ⁇ turn ⁇ , it does not participate as a function to a class G and can therefore be processed (step 1/1).
- the scope of its parameter stops at CR2 ⁇ " ⁇ , indeed the entity to be treated is the parameter:
- the dichotomizer transforms ⁇ turn right ⁇ into ⁇ [turn ⁇ "right ⁇ ] ⁇ in the sequence (step 1/3).
- the Dichotomizer will then resume the function search (step
- ⁇ SETWHEEL1 ⁇ is transformed in the sequence into ⁇ [ ⁇ SETWHEEL ⁇ »1 ⁇ ] ⁇ (step 1/3).
- the Dichotomizer resumes the function search and will no longer find any unexpressed.
- the Dichotomizer has received a sequence already containing IG classes.
- IG sequence A sequence including the IG classes for all known functions of the process and present in the sequence will be referred to as "IG sequence”.
- This sequence format is the output format of the sequences for the Dichotomizer. This will also be the input format of the sequences for the Identification-Creation algorithm.
- the mission of the identification-creation stage is either to identify (if known) the component classes of the sequence (delimited by the IG classes), or to create them on the database.
- ⁇ 1 ⁇ being a character, it is by construction already known ( ⁇ 1 ⁇ will be the class 49 in the cases Ascii extended, Utf8).
- This IG sequence will be gradually converted, as identifications or creations of classes, in LIG sequence in which:
- LIG's "L” comes from Lead class in English terminology for “leading class number” in French. In other words "sequence with Lead and Indication G classes". In this format a sequence carries in itself all the information necessary for its Application / execution (step 3).
- a class G can be created only if one already knows on the base its Function AND its Parameter. In other words, if a class number has already been assigned to them. The sequences of the Dichotomizer stage take this into account.
- step 2/7 class G unknown on the sequence whose function and parameter are also unknown on the sequence -> return iteration.
- step 2/1) Following iteration for class ⁇ t ⁇ in the 3 rd position in (1) (step 2/1)
- step 2/3 we are in the case of step 2/3.
- second iteration - that is the position of the beginning of the class G corresponding to the CR3 in sixth position and in the case of absence of lead.
- the LIG sequences have the extreme advantage of linearizing the class structure (see Figure 6). Consequently, an iteration on the classes of a LIG sequence makes it possible to "scan" all the classes involved in the structure starting from the largest ones and respecting the chronological order.
- the LIG sequences account for how the process has learned and understood a sequence.
- F1 be an upstream function
- F2 a swallow function
- Pi a parameter
- the process takes the hierarchy most often known. That is, if Ni> Nj.
- the process takes F1 >>> F2 and vice versa.
- ⁇ Turn ⁇ is an upstream
- ⁇ turn left means ⁇ SETWHEEL-1 ⁇
- the dichotomy step finds the function ⁇ turn ⁇ with as parameter all or part of ⁇ left means ⁇ SETWHEEL-1 ⁇
- the method accordingly limits the scope of the parameter to ⁇ left ⁇ .
- the dichotomy step is not found in this Downstream parameter, so it will not reduce the scope of the parameter and will insert the IG classes such as:
- the class to create is No G and known G -> no update to make, the class remains G on the database
- the class to be created is G and known G in a different form (the meaning and / or the function and the parameter are different) -> we update the expression in terms of class and meaning (the new version takes precedence over the 'Ancient),
- the dichotomy step may establish an erroneous dichotomy for a sequence. It is then enough to re-enter said sequence by mentioning the classes IG according to what must be understood. The sequence will then be corrected.
- ⁇ turn left means ⁇ SETWHEEL-1 ⁇
- ⁇ SETENGINE is linked in the execution stage to the engine control.
- the corresponding orders are:
- ⁇ go ahead means ⁇ SETENGINE1 ⁇
- the identification-creation step is performed as the dichotomy step.
- a function class is also called just function.
- a parameter class is also called just parameter.
- the principle of the dichotomy is that it is the hierarchical relationships between two functions that make it possible to dichotomise a sequence.
- Example 1 Case of a Known Hierarchy in the Hierarchical Reporting Table
- step b) of dichotomy it is recognized the function upstream ⁇ turn ⁇ which is correctly bordered and can thus be considered.
- ⁇ turn ⁇ is an upstream function
- ⁇ turn ⁇ is an upstream function
- the System does not have a direct / simple reporting relationship
- ⁇ turn right means ⁇ SETWHEEL1 ⁇
- ⁇ turn ⁇ is an upstream function
- ⁇ turn right means ⁇ SETWHEEL1 ⁇
- the upstream ⁇ turn ⁇ function will be processed because when it is well defined, when searching for the scope of its parameter, it is met with the upstream function ⁇ means ⁇ , it is then evaluated the hierarchical relationship of the function ⁇ turn ⁇ versus the function list ParDep of ⁇ means ⁇ is S6, the generalized class S5 gives us an occurrence such that ⁇ turn ⁇ ⁇ S6.
- the upstream function ⁇ turn ⁇ will be processed because, when searching for the range of its parameter, the upstream Dep function ⁇ is met, it is then evaluated the hierarchical relationship of the function ⁇ turn ⁇ versus the list of ParDep functions of ⁇ signifies ⁇ , which list is empty, so there is no hierarchy via this mode, it is then searched for the related functions of ⁇ signifies ⁇ , the system will then find via S7 and S8 the function ⁇ means ⁇ , it will look for the hierarchy of ⁇ turn ⁇ versus ⁇ means ⁇ , which the system knows as being
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Automation & Control Theory (AREA)
- Stored Programmes (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1850996A FR3077656A1 (fr) | 2018-02-07 | 2018-02-07 | Procede de transformation d’une sequence pour la rendre executable par une machine |
PCT/FR2019/050272 WO2019155168A1 (fr) | 2018-02-07 | 2019-02-07 | Procédé de transformation d'une séquence pour la rendre exécutable par une machine |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3750051A1 true EP3750051A1 (fr) | 2020-12-16 |
Family
ID=62749079
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP19710040.7A Ceased EP3750051A1 (fr) | 2018-02-07 | 2019-02-07 | Procédé de transformation d'une séquence pour la rendre exécutable par une machine |
Country Status (4)
Country | Link |
---|---|
US (1) | US11009845B2 (fr) |
EP (1) | EP3750051A1 (fr) |
FR (1) | FR3077656A1 (fr) |
WO (1) | WO2019155168A1 (fr) |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4499553A (en) * | 1981-09-30 | 1985-02-12 | Dickinson Robert V | Locating digital coded words which are both acceptable misspellings and acceptable inflections of digital coded query words |
US6108666A (en) * | 1997-06-12 | 2000-08-22 | International Business Machines Corporation | Method and apparatus for pattern discovery in 1-dimensional event streams |
US6125342A (en) * | 1997-11-18 | 2000-09-26 | L & H Applications Usa, Inc. | Pronoun semantic analysis system and method |
JP3004254B2 (ja) * | 1998-06-12 | 2000-01-31 | 株式会社エイ・ティ・アール音声翻訳通信研究所 | 統計的シーケンスモデル生成装置、統計的言語モデル生成装置及び音声認識装置 |
US6243670B1 (en) | 1998-09-02 | 2001-06-05 | Nippon Telegraph And Telephone Corporation | Method, apparatus, and computer readable medium for performing semantic analysis and generating a semantic structure having linked frames |
US6173441B1 (en) * | 1998-10-16 | 2001-01-09 | Peter A. Klein | Method and system for compiling source code containing natural language instructions |
US7085708B2 (en) * | 2000-09-23 | 2006-08-01 | Ravenflow, Inc. | Computer system with natural language to machine language translator |
US7552124B2 (en) * | 2004-06-17 | 2009-06-23 | Ixi Mobile (R&D), Ltd. | Natural language for programming a specialized computing system |
JP4459267B2 (ja) * | 2005-02-28 | 2010-04-28 | パイオニア株式会社 | 辞書データ生成装置及び電子機器 |
US9846739B2 (en) * | 2006-10-23 | 2017-12-19 | Fotonation Limited | Fast database matching |
US7809765B2 (en) * | 2007-08-24 | 2010-10-05 | General Electric Company | Sequence identification and analysis |
US8077984B2 (en) * | 2008-01-04 | 2011-12-13 | Xerox Corporation | Method for computing similarity between text spans using factored word sequence kernels |
JP5391583B2 (ja) * | 2008-05-29 | 2014-01-15 | 富士通株式会社 | 検索装置、生成装置、プログラム、検索方法および生成方法 |
JP4701292B2 (ja) * | 2009-01-05 | 2011-06-15 | インターナショナル・ビジネス・マシーンズ・コーポレーション | テキスト・データに含まれる固有表現又は専門用語から用語辞書を作成するためのコンピュータ・システム、並びにその方法及びコンピュータ・プログラム |
ITRM20130022A1 (it) * | 2013-01-11 | 2014-07-12 | Natural Intelligent Technologies S R L | Procedimento e apparato di riconoscimento di scrittura a mano |
US10843080B2 (en) * | 2016-02-24 | 2020-11-24 | Virginia Tech Intellectual Properties, Inc. | Automated program synthesis from natural language for domain specific computing applications |
-
2018
- 2018-02-07 FR FR1850996A patent/FR3077656A1/fr not_active Withdrawn
-
2019
- 2019-02-05 US US16/268,247 patent/US11009845B2/en active Active
- 2019-02-07 EP EP19710040.7A patent/EP3750051A1/fr not_active Ceased
- 2019-02-07 WO PCT/FR2019/050272 patent/WO2019155168A1/fr unknown
Non-Patent Citations (1)
Title |
---|
CHEN YAO-HSIANG ET AL: "Voice control design of a mobile robot using shared-control approach", 2017 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), IEEE, 5 October 2017 (2017-10-05), pages 105 - 110, XP033270915, DOI: 10.1109/SMC.2017.8122586 * |
Also Published As
Publication number | Publication date |
---|---|
US20190243324A1 (en) | 2019-08-08 |
FR3077656A1 (fr) | 2019-08-09 |
WO2019155168A1 (fr) | 2019-08-15 |
US11009845B2 (en) | 2021-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chang et al. | The prompt artists | |
EP1836651B1 (fr) | Procédé de recherche, reconnaissance et localisation d'un terme dans l'encre, dispositif, programme d'ordinateur correspondants | |
EP1364316A2 (fr) | Dispositif d'extraction d'informations d'un texte a base de connaissances | |
FR2982388A1 (fr) | Systeme et procede pour la generation productive de mots composes dans une traduction automatique statistique | |
US20230118966A1 (en) | Generation of story videos corresponding to user input using generative models | |
BE1024194A9 (fr) | Procédé d'indentification d'un caractère dans une image numérique | |
WO2007082948A1 (fr) | Procede et dispositif pour extraire des informations et les transformer en donnees qualitatives d'un document textuel | |
BE1008621A3 (fr) | Procedure et procede de communication entre machines et procede generalise de preparation de programmes afferents. | |
FR3017474A1 (fr) | Saisie assistee de regles dans une base de connaissance | |
EP2564362A1 (fr) | Procede et systeme d'adaptation d'un contenu textuel au comportement langagier d'une communaute en ligne | |
EP2126735A1 (fr) | Procede de traduction automatique | |
WO2014173882A1 (fr) | Procédé et dispositif d'extraction automatique de thèmes à partir d'au moins un document contenant du texte | |
WO2016116459A1 (fr) | Procédé de lemmatisation, dispositif et programme correspondant | |
EP3750051A1 (fr) | Procédé de transformation d'une séquence pour la rendre exécutable par une machine | |
WO1995008154A1 (fr) | Procede de demonstration automatique | |
US10878005B2 (en) | Context aware document advising | |
FR2880708A1 (fr) | Procede de recherche dans l'encre par conversion dynamique de requete. | |
Alcolei | Enriched concurrent games: witnesses for proofs and resource analysis | |
FR3144335A1 (fr) | Procédé de génération d'un signal comprenant des graphes de connaissances fusionnés et produit programme d'ordinateur associé | |
EP3114597A1 (fr) | Procédé d'analyse d'une pluralité de messages, produit programme d'ordinateur et dispositif associés | |
Turbin | Comparing Natural Language Models for Software Category Classification | |
EP4155967A1 (fr) | Procédé d'échanges d'informations sur un objet d'intérêt entre une première et une deuxième entités, dispositif électronique d'échange d'informations et produit programme d'ordinateur associés | |
WO2020079109A1 (fr) | Dispositif de traitement automatique de texte par ordinateur | |
FR2939538A1 (fr) | Procede de recherche de correspondances entres differentes sources de donnees. | |
FR3110740A1 (fr) | Procédé de conversion automatique de fichiers numériques |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20200810 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20210906 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R003 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED |
|
18R | Application refused |
Effective date: 20230705 |