CN111309383B - Instruction identification method, device, equipment and storage medium - Google Patents
Instruction identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111309383B CN111309383B CN202010129986.8A CN202010129986A CN111309383B CN 111309383 B CN111309383 B CN 111309383B CN 202010129986 A CN202010129986 A CN 202010129986A CN 111309383 B CN111309383 B CN 111309383B
- Authority
- CN
- China
- Prior art keywords
- instruction
- target
- formatting
- score
- target reference
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/30007—Arrangements for executing specific machine instructions to perform operations on data operands
- G06F9/30025—Format conversion instructions, e.g. Floating-Point to Integer, decimal conversion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30145—Instruction analysis, e.g. decoding, instruction word fields
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a method, a device, equipment and a storage medium for identifying an instruction, wherein the method comprises the following steps: acquiring an input instruction to be identified, identifying the instruction to be identified, and acquiring at least one target keyword corresponding to the instruction to be identified; searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database as a target reference instruction; respectively calculating an instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm; and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score. By adopting the method and the device, the accuracy of instruction identification can be improved.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for instruction recognition, a computer device, and a computer-readable storage medium.
Background
With the development of computer technology and the popularization of intelligent home, more and more household appliances can be intelligently controlled through voice so as to improve the interactivity and intelligence of product control.
Generally, the control of the household appliances is realized by inputting voice by a user, and then performing voice recognition and semantic recognition on natural voice input by the user to convert the natural voice into machine-understandable executable instructions. Typically, the identification of the instructions is by keyword matching. The method comprises the steps of defining related keywords for a preset control instruction, and then matching the corresponding control instruction according to the keywords corresponding to voice input by a user in the scene of instruction recognition.
However, if the keyword input by the user is inaccurate or there are a plurality of instructions for matching the keyword input by the user, the corresponding control instruction may not be determined according to the keyword matching, resulting in failure or error in instruction recognition. In particular, when there is an inclusion relationship among keywords of a plurality of instructions, there is a case where there are a plurality of matching instructions but the instructions cannot be specified. For example, when a keyword corresponding to a voice input by a user is "open" or "air conditioning", it is not possible to determine whether the matched command is a or B, and a command recognition failure occurs.
In summary, the existing instruction recognition technology is not accurate enough for user instruction recognition, and the late-stage control based on the conclusion obtained by the inaccurate analysis and recognition technology also generates a large deviation.
Disclosure of Invention
In view of the above, it is necessary to provide an instruction identification method, an instruction identification apparatus, a computer device, and a computer-readable storage medium.
An instruction recognition method, the method comprising:
acquiring an input instruction to be recognized, recognizing the instruction to be recognized, and acquiring at least one target keyword corresponding to the instruction to be recognized;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
Optionally, the identifying the instruction to be identified to obtain at least one target keyword corresponding to the instruction to be identified further includes:
determining an initial keyword corresponding to the instruction to be identified according to a preset keyword database;
obtaining synonyms corresponding to the instructions to be recognized according to a preset synonym database, and determining synonym keywords corresponding to the synonyms;
and determining the target keywords according to the initial keywords and the synonymous keywords.
Optionally, the preset formatting instruction database includes an instruction index, where the instruction index includes an index identifier corresponding to each formatting instruction;
the searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database as a target reference instruction further comprises:
comparing each target keyword with an index identifier contained in the base instruction index respectively to determine a target index identifier corresponding to each target keyword;
and taking the formatting instruction corresponding to the searched target index identification as the target reference instruction.
Optionally, each formatting instruction contained in the preset formatting instruction database is provided with at least one corresponding instruction reference word;
the searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database as a target reference instruction further comprises:
traversing the formatted instruction database;
for the traversed formatting instruction, judging whether an instruction reference word corresponding to the traversed formatting instruction is matched with the target keyword or not;
and if so, taking the traversed formatting instruction as the target reference instruction.
Optionally, each formatting instruction contained in the preset formatting instruction database is further provided with a corresponding default vector, and the default vector is used for identifying whether each instruction reference word corresponding to the formatting instruction can be defaulted;
the step of calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm further comprises:
for each target reference instruction:
determining an instruction coverage vector corresponding to the target reference instruction, and calculating a first score corresponding to the target reference instruction according to the instruction coverage vector, wherein the element value of the instruction coverage vector is 0 or 1;
determining an add-subtract-divide coverage vector corresponding to the target reference instruction, and calculating a second score corresponding to the target reference instruction according to the add-subtract-divide coverage vector and a default vector corresponding to the target reference instruction, wherein the element value of the add-subtract-divide coverage vector is 1 or-1;
and determining an instruction score corresponding to the target reference instruction according to the first score and the second score.
Optionally, the determining an instruction coverage vector corresponding to the target reference instruction further includes:
and judging whether each instruction reference word corresponding to the target reference instruction has a matched target keyword or not, and determining an instruction coverage vector corresponding to the target reference instruction according to a judgment result.
Optionally, the determining, according to the instruction score, a target instruction corresponding to the instruction to be recognized in at least one target reference instruction further includes:
determining the highest value in the instruction scores of at least one target reference instruction, and taking the target reference instruction corresponding to the highest value as the target instruction.
Optionally, before the target reference instruction corresponding to the highest value is taken as the target instruction, the method further includes:
determining whether the instruction score corresponding to the highest value is greater than or equal to a preset score threshold value;
and taking the target reference instruction corresponding to the highest value as the target instruction under the condition that the instruction score corresponding to the highest value is greater than or equal to a preset score threshold value.
An instruction recognition apparatus, the apparatus comprising:
the target keyword acquisition module is used for acquiring an input instruction to be identified, identifying the instruction to be identified and acquiring at least one target keyword corresponding to the instruction to be identified;
the target reference instruction determining module is used for searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
the instruction score calculation module is used for calculating an instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and the target instruction determining module is used for determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of:
acquiring an input instruction to be recognized, recognizing the instruction to be recognized, and acquiring at least one target keyword corresponding to the instruction to be recognized;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
acquiring an input instruction to be recognized, recognizing the instruction to be recognized, and acquiring at least one target keyword corresponding to the instruction to be recognized;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
The embodiment of the invention has the following beneficial effects:
after the instruction identification method, the instruction identification device, the computer equipment and the computer readable storage medium are adopted, after a user inputs an instruction, the instruction is identified and a corresponding target keyword is determined, then a plurality of target reference instructions corresponding to the target keyword are searched in an instruction database, and an instruction score corresponding to each target reference instruction is calculated; and finally, determining a target instruction in the target reference instruction according to the instruction score, wherein the target instruction is used for being sent to the corresponding control equipment and controlling the control equipment. That is, the corresponding target instruction is not directly determined according to the keyword matching aiming at the instruction input by the user, so that the condition that the keyword matching possibly exists in the scheme is not accurate is avoided; the target reference instructions are searched and determined through the keywords, so that all relevant instructions are considered, and then the target instructions which are most matched with the instructions input by the user are determined in the target reference instructions through calculation of instruction scores, so that the accuracy of instruction identification is improved, the accuracy of subsequent control is improved, and the user experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Wherein:
FIG. 1 is a flow diagram illustrating a method for instruction recognition in accordance with one embodiment;
FIG. 2 is a diagram of a synonym table in one embodiment;
FIG. 3 is a flow diagram illustrating the determination of a target keyword in one embodiment;
FIG. 4 is a diagram of an instruction database in one embodiment;
FIG. 5 is a diagram of a format instruction in one embodiment;
FIG. 6 is a flow diagram illustrating the determination of a target reference instruction in one embodiment;
FIG. 7 is a diagram of an index table in one embodiment;
FIG. 8 is a flow diagram illustrating the calculation of instruction scores according to one embodiment;
FIG. 9 is a flow diagram illustrating the determination of a target instruction in one embodiment;
FIG. 10 is a diagram illustrating calculation of an instruction score for a target reference instruction in one embodiment;
FIG. 11 is a block diagram of an instruction recognition device in one embodiment;
FIG. 12 is a block diagram that illustrates the structure of an instruction score calculation module in one embodiment;
FIG. 13 is a block diagram of a computer device for executing the instruction recognition method in one embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in FIG. 1, in one embodiment, an instruction recognition method is provided. The method is used for identifying the command of voice or text input by a user and determining a target command which can be identified by a machine for controlling so as to realize the control of target equipment.
In this embodiment, the target device controlled by the target command identified by the command identification method according to the embodiment of the present invention may be any device based on a computer system, for example, a home appliance such as an air conditioner.
The operation of the instruction recognition method may be based on a computer system of the target device (that is, a computer system is provided in the target device, and the command recognition is performed based on the computer system), or may be based on an external computer device connected to the target device, for example, a device such as a smart phone, a personal computer, or a server. Specifically, after receiving an instruction input by a user, the instruction is analyzed and recognized to determine a corresponding target instruction, and the target instruction is used for controlling the target device to achieve a corresponding control effect.
In this embodiment, the target device (e.g., smart air conditioner) is connected to a control device (e.g., smart speaker), and the control device is connected to the server.
The control device is used for receiving an instruction input by a user and sending the instruction to the server so that the server can identify the received instruction and determine a corresponding target instruction, then the target instruction is returned to the control device, and the control device sends the corresponding target instruction to the target device to control the target device.
In a specific embodiment, the instruction identification method specifically includes the following steps:
step S102: the method comprises the steps of obtaining an input instruction to be recognized, recognizing the instruction to be recognized, and obtaining at least one target keyword corresponding to the instruction to be recognized.
In this embodiment, the instruction to be recognized input by the user may be input by voice, for example, by a microphone device of the control device, i.e. a smart speaker; in other application scenarios, the input may be via microphone means on other devices.
When the user inputs the instruction to be recognized in a voice manner, the corresponding voice needs to be subjected to voice recognition. Specifically, in step S102, the process of acquiring the input to-be-recognized instruction further includes: the method comprises the steps of obtaining input voice to be recognized, converting the voice to be recognized into corresponding characters according to a preset voice recognition algorithm, and determining a command to be recognized corresponding to the input voice to be recognized according to the converted characters.
For example, the instruction to be recognized may be "please turn on the cooling mode of the air conditioner".
In another embodiment, the instruction to be recognized input by the user may also be input by text, for example, the user may input the instruction to be recognized by inputting text on a mobile phone. In this case, the corresponding instruction to be recognized is directly determined for the text input by the user.
In this embodiment, the process of determining the keyword corresponding to the instruction to be recognized may be a process of determining the keyword in the instruction to be recognized according to a preset keyword database.
Specifically, keywords that may correspond to a control instruction corresponding to the target device are defined in advance, and a corresponding keyword database is determined. And then in the process of determining the target keyword corresponding to the instruction to be recognized, matching the instruction to be recognized with the keyword database, and extracting the keyword matched with the keyword contained in the keyword database from the instruction to be recognized so as to determine the target keyword. In a specific embodiment, a word or a word corresponding to a keyword contained in a keyword database is searched in the instruction to be recognized, if the word or the word is found, the found word or word is used as the keyword extracted from the instruction to be recognized, and the target keyword is determined according to the extracted keyword.
It should be further considered that the terms of different users in the process of inputting the instruction may be different, and in the present embodiment, in the process of determining the target keyword, it is also required to consider whether there is a corresponding synonym for each keyword. In a specific embodiment, synonyms that may exist in keywords that may correspond to control instructions corresponding to the target device need to be defined in advance, and a synonym database corresponding to the keyword database is determined. Then, in the process of determining the target keyword, it is also necessary to determine whether there is a matching keyword from the synonym database.
Specifically, as shown in FIG. 2, an example of a synonym database is given.
Specifically, as shown in fig. 3, the identifying the instruction to be identified to obtain at least one target keyword corresponding to the instruction to be identified further includes:
step S201: determining an initial keyword corresponding to the instruction to be identified according to a preset keyword database;
step S202: obtaining synonyms corresponding to the instructions to be recognized according to a preset synonym database, and determining synonym keywords corresponding to the synonyms;
step S203: and taking the initial keywords and the synonymous keywords as the target keywords.
That is, in the process of determining the target keyword, it is also necessary to consider whether there is a synonym corresponding to the keyword included in the keyword database in the instruction to be recognized. Specifically, the synonym corresponding to the instruction to be recognized is determined according to the synonym database, and then the keyword corresponding to the synonym is determined according to the synonym database, namely the keyword is the synonym keyword, wherein the synonym keyword is also contained in the keyword database.
In this embodiment, in order to improve the accuracy of extracting the keywords, in the process of extracting the keywords, the long keywords need to be prioritized as an extraction principle. For example, when the control instruction is "please open the bedroom air conditioner", and the keyword included in the keyword database includes 3 keywords of the bedroom air conditioner, the bedroom air conditioner and the bedroom air conditioner, the extracted keyword is the "bedroom air conditioner" instead of the bedroom or the bedroom air conditioner on the principle that the long keyword is preferred, and therefore control errors caused by failure in extracting the keyword are avoided.
Step S104: and searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction.
In this embodiment, an instruction database needs to be constructed in advance for all control instructions corresponding to the target device.
Specifically, for example, command open _ air _ conditioner corresponds to a control command "turn on the air conditioner", and command set _ air _ conditioner _ mode { value ═ hot } corresponds to a control command "turn on the air conditioner left and right windrows".
In the formatting command database, standard processing is performed on all control commands corresponding to the target device, and the formatting command database is constructed for the standard control commands (in the present embodiment, "formatting commands"). Specifically, for each formatting instruction, a combination of corresponding four-segment instruction reference words, that is, a combination of 4 instruction reference words, is constructed. Specifically, the instruction reference words include: "device" (device name), "function" (target function), "operation" (target operation), "value" (additional information). The combination of the 4 instruction reference words determines a unique formatting instruction.
For example, fig. 4 shows a formatting command database corresponding to formatting commands for the smart air conditioner.
Further, in some formatting commands, one or more command reference words in the 4 command reference words may be default, and some of the 4 command reference words may not be default. In the instruction formatting database, a default vector corresponding to each formatting instruction is further included to mark whether (device, function, operation, value)4 instruction reference words can be defaulted, 0 indicates that the instruction reference word at the position can be defaulted, 1 indicates that the instruction reference word cannot be defaulted, and the marked values are separated by commas. For example, if the default vector is "1, 0,0, 0", it means that device cannot be defaulted, but all the other three segments can be defaulted.
In specific implementation, according to a plurality of target keywords corresponding to the instruction to be recognized, formatting instructions corresponding to the target keywords are searched in a formatting instruction database to serve as target reference instructions, and the target reference instructions are all formatting instructions associated with the target keywords.
For example, in the case that the target keyword corresponding to the instruction to be recognized is "open, air-conditioning", the target reference instruction includes all formatting instructions matching one or more of the target keywords, that is, includes all formatting instructions matching the target keywords "air-conditioning" and/or "open". That is to say, in this embodiment, by searching for all formatting instructions matched with the target keyword, it is avoided that formatting instructions possibly related to the instruction to be recognized are missed, and the comprehensiveness and accuracy of the instruction to be recognized are improved.
In a specific implementation, in the process of determining the target reference instruction, a formatting instruction matched with one or more target keywords in the target keywords is searched in a preset formatting instruction database to serve as the target reference instruction.
As shown in fig. 5, an example of a corresponding target reference command in the case where the target keyword is "open, air-conditioning" is given.
As described above, each formatting instruction included in the pre-constructed formatting instruction database includes a plurality of instruction reference words such as device, function, operation, value, and the like, and in the process of searching and determining the target reference instruction specifically according to the target keyword, the process of searching and matching each target keyword is performed respectively, so as to ensure that the target reference instruction can include all related formatting instructions comprehensively.
Specifically, as shown in fig. 6, the searching, in a preset formatting instruction database, for at least one formatting instruction corresponding to the at least one target keyword as a target reference instruction further includes:
step S301: traversing the formatted instruction database;
step S302: for the traversed formatting instruction, judging whether an instruction reference word corresponding to the traversed formatting instruction is matched with the target keyword or not;
if yes, go to step S303: taking the traversed formatting instruction as the target reference instruction;
if not, continuing to execute the step S301;
and circularly executing the steps S301-S303 until all the formatting instructions in the formatting instruction database are traversed.
Through the steps S301-S303, all formatting instructions matched with one or more of all target keywords corresponding to the instruction to be recognized can be searched as the target reference instruction, the comprehensiveness of the target reference instruction in the determination process is guaranteed, and the accuracy of subsequent target instruction determination is improved.
In one embodiment, in order to improve the search efficiency of the target reference instruction search, in this embodiment, an index needs to be constructed for a preset formatted instruction database, so as to improve the search efficiency through the index.
Specifically, in a preset formatting instruction database, an instruction index table of "instruction reference word-instruction" is constructed according to instruction reference words corresponding to all formatting instructions contained in the formatting instruction database, where the instruction index table is an instruction index based on the instruction reference words, and the instruction reference words (i.e. corresponding index identifiers) contained in the instruction index are used to determine corresponding instructions (i.e. corresponding formatting instructions). For example, as shown in fig. 7, an example of an instruction index table is given.
In a specific embodiment, in the process of determining the target reference instruction, the method further includes: comparing each target keyword with an index identifier contained in the instruction index respectively, and determining a target index identifier corresponding to each target keyword; and taking the formatting instruction corresponding to the searched target index identification as the target reference instruction.
Specifically, for each target keyword, all matched formatting instructions are determined as target reference instructions according to the instruction index. The introduction of the instruction index can accelerate the matching speed in the searching process of the target reference instruction, the whole formatting instruction database does not need to be searched every time, and only the formatting instruction corresponding to the corresponding target keyword can be searched according to the instruction index.
Step S106: and respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm.
In this embodiment, for all target reference instructions related to a target keyword, an instruction score corresponding to the target reference instruction is determined according to a preset instruction scoring algorithm, where the instruction score is used to identify a matching degree of the target reference instruction and an instruction to be recognized, so that which instruction in the target reference instruction is most matched with the instruction to be recognized can be determined.
Specifically, as mentioned above, the format instruction contained in the preset format instruction database is further provided with a corresponding default vector for identifying whether at least one instruction reference word corresponding to the format instruction can be defaulted. Whether each non-defaultable instruction reference word has a corresponding target keyword can be determined according to the default vector.
Specifically, as shown in fig. 8, the step S106 further includes:
step S401: and determining an instruction coverage vector corresponding to the target reference instruction, and calculating a first score corresponding to the target reference instruction according to the instruction coverage vector.
And aiming at each target reference instruction, determining whether each instruction reference word has a matched target keyword in the item label reference instruction according to the item label reference instruction, and determining a corresponding first score S according to a matching result, wherein S can determine the score S _ i of each instruction reference word according to whether the matched target keyword exists in each instruction reference word, wherein i represents the ith instruction reference word, and then calculating the corresponding first score S according to S _ i.
Specifically, the instruction coverage vector s corresponding to the target reference instruction may be determined according to whether a matching target keyword exists in each instruction reference word. In a specific embodiment, S ═ { S _ i, i denotes the ith instruction reference word }, and S ═ Σ S _ i. Specifically, the value of the element of the instruction coverage vector is 0 or 1. And for each instruction reference word, determining whether the target keyword extracted from the instruction to be recognized contains the instruction reference word, if so, setting the element value of the corresponding instruction coverage vector to be 1, otherwise, setting the element value to be 0. That is, in the case where there is a corresponding target keyword for the instruction reference word, the score corresponding to the first score is 1, and otherwise, no score is given. The degree of match between the instruction to be identified and the target reference instruction can be determined according to the first score.
Step S402: and determining an add-subtract vector corresponding to the target reference instruction according to the instruction coverage vector and a default vector corresponding to the target reference instruction, and calculating a second score corresponding to the target reference instruction according to the add-subtract vector.
The default vector corresponding to the target reference instruction is denoted as v2, v2 ═ { v2_ i, where i denotes the ith instruction reference word }, and v2_ i is whether the ith instruction reference word can be defaulted.
In addition, for each instruction reference word of the target reference instruction, the plus-minus score covering vector v1 is determined according to whether the instruction reference word has a corresponding target keyword. Specifically, if the corresponding instruction reference word has the corresponding target keyword, the coverage value in the corresponding add-subtract-divide coverage vector is 1, otherwise, the coverage value is-1, that is, v1 ═ v1_ i, i represents the ith instruction reference word, and v1_ i is the coverage value corresponding to the ith instruction reference word.
And determining whether the keywords which cannot be defaulted are extracted according to the default vector corresponding to each target reference instruction to increase or decrease the initial score, namely determining a corresponding second score.
The second score delta corresponding to the target reference instruction is determined by, in particular, adding and subtracting the dot product of the override vector v1 and the default vector v2 corresponding to the target reference instruction.
Namely:
delta＝v1·v2，
delta _ i-v 1_ i × v2_ i, i denoting the i-th instruction reference word.
That is, the plus-minus score covers the position of 1, which means that when the instruction reference word of the position is not defaultable, it can be used for adding score, and the plus-minus score covers the position of 1, which means that when the instruction reference word of the position is not defaultable, it can be used for subtracting score. When the instruction reference word at a certain position is available, the position in the default vector is 0, so that the element value of the position in the addition and subtraction coverage can be ignored. If the instruction reference word at a certain position is not defaultable, adding points when the instruction reference word appears in the instruction to be recognized input by the user, and subtracting points when the instruction reference word does not appear; if the instruction reference word at the position is defaultable, the appearance or non-appearance of the instruction reference word in the instruction to be recognized input by the user has no influence on the final instruction score.
Step S403: and determining the command score corresponding to the target reference command according to the first score and the second score.
In this embodiment, in order to improve the accuracy of determining the target instruction, the final instruction score of each target reference instruction is obtained by comprehensively considering the first score and the second score, for example, by calculating according to a preset formula.
In a particular embodiment, an instruction score corresponding to the target reference instruction is determined based on a sum of the first score and the second score.
Namely, according to the following formula
Sorce＝S+delta
And calculating the instruction score corresponding to the target reference instruction.
The method for calculating the instruction score of the target reference instruction is more flexible, whether the matching degree between the keyword corresponding to the instruction to be recognized and the preset formatting instruction is calculated can be accurately calculated, and the accuracy of instruction recognition is improved.
For example, in fig. 5, when the target keyword is "open/air-conditioned", the calculation of the score corresponding to the target reference command included in fig. 5 will be described as an example.
Reference instruction as item 1 in FIG. 5:
The add-subtract override command v1 is [1, -1,1, -1], and its default vector is v2 ═ 1,0,1,0], and the corresponding second score is delta v1 · v2 ═ 1+ -1 ± -0 +1 ═ 0 ═ 2.
The item 1 indicates that the command score corresponding to the reference command is Sorce + delta 2+ 4.
Reference instruction as item 5 in FIG. 5:
entry 5 marks the instruction coverage vector corresponding to the reference instruction as: s ═ 1,0,1,0], the first score is S ═ 1+0+1+0 ═ 2.
The add-subtract override command v1 is [1, -1,1, -1], and its default vector is v2 ═ 1,0,0,1], and the corresponding second score is delta v1 · v2 ═ 1+ -1 ═ 0+ — 1 ═ 0.
The 5 th entry indicates that the command score corresponding to the reference command is Sorce + delta 2+ 0-0.
In another embodiment, if the to-be-recognized instruction is "temperature is raised a little bit", the user does not specify the device object in the to-be-recognized instruction, in this case, the value of the element of the position corresponding to the device reference word of the default vector corresponding to the format instruction "device is air conditioner, function is temperature setting, operation is raised, value is 1 degree" is 0, that is, the instruction reference word of the device position may be default, that is, the default raised temperature refers to raised temperature of the air conditioner, so that the instruction score corresponding to the to-be-recognized instruction is raised, and accuracy of subsequently determining the target instruction is improved. In other embodiments, for a formatting instruction, if the element value of the position corresponding to the instruction reference word device in the corresponding default vector is set to 1, the instruction score of other formatting instructions is reduced, so that even in the case of missing the control device name, the instruction to be recognized can be correctly recognized, and the accuracy of instruction recognition is improved.
That is, according to the above method for calculating the instruction score of the target reference instruction, even when there is an inclusion relationship between the instructions or when there is a lack or a blur in the instruction to be recognized, the matching degree between the formatted instruction and the instruction to be recognized can be accurately calculated, thereby improving the accuracy of the instruction recognition.
As shown in fig. 10, the instruction score calculation results of all the instructions corresponding to the target reference instruction shown in fig. 10 are given.
Step S108: and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
After the instruction score corresponding to each target reference instruction is determined, the instruction which is most matched with the instruction to be identified in all the target reference instructions can be determined as the target instruction according to the instruction score. In a specific embodiment, the highest value in the instruction scores may be used to determine the target instruction, that is, as shown in fig. 9, step S108 further includes:
step S1081: determining a highest value of the instruction scores of the at least one target reference instruction;
step S1083: and taking the target reference instruction corresponding to the highest value as the target instruction.
Further, in order to ensure the effectiveness of the target instruction, it is also required to consider whether the instruction score corresponding to the determined target instruction meets a preset condition, for example, whether the instruction score is greater than or equal to a preset score threshold. And only when the corresponding instruction score is greater than or equal to the preset score threshold value, the corresponding instruction is taken as the target instruction, so that the validity of the instruction is ensured, and the instruction with lower instruction score is prevented from being confirmed as the target instruction.
Specifically, after the step S1081, the method further includes:
step S1082: determining whether the instruction score corresponding to the highest value is greater than or equal to a preset score threshold value;
if yes, go to step S1083: taking the target reference instruction corresponding to the highest value as a target instruction;
if not, go to step S1084: and returning a message of failure of instruction identification.
For example, in the instruction score shown in fig. 9 described above, the preset score threshold value is 2 points. In the scores corresponding to the target reference commands shown in fig. 9, the highest value of the command score is 4 points, and since 4 points >2 points, the command corresponding to the highest value command score is taken as the target command, that is, the target command corresponding to the case where the control command is "turn on the air conditioner" is open _ air _ conditioner, the corresponding command reference words are "air conditioner, function switch, operation switch on, and value.
In one embodiment, as shown in fig. 11, an instruction recognition apparatus is also provided.
Specifically, as shown in fig. 11, the command recognition apparatus includes:
the target keyword acquisition module 102 is configured to acquire an input instruction to be recognized, recognize the instruction to be recognized, and acquire at least one target keyword corresponding to the instruction to be recognized;
a target reference instruction determining module 104, configured to search, in a preset formatting instruction database, at least one formatting instruction corresponding to the at least one target keyword as a target reference instruction;
the instruction score calculation module 106 is configured to calculate an instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and a target instruction determining module 108, configured to determine, according to the instruction score, a target instruction corresponding to the instruction to be identified in at least one target reference instruction.
Optionally, in an embodiment, the target keyword obtaining module 102 is further configured to: determining an initial keyword corresponding to the instruction to be identified according to a preset keyword database; obtaining synonyms corresponding to the instructions to be recognized according to a preset synonym database, and determining synonym keywords corresponding to the synonyms; and determining the target keywords according to the initial keywords and the synonymous keywords.
Optionally, in an embodiment, each formatting instruction included in the preset formatting instruction database is provided with at least one instruction reference word; the target reference instruction determination module 104 is further configured to traverse the formatted instruction database; for the traversed formatting instruction, judging whether an instruction reference word corresponding to the traversed formatting instruction is matched with the target keyword or not; and if so, taking the traversed formatting instruction as the target reference instruction.
Optionally, in an embodiment, the preset formatting instruction database includes an instruction index, where the instruction index includes an index identifier corresponding to each formatting instruction; the target reference instruction determining module 104 is further configured to compare each target keyword with an index identifier included in the instruction index, so as to determine a target index identifier corresponding to each target keyword; and taking the formatting instruction corresponding to the searched target index identification as the target reference instruction.
Optionally, in an embodiment, each formatting instruction included in the preset formatting instruction database is further provided with a corresponding default vector, where the default vector is used to identify whether each instruction reference word corresponding to the formatting instruction can be defaulted; as shown in fig. 12, the instruction score calculating module 106 further includes a first score calculating unit 1061, a second score calculating unit 1062, and a target score calculating unit 1063;
the first score calculation unit 1061 is configured to calculate, according to an instruction coverage vector corresponding to each target reference instruction, a first score corresponding to the target reference instruction according to the instruction coverage vector, where an element value of the instruction coverage vector is 0 or 1;
the second score calculating unit 1062 is configured to calculate, according to the plus-minus score coverage vector corresponding to each target reference instruction, a second score corresponding to the target reference instruction according to the plus-minus score coverage vector and the default vector corresponding to the target reference instruction, where an element value of the plus-minus score coverage vector is 1 or-1;
the target score calculating unit 1063 is configured to determine an instruction score corresponding to the target reference instruction according to the first score and the second score.
Optionally, in an embodiment, the target instruction determining module 108 is further configured to determine whether each instruction reference word corresponding to the target reference instruction has a matched target keyword, and determine an instruction coverage vector corresponding to the target reference instruction according to a determination result.
Optionally, in an embodiment, the target instruction determining module 108 is further configured to determine a highest value in the instruction scores of the at least one target reference instruction, and use the target reference instruction corresponding to the highest value as the target instruction.
FIG. 13 is a diagram illustrating an internal structure of a computer device in one embodiment. The computer device may be specifically a terminal, and may also be a server. As shown in fig. 13, the computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program that, when executed by the processor, causes the processor to implement the instruction recognition method. The internal memory may also have a computer program stored therein, which when executed by the processor, causes the processor to perform the age identification method. Those skilled in the art will appreciate that the architecture shown in fig. 13 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is proposed, comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
acquiring an input instruction to be identified, identifying the instruction to be identified, and acquiring at least one target keyword corresponding to the instruction to be identified;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
In one embodiment, a computer-readable storage medium is proposed, in which a computer program is stored which, when executed by a processor, causes the processor to carry out the steps of:
acquiring an input instruction to be identified, identifying the instruction to be identified, and acquiring at least one target keyword corresponding to the instruction to be identified;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
and determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score.
After the instruction identification method, the instruction identification device, the computer equipment and the computer readable storage medium are adopted, after a user inputs an instruction, the instruction is identified and a corresponding target keyword is determined, then a plurality of target reference instructions corresponding to the target keyword are searched in an instruction database, and an instruction score corresponding to each target reference instruction is calculated; and finally, determining a target instruction in the target reference instruction according to the instruction score, wherein the target instruction is used for being sent to the corresponding control equipment and controlling the control equipment. That is, the corresponding target instruction is not directly determined according to the keyword matching aiming at the instruction input by the user, so that the condition that the keyword matching possibly exists in the scheme is not accurate is avoided; the target reference instructions are searched and determined through the keywords, so that all relevant instructions are considered, and then the target instructions which are most matched with the instructions input by the user are determined in the target reference instructions through calculation of instruction scores, so that the accuracy of instruction identification is improved, the accuracy of subsequent control is improved, and the user experience is improved.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a non-volatile computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the program is executed. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (8)
1. An instruction recognition method, the method comprising:
acquiring an input instruction to be recognized, recognizing the instruction to be recognized, and acquiring at least one target keyword corresponding to the instruction to be recognized;
searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
respectively calculating an instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score;
each formatting instruction contained in the preset formatting instruction database is provided with at least one corresponding instruction reference word; the searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database as a target reference instruction further comprises: traversing the formatting instruction database; for the traversed formatting instruction, judging whether an instruction reference word corresponding to the traversed formatting instruction is matched with the target keyword or not; if so, taking the traversed formatting instruction as the target reference instruction;
each formatting instruction contained in the preset formatting instruction database is also provided with a corresponding default vector, and the default vector is used for identifying whether each instruction reference word corresponding to the formatting instruction can be defaulted or not; the step of calculating the instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm further comprises: for each target reference instruction: determining an instruction coverage vector corresponding to the target reference instruction, and calculating a first score corresponding to the target reference instruction according to the instruction coverage vector, wherein the element value of the instruction coverage vector is 0 or 1; determining an add-subtract-divide coverage vector corresponding to the target reference instruction, and calculating a second score corresponding to the target reference instruction according to the add-subtract-divide coverage vector and a default vector corresponding to the target reference instruction, wherein the element value of the add-subtract-divide coverage vector is 1 or-1; and determining an instruction score corresponding to the target reference instruction according to the first score and the second score.
2. The instruction recognition method according to claim 1, wherein the recognizing the instruction to be recognized to obtain at least one target keyword corresponding to the instruction to be recognized further comprises:
determining an initial keyword corresponding to the instruction to be identified according to a preset keyword database;
obtaining synonyms corresponding to the instructions to be recognized according to a preset synonym database, and determining synonym keywords corresponding to the synonyms;
and determining the target keywords according to the initial keywords and the synonymous keywords.
3. The instruction recognition method according to claim 1, wherein the preset formatting instruction database comprises an instruction index, and the instruction index comprises an index identifier corresponding to each formatting instruction;
the searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database as a target reference instruction further comprises:
comparing each target keyword with an index identifier contained in the instruction index respectively to determine a target index identifier corresponding to each target keyword;
and taking the formatting instruction corresponding to the searched target index identification as the target reference instruction.
4. The instruction recognition method according to claim 1, wherein the determining, in at least one target reference instruction according to the instruction score, a target instruction corresponding to the instruction to be recognized further comprises:
determining the highest value in the instruction scores of at least one target reference instruction, and taking the target reference instruction corresponding to the highest value as the target instruction.
5. The instruction identification method according to claim 4, wherein before the step of using the target reference instruction corresponding to the highest value as the target instruction, the step of further comprising:
determining whether the instruction score corresponding to the highest value is greater than or equal to a preset score threshold value;
and taking the target reference instruction corresponding to the highest value as the target instruction under the condition that the instruction score corresponding to the highest value is greater than or equal to a preset score threshold value.
6. An instruction recognition apparatus, characterized in that the apparatus comprises:
the target keyword acquisition module is used for acquiring an input instruction to be identified, identifying the instruction to be identified and acquiring at least one target keyword corresponding to the instruction to be identified;
the target reference instruction determining module is used for searching at least one formatting instruction corresponding to the at least one target keyword in a preset formatting instruction database to serve as a target reference instruction;
the instruction score calculation module is used for calculating an instruction score corresponding to each target reference instruction according to a preset instruction scoring algorithm;
the target instruction determining module is used for determining a target instruction corresponding to the instruction to be identified in at least one target reference instruction according to the instruction score;
each formatting instruction contained in the preset formatting instruction database is provided with at least one corresponding instruction reference word; the target reference instruction determining module is specifically used for traversing the formatting instruction database; for the traversed formatting instruction, judging whether an instruction reference word corresponding to the traversed formatting instruction is matched with the target keyword or not; if yes, the traversed formatting instruction is used as the target reference instruction;
each formatting instruction contained in the preset formatting instruction database is also provided with a corresponding default vector, and the default vector is used for identifying whether each instruction reference word corresponding to the formatting instruction can be defaulted or not; the instruction score calculation module is specifically configured to: for each target reference instruction: determining an instruction coverage vector corresponding to the target reference instruction, and calculating a first score corresponding to the target reference instruction according to the instruction coverage vector, wherein the element value of the instruction coverage vector is 0 or 1; determining an add-subtract-divide coverage vector corresponding to the target reference instruction, and calculating a second score corresponding to the target reference instruction according to the add-subtract-divide coverage vector and a default vector corresponding to the target reference instruction, wherein the element value of the add-subtract-divide coverage vector is 1 or-1; and determining an instruction score corresponding to the target reference instruction according to the first score and the second score.
7. A computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 5.
8. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010129986.8A CN111309383B (en) | 2020-02-28 | 2020-02-28 | Instruction identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010129986.8A CN111309383B (en) | 2020-02-28 | 2020-02-28 | Instruction identification method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111309383A CN111309383A (en) | 2020-06-19 |
| CN111309383B true CN111309383B (en) | 2022-05-20 |
Family
ID=71156636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010129986.8A Active CN111309383B (en) | 2020-02-28 | 2020-02-28 | Instruction identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111309383B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815314A (en) * | 2019-01-04 | 2019-05-28 | 平安科技（深圳）有限公司 | A kind of intension recognizing method, identification equipment and computer readable storage medium |
| CN110135692A (en) * | 2019-04-12 | 2019-08-16 | 平安普惠企业管理有限公司 | Intelligence grading control method, device, computer equipment and storage medium |
| CN110675870A (en) * | 2019-08-30 | 2020-01-10 | 深圳绿米联创科技有限公司 | Voice recognition method and device, electronic equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8788260B2 (en) * | 2010-05-11 | 2014-07-22 | Microsoft Corporation | Generating snippets based on content features |
| US10452842B2 (en) * | 2017-06-07 | 2019-10-22 | International Business Machines Corporation | Cognitive learning to counter security threats for kinematic actions in robots |
| CN107995376A (en) * | 2017-11-07 | 2018-05-04 | 平安科技（深圳）有限公司 | A kind of user reports a case to the security authorities processing method and terminal device |
| CN109872053A (en) * | 2019-01-28 | 2019-06-11 | 平安科技（深圳）有限公司 | Assets suggest generation method, device, computer equipment and storage medium |
| CN109767787B (en) * | 2019-01-28 | 2023-03-10 | 腾讯科技（深圳）有限公司 | Emotion recognition method, device and readable storage medium |
-
2020
- 2020-02-28 CN CN202010129986.8A patent/CN111309383B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815314A (en) * | 2019-01-04 | 2019-05-28 | 平安科技（深圳）有限公司 | A kind of intension recognizing method, identification equipment and computer readable storage medium |
| CN110135692A (en) * | 2019-04-12 | 2019-08-16 | 平安普惠企业管理有限公司 | Intelligence grading control method, device, computer equipment and storage medium |
| CN110675870A (en) * | 2019-08-30 | 2020-01-10 | 深圳绿米联创科技有限公司 | Voice recognition method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Automatic Lecture Video Content Summarizationwith Attention-based Recurrent Neural Network;M. B. Andra;《2019 International Conference of Artificial Intelligence and Information Technology (ICAIIT)》;20190912;54-59 * |
| 语音关键词检索若干问题的研究;***;《中国博士学位论文全文数据库 信息科技辑》;20140115(第01期);I136-16 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111309383A (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10847146B2 (en) | Multiple voice recognition model switching method and apparatus, and storage medium | |
| CN111325037B (en) | Text intention recognition method and device, computer equipment and storage medium | |
| CN108595695B (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN111104495A (en) | Information interaction method, device, equipment and storage medium based on intention recognition | |
| CN110413730B (en) | Text information matching degree detection method, device, computer equipment and storage medium | |
| CN108427707B (en) | Man-machine question and answer method, device, computer equipment and storage medium | |
| CN110689881B (en) | Speech recognition method, speech recognition device, computer equipment and storage medium | |
| CN112668313A (en) | Intelligent sentence error correction method and device, computer equipment and storage medium | |
| CN110688853B (en) | Sequence labeling method and device, computer equipment and storage medium | |
| CN112446218A (en) | Long and short sentence text semantic matching method and device, computer equipment and storage medium | |
| WO2022052505A1 (en) | Method and apparatus for extracting sentence main portion on the basis of dependency grammar, and readable storage medium | |
| CN111737987B (en) | Intention recognition method, device, equipment and storage medium | |
| US20190095447A1 (en) | Method, apparatus, device and storage medium for establishing error correction model based on error correction platform | |
| CN109614627B (en) | Text punctuation prediction method and device, computer equipment and storage medium | |
| CN113506574A (en) | Method and device for recognizing user-defined command words and computer equipment | |
| CN110970022B (en) | Terminal control method, device, equipment and readable medium | |
| CN111191032A (en) | Corpus expansion method and device, computer equipment and storage medium | |
| CN111368061B (en) | Short text filtering method, device, medium and computer equipment | |
| CN112766319A (en) | Dialogue intention recognition model training method and device, computer equipment and medium | |
| CN114678027A (en) | Error correction method and device for voice recognition result, terminal equipment and storage medium | |
| CN111309383B (en) | Instruction identification method, device, equipment and storage medium | |
| CN116644183B (en) | Text classification method, device and storage medium | |
| CN113160800A (en) | Voice data labeling method, electronic equipment and storage device | |
| CN115858776B (en) | Variant text classification recognition method, system, storage medium and electronic equipment | |
| CN110764423A (en) | Household appliance control method and device and household appliance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |