Intelligent access control scheduling method, device and equipment based on natural language processing

Abstract

The invention provides an intelligent access control scheduling method, device and equipment based on natural language processing. The method comprises the following steps: matching a question generation text corresponding to a question generation parameter from a resident information database based on the question generation parameter input by a resident; generating a question-answer pair set based on the question generation text, and sending a playing instruction to a voice playing device; receiving a resident answer text collected by the input device, and obtaining a first similarity between the resident answer text and the standard answer text by using a text similarity algorithm; when the first similarity is greater than a preset threshold value, sending an opening instruction to the access controller; random question and answer pairs are generated through question generation texts corresponding to the question generation parameters, and an access control scheduling scheme which is high in safety and timely in response is further provided for residents without door opening cards.

Description

Technical Field

[0001] The present invention relates to the field of natural language processing technology, and more specifically, to an intelligent access control scheduling method, device, computer equipment and readable storage medium based on natural language processing. Background Art

[0002] With the rapid development of my country's economy, smart access control has become widely used due to its high security advantages. Unlike traditional mechanical locks, smart access control refers to comprehensive products such as electromagnetic locks and combination locks that integrate electronic and computer technologies using a large number of electronic components. However, electromagnetic locks require the production and distribution of key cards, and combination locks often use passwords that are shared by multiple people and difficult to customize. Therefore, access credentials for electromagnetic and combination locks are easily lost or leaked, resulting in low security.

[0003] As people's awareness of security continues to grow, the application of access control systems is becoming increasingly widespread. Furthermore, with the continuous maturity of RFID technology, access control systems have also experienced rapid development and entered a mature stage. Access control systems using various technologies have emerged, including proximity card access control systems, fingerprint access control systems, iris access control systems, facial recognition access control systems, and voice-activated access control systems. Each of these systems has its own strengths in security, convenience, and manageability, and their application areas are also expanding. Different access control systems are usually selected according to different building types to meet the requirements.

[0004] The emergence of mobile door-opening technologies, such as WeChat, has brought significant convenience to people, eliminating the need to carry traditional keys and access cards. Instead, they can use a smartphone with WeChat installed to unlock access points such as residential complexes, building gates, apartment doors, and even room doors. However, many residents dislike or forget to carry their access cards or phones, making existing access control scheduling methods ineffective in providing timely assistance to these residents. Summary of the Invention

[0005] In view of the above, the present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an intelligent access control scheduling method, apparatus, computer device, and readable storage medium based on natural language processing, which provides a highly secure and responsive access control scheduling method for residents who do not have access cards.

[0006] To this end, in a first aspect, an embodiment of the present invention provides an intelligent access control scheduling method based on natural language processing, the method specifically comprising: Based on the question generation parameters input by the resident, a question generation text corresponding to the question generation parameters is matched from a resident information database; the question generation parameters include at least one of the resident's name, ID number or house number; Generate a set of question-answer pairs based on the question-generated text, and send a play instruction to the voice playback device; the play instruction includes a first audio file converted from the verification question text in the question-answer pair; receiving the household answer text collected by the input device, and obtaining a first similarity between the household answer text and the standard answer text using a text similarity algorithm; When the first similarity is greater than a preset threshold, an opening instruction is sent to the access controller. Preferably, the question generation parameter includes at least one of a resident's name, ID number, or house number. Preferably, the question generation text generates a question-answer pair, including: Automatically extracting a key phrase set from the question generation text through named entity recognition (NER) and dependency analysis (DP); the key phrase set includes multiple key phrases; Inputting the title generation text and the key phrase set into a question-answer pair generation model to obtain question-answer pairs corresponding to the key phrases; The key phrases are filtered for similarity with the answers in the corresponding question-answer pairs to obtain a filtered question-answer pair set.

[0007] Preferably, the question-answer pair generation model includes a question generator and an answer generator, and the question-answer pair generation model includes a text-to-text transfer Transformer model.

[0008] Preferably, the similarity filtering of the key phrases and the answers in the corresponding question-answer pairs to obtain a filtered question-answer pair set comprises the following steps: Obtaining the precision and recall between the key phrase and the answer in the corresponding question-answer pair; If the precision or recall corresponding to the target key phrase in the key phrase set is less than the overlap threshold, the target question-answer pair corresponding to the target key phrase is filtered.

[0009] Preferably, the method further comprises: If the precision or recall corresponding to the target key phrase in the key phrase set is not less than the overlap threshold, obtaining a first word frequency vector of the target key phrase and a second word frequency vector of the answer in the corresponding question-answer pair respectively; Obtaining a second similarity between the target key phrase and its corresponding answer based on the first word frequency vector and the second word frequency vector; If the second similarity is less than a preset similarity threshold, the target question-answer pair corresponding to the target keyword phrase is filtered.

[0010] Preferably, the method further comprises: When the first similarity is not greater than a preset threshold, a reminder instruction is sent to the voice playback device, where the reminder instruction includes a reminder audio file.

[0011] In a second aspect, an embodiment of the present invention further provides an intelligent access control scheduling device based on natural language processing, the device specifically comprising: A generated text acquisition module is used to match a topic generation text corresponding to the topic generation parameters input by the resident from a resident information database; the topic generation parameters include at least one of the resident's name, ID number, or house number; A playback instruction sending module, which is used to generate a question-answer pair set based on the question generation text and send a playback instruction to the voice playback device; the playback instruction includes a first audio file converted from the verification question text in the question-answer pair; a similarity calculation module, configured to receive the household answer text collected by the input device and obtain a first similarity between the household answer text and the standard answer text using a text similarity algorithm; The opening instruction sending module is used to send an opening instruction to the access controller when the first similarity is greater than a preset threshold.

[0012] In a third aspect, an embodiment of the present invention further provides a computer device comprising: a memory and a processor and a computer program stored in the memory. When the computer program is executed on the processor, the intelligent access control scheduling method based on natural language processing as described in the first aspect is implemented.

[0013] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the intelligent access control scheduling method based on natural language processing as described in the first aspect is implemented.

[0014] The embodiments of the present invention provide an intelligent access control scheduling method and device based on natural language processing. Based on the question generation parameters input by the resident, the method matches the question generation text corresponding to the question generation parameters from the resident information database; generates a question and answer pair set based on the question generation text, and sends a playback instruction to the voice playback device; receives the resident answer text collected by the input device, and uses a text similarity algorithm to obtain a first similarity between the resident answer text and the standard response text; when the first similarity is greater than a preset threshold, sends an opening instruction to the access control controller; generates random question and answer pairs through the question generation text corresponding to the question generation parameters, further providing a highly secure and responsive access control scheduling solution for residents who do not bring door opening cards. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flowchart of an intelligent access control scheduling method based on natural language processing provided by an embodiment of the present invention; Figure 2 A schematic diagram of the framework structure of the playback instruction sending module provided in an embodiment of the present invention; Figure 3 A schematic diagram of a framework structure for generating question-answer pairs based on the question-generated text provided by an embodiment of the present invention; Figure 4 A schematic diagram of a process for extracting a set of key phrases from a topic-generated text according to an embodiment of the present invention; Figure 5 A schematic diagram of the framework structure of the filtering submodule provided in an embodiment of the present invention.

[0016] Figure 6 A schematic diagram of the framework structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0018] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those skilled in the art will appreciate the applicability of other processes and / or the use of other materials.

[0019] Please refer to Figure 1 The present invention aims to provide an intelligent access control scheduling method based on natural language processing, which is applied to an intelligent access control scheduling system. The intelligent access control scheduling system includes an access control controller, a voice playback device, and an input device. The method specifically includes the following steps: S10, based on the question generation parameters input by the resident, matching the question generation text corresponding to the question generation parameters from the resident information database; The question generation parameters are input by the resident through an input device, and the question generation parameters include at least one of the resident's name, ID number or house number.

[0020] S20, generating a set of question-answer pairs based on the question-generated text, and sending a playback instruction to the voice playback device; Specifically, the server automatically extracts a set of key phrases from the question-generated text through named entity recognition (NER) and dependency analysis (DP); the key phrase set contains multiple key phrases; the question-generated text and the key phrase set are input into the question-answer pair generation model to obtain question-answer pairs corresponding to the key phrases; the key phrases are similarly filtered with the answers in the corresponding question-answer pairs to obtain a filtered question-answer pair set. S30, receiving the resident's answer text collected by the input device, and using a text similarity algorithm to obtain a first similarity between the resident's answer text and the standard answer text; S40: When the first similarity is greater than a preset threshold, an opening instruction is sent to the access controller.

[0021] Specifically, when the first similarity is not greater than the preset threshold, a reminder instruction is sent to the voice playback device, wherein the reminder instruction includes a reminder audio file; when the first similarity is not greater than the preset threshold, a reminder instruction is sent to the voice playback device, wherein the reminder instruction includes a reminder audio file.

[0022] The embodiments of the present invention provide an intelligent access control scheduling method and device based on natural language processing. Based on the question generation parameters input by the resident, the method matches the question generation text corresponding to the question generation parameters from the resident information database; generates a question and answer pair set based on the question generation text, and sends a playback instruction to the voice playback device; receives the resident answer text collected by the input device, and uses a text similarity algorithm to obtain a first similarity between the resident answer text and the standard response text; when the first similarity is greater than a preset threshold, sends an opening instruction to the access control controller; generates random question and answer pairs through the question generation text corresponding to the question generation parameters, further providing a highly secure and responsive access control scheduling solution for residents who do not bring door opening cards.

[0023] Please refer to Figure 2 as well as Figure 3 In one embodiment of the application, step S20 specifically includes: S21. Automatically extract a set of key phrases from the question generation text through named entity recognition (NER) and dependency analysis (DP); Among them, the key phrase set contains multiple key phrases; P is the text input generated by the topic, which can be an article, a paragraph or a sentence; K={k1, k2,…,k n} is the set of key phrases extracted from P; Q = {q 1， q2,…,q m} is due to k i (, i = 1, 2, ..., n) and P generate a set of corresponding problems, A = {a1, a2, ..., a m} is q i (i = 1, 2, ..., n) corresponds to the set of answers in P, Q' is the filtered question set, and A' is the filtered answer set.

[0024] The main workflow is as follows: Input P, ​​the key phrase extraction module automatically extracts K from P through named entity recognition (NER) and dependency analysis (DP), and<P,K> Input into the question-answer pair generation module. The question-answer pair generation module has a question generator and an answer generator.<P,K> Generate Q and A combination<P,Q,A> , and then with<P,K> The similarity filtering module performs overlap and similarity filtering on the answers to each question-answer pair and the key phrases that generate the corresponding question to ensure the relevance and quality of the question-answer pair, and finally outputs the filtered question-answer pair.<P,Q',A'> Specifically, when the candidate answers in the question generation text can generate highly relevant questions, such candidate answers are called key phrases of the text. The method of extracting candidate answers from the question generation text is completed by using named entity recognition (NER) and dependency parsing (DP). NER is responsible for marking all named entities in the article, and DP is responsible for analyzing the dependency relationship of the named entity in order to find phrases suitable for generating questions.

[0025] It is understandable that, combined with the way humans ask questions, named entities whose relationship labels with dependent words are: nsubj (noun subject), nsubjpass (passive noun subject), nummod (numerical modifier), advmod (adverbial modifier), amod (adjective modifier), npadvmod (noun as adverbial modifier), appos (appositive modifier), pobj (prepositional object) are directly extracted as key phrases; named entities whose relationship labels with dependent words are poss (possessive modifier) ​​and compound (compound word) are combined according to the position of their dependent words to generate new key phrases.

[0026] According to the above rules, the set is defined as follows: Label1 ={nsubj,nsubjpass,nummod,advmod,amod,npadvmod,appos,pobj} Label2 = {poss, compound} by Figure 4 Take the following example: the shaded area represents the extracted named entities, the arrows point to the entity's dependent words, the labels on the arrows are the entity's dependent labels, and the rounded rectangles represent the key phrases extracted from the text. NER tags four named entities in this text. DP analysis reveals that the dependent labels for the named entities "2015-2016," "Notre Dame," and "18th" are pobj, nsubj, and advmod, respectively, belonging to Label 1 and thus directly becoming key phrases. The dependent label for the named entity "U.S. News & World Report's" is poss, belonging to Label 2. Therefore, when combined with its dependent word "Colleges," it becomes "U.S. News & World Reports Best Colleges," a key phrase.

[0027] S22: input the title generation text and the key phrase set into the question-answer pair generation model to obtain question-answer pairs corresponding to the key phrases; The question-answer generation model consists of a question generator and an answer generator. First, the text and a set of key phrases extracted from it are combined and input into the question generator. The question generator will generate a question corresponding to each key phrase. Then, Q and P are combined and input into the answer generator. The answer generator will generate the answer corresponding to each question. Finally, the text and the corresponding question-answer pair are output. The question-answer generation model can be a fine-tuned text-to-text transfer Transformer (T5) model.

[0028] S23. Filter the key phrases and the answers in the corresponding question-answer pairs based on similarity to obtain a filtered question-answer pair set.

[0029] Specifically, the precision and recall between the key phrase and the answer in the corresponding question-answer pair are obtained; if the precision or recall corresponding to the target key phrase in the key phrase set is less than the overlap threshold, the target question-answer pair corresponding to the target key phrase is filtered.

[0030] In an embodiment of the present application, a method for extracting and filtering key phrases is proposed, which aims to extract key phrases for generating highly relevant questions from unlabeled articles as candidate answers, filter out question-answer pairs with poor correlation between keywords and generated answers, and aim to ensure the quality of the question-answer pairs finally generated. Unlike the round-trip consistency filtering method, it can retain question-answer pairs with inconsistent keywords and answers but similar meanings.

[0031] Please refer to Figure 5 As shown, in one embodiment of the application, in order to solve the problem of poor correlation between question and answer pairs, the similarity between the key phrases and answers of the generated question and answer pairs is compared to determine whether the question and answer pairs are relevant; if the key phrases are consistent with the answers or the similarity is high, then the question and answer pair is considered to have high correlation, otherwise, the question and answer pair is filtered out.

[0032] S41. Obtain the precision and recall between the key phrase and the answer in the corresponding question-answer pair; It should be noted that precision and recall The calculation formula is as follows: Among them, 1− gra m ki ,ai for k i and a i The number of overlapping words in , len is the length of the sentence.

[0033] S42: If the precision or recall corresponding to the target key phrase in the key phrase set is less than the overlap threshold, the target question-answer pair corresponding to the target key phrase is filtered.

[0034] It should be noted that if the precision or recall corresponding to the target key phrase in the key phrase set is not less than the overlap threshold, the first word frequency vector of the target key phrase and the second word frequency vector of the answer in the corresponding question and answer pair are obtained respectively; based on the first word frequency vector and the second word frequency vector, the second similarity between the target key phrase and the corresponding answer is obtained; if the second similarity is less than the preset similarity threshold, the target question and answer pair corresponding to the target key phrase is filtered.

[0035] The calculation formula for the second similarity is: in, , are the word frequency vectors of length m for key phrases and answers respectively.

[0036] In a second aspect, an embodiment of the present invention further provides an intelligent access control scheduling device 100 based on natural language processing, the device specifically comprising: A generated text acquisition module is used to match a topic generation text corresponding to the topic generation parameters input by the resident from a resident information database; the topic generation parameters include at least one of the resident's name, ID number, or house number; A playback instruction sending module, which is used to generate a question-answer pair set based on the question generation text and send a playback instruction to the voice playback device; the playback instruction includes a first audio file converted from the verification question text in the question-answer pair; a similarity calculation module, configured to receive the household answer text collected by the input device and obtain a first similarity between the household answer text and the standard answer text using a text similarity algorithm; The opening instruction sending module is used to send an opening instruction to the access controller when the first similarity is greater than a preset threshold.

[0037] In one embodiment of the application, the playback instruction sending module includes: A phrase extraction submodule, which is used to automatically extract a key phrase set from the question-generated text through named entity recognition (NER) and dependency analysis (DP); the key phrase set includes multiple key phrases; An answer pair acquisition submodule, configured to input the question generation text and the key phrase set into a question-answer pair generation model to obtain question-answer pairs corresponding to the key phrases; The filtering submodule is used to perform similarity filtering on the key phrases and the answers in the corresponding question-answer pairs to obtain a filtered question-answer pair set.

[0038] In one embodiment of the application, the question-answer pair generation model includes a question generator and an answer generator, and the question-answer pair generation model includes a text-to-text transfer Transformer model.

[0039] In one embodiment of the application, the filtering submodule includes: a computing unit configured to obtain a precision and a recall between the key phrase and the answer in the corresponding question-answer pair; The first determination unit is configured to filter the target question-answer pair corresponding to the target key phrase in the key phrase set if the precision or recall corresponding to the target key phrase in the key phrase set is less than an overlap threshold.

[0040] In one embodiment of the application, the filtering submodule further includes: a second determining unit configured to obtain, if the precision or recall corresponding to the target key phrase in the key phrase set is not less than the overlap threshold, a first term frequency vector of the target key phrase and a second term frequency vector of the answer in the corresponding question-answer pair; a similarity calculation unit, configured to obtain a second similarity between the target key phrase and its corresponding answer based on the first word frequency vector and the second word frequency vector; The third determination unit is configured to filter the target question-answer pair corresponding to the target key phrase if the second similarity is less than a preset similarity threshold.

[0041] In one embodiment of the application, the device further includes: When the first similarity is not greater than a preset threshold, a reminder instruction is sent to the voice playback device, where the reminder instruction includes a reminder audio file.

[0042] Thirdly, as Figure 6 As shown, an embodiment of the present invention further provides a computer device 200, comprising: a memory 201 and a processor 202 and a computer program 203 stored on the memory 201. When the computer program 203 is executed on the processor 202, the intelligent access control scheduling method based on natural language processing as described in the first aspect is implemented.

[0043] The computer device 200 may be a desktop computer, a notebook computer, a PDA, a cloud server, or other computing devices. The computer device 200 may include, but is not limited to, a processor 202 and a memory 201. Those skilled in the art will appreciate that Figure 3This is merely an example of the computer device 200 and does not constitute a limitation on the computer device 200 . The computer device 200 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device 200 may also include input and output devices, network access devices, etc.

[0044] In some embodiments, the memory 201 may be an internal storage unit of the computer device 200, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 201 may also be an external storage device of the computer device 200, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the computer device 200. Furthermore, the memory 201 may also include both an internal storage unit of the computer device 200 and an external storage device. The memory 201 is used to store operating devices, application programs, boot loaders, data, and other programs, such as the program code of the computer program. The memory 201 may also be used to temporarily store data that has been output or is about to be output.

[0045] The processor 202 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or any conventional processor.

[0046] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the intelligent access control scheduling method based on natural language processing as described in the first aspect is implemented.

[0047] In this embodiment, if the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process of the above-mentioned method embodiment by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can at least include: any entity or device capable of carrying computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, mobile hard drive, magnetic disk, or optical disk. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electric carrier signals or telecommunication signals.

[0048] The embodiments of the present invention provide an intelligent access control scheduling method and device based on natural language processing. Based on the question generation parameters input by the resident, the method matches the question generation text corresponding to the question generation parameters from the resident information database; generates a question and answer pair set based on the question generation text, and sends a playback instruction to the voice playback device; receives the resident answer text collected by the input device, and uses a text similarity algorithm to obtain a first similarity between the resident answer text and the standard response text; when the first similarity is greater than a preset threshold, sends an opening instruction to the access control controller; generates random question and answer pairs through the question generation text corresponding to the question generation parameters, further providing a highly secure and responsive access control scheduling solution for residents who do not bring door opening cards.

[0049] In the description of this specification, the reference terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0050] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. An intelligent access control scheduling method based on natural language processing, the method is applied to an intelligent access control scheduling system, the intelligent access control scheduling system includes an access control controller, a voice playback device and an input device, characterized in that: The method specifically includes: Based on the question generation parameters input by the resident, a question generation text corresponding to the question generation parameters is matched from a resident information database; the question generation parameters include at least one of the resident's name, ID number or house number; Generate a set of question-answer pairs based on the question-generated text, and send a play instruction to the voice playback device; the play instruction includes a first audio file converted from the verification question text in the question-answer pair; receiving the household answer text collected by the input device, and obtaining a first similarity between the household answer text and the standard answer text using a text similarity algorithm; When the first similarity is greater than a preset threshold, an opening instruction is sent to the access controller.

2. The intelligent access control scheduling method based on natural language processing according to claim 1 is characterized in that: The question generation parameters include at least one of a resident's name, an ID number, or a house number.

3. The intelligent access control scheduling method based on natural language processing according to claim 2 is characterized in that: Generating a question-answer pair based on the title text includes: Automatically extracting a key phrase set from the question generation text through named entity recognition (NER) and dependency analysis (DP); the key phrase set includes multiple key phrases; Inputting the title generation text and the key phrase set into a question-answer pair generation model to obtain question-answer pairs corresponding to the key phrases; The key phrases are filtered for similarity with the answers in the corresponding question-answer pairs to obtain a filtered question-answer pair set.

4. The intelligent access control scheduling method based on natural language processing according to claim 3 is characterized in that: The question-answer pair generation model includes a question generator and an answer generator, and the question-answer pair generation model includes a text-to-text transfer Transformer model.

5. The intelligent access control scheduling method based on natural language processing according to claim 1 is characterized in that: The step of filtering the key phrases and the answers in the corresponding question-answer pairs by similarity to obtain a filtered question-answer pair set includes the following steps: Obtaining the precision and recall between the key phrase and the answer in the corresponding question-answer pair; If the precision or recall corresponding to the target key phrase in the key phrase set is less than the overlap threshold, the target question-answer pair corresponding to the target key phrase is filtered.

6. The intelligent access control scheduling method based on natural language processing according to claim 5 is characterized in that: The method further includes: if the precision or recall corresponding to the target key phrase in the key phrase set is not less than the overlap threshold, respectively obtaining a first word frequency vector of the target key phrase and a second word frequency vector of the answer in the corresponding question-answer pair; Obtaining a second similarity between the target key phrase and its corresponding answer based on the first word frequency vector and the second word frequency vector; If the second similarity is less than a preset similarity threshold, the target question-answer pair corresponding to the target keyword phrase is filtered.

7. The intelligent access control scheduling method based on natural language processing according to claim 1 is characterized in that: The method further comprises: When the first similarity is not greater than a preset threshold, a reminder instruction is sent to the voice playback device, where the reminder instruction includes a reminder audio file.

8. An intelligent access control scheduling device based on natural language processing, characterized in that: The device specifically includes: A generated text acquisition module is used to match a topic generation text corresponding to the topic generation parameters input by the resident from a resident information database; the topic generation parameters include at least one of the resident's name, ID number, or house number; A playback instruction sending module, which is used to generate a question-answer pair set based on the question generation text and send a playback instruction to the voice playback device; the playback instruction includes a first audio file converted from the verification question text in the question-answer pair; a similarity calculation module, configured to receive the household answer text collected by the input device and obtain a first similarity between the household answer text and the standard answer text using a text similarity algorithm; The opening instruction sending module is used to send an opening instruction to the access controller when the first similarity is greater than a preset threshold.

9. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory, which, when executed on the processor, implements the intelligent access control scheduling method based on natural language processing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed by the processor, the intelligent access control scheduling method based on natural language processing as described in any one of claims 1 to 7 is implemented.

Images