A semantic parsing method and system, an electronic device, and a medium

Abstract

The application belongs to the technical field of data processing, and aims to provide a semantic analysis method and system, an electronic device and a medium. The method comprises the following steps: receiving consultation data input by a user; inputting the consultation data into a preset sentence conversion model for sentence conversion processing to obtain standardized data corresponding to the consultation data; performing analysis processing on the standardized data to obtain a logical expression corresponding to the standardized data; according to the logical expression, retrieving a consultation result corresponding to the logical expression from a preset question and answer database; and converting the consultation result into audio data and outputting the audio data. The application can realize online guiding, has strong interaction with the user, can facilitate the user to quickly obtain the consultation result, is beneficial to improving the convenience of the user during the tour in the scenic spot, and has high accuracy of the obtained consultation result.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing, and particularly relates to a semantic analysis method and system, an electronic device and a medium. BACKGROUND

[0002] In the business process of a scenic spot or the like, since the area is wide, tourists in the scenic spot usually master the locations of various scenic spots in the scenic spot and the background stories of the various scenic spots through paper maps and brochures and guide signs set in designated areas in the scenic spot. However, the paper maps, brochures and guide signs all have the problem of poor interactivity with users. To solve the above problem, some scenic spots have online platforms that can guide tourists online, which usually set multiple blocks, such as a ticket purchase module, a map guide module and a scenic spot query module.

[0003] The existing online guide platform usually requires users to click the corresponding module by themselves and find the data to be queried by themselves, which will bring unnecessary burden to users and result in poor user experience in the case of a too large area of the scenic spot or too many scenic spots.

[0004] To facilitate tourists to quickly obtain corresponding information, some scenic spots have also appeared electronic explainers that can be used for tourist Q&A.

[0005] However, in the process of using the prior art, the inventors have found that the prior art at least has the following problems:

[0006] In the prior art, a standardized data with a structured specification is usually used to train a Q&A model, and the standardized data and natural sentences are quite different, so that the accuracy of the finally obtained Q&A result is poor. SUMMARY

[0007] The present application aims to at least partially solve the above technical problems, and provides a semantic analysis method, system, electronic device and medium.

[0008] To achieve the above purpose, the present application adopts the following technical solutions:

[0009] In a first aspect, a semantic analysis method is provided, comprising:

[0010] receiving user input consultation data;

[0011] inputting the consultation data into a preset sentence conversion model for sentence conversion processing to obtain standardized data corresponding to the consultation data;

[0012] analyzing the standardized data to obtain a logical expression corresponding to the standardized data;

[0013] According to the logical expression, a consultation result corresponding to the logical expression is retrieved from a preset question and answer database;

[0014] The consultation result is converted into audio data and output.

[0015] The present application can realize online navigation, has strong interaction with users, and can facilitate users to quickly obtain a consultation result, thereby improving the convenience of users during the process of visiting a scenic spot. In addition, the accuracy of the obtained consultation result is high. Specifically, in the implementation process of the present application, sentence conversion processing and analysis processing can be performed based on the consultation data input by the user to obtain a corresponding logical expression, and a corresponding consultation result is retrieved from a preset question and answer database and output in the form of audio. In this process, since the present application converts the consultation data input by the user into a standard form by inputting it into a preset sentence conversion model before analyzing it, the accuracy of subsequent semantic analysis and result output is improved, thereby improving the user experience.

[0016] In one possible design, the consultation data is in the form of text data; the consultation data input by the user includes:

[0017] The voice consultation data input by the user is received;

[0018] The voice consultation data is converted into text format consultation data;

[0019] Converting the voice consultation data into text format consultation data includes:

[0020] A voice-text database is constructed; the voice-text database includes a plurality of voice sample data and text data corresponding to the plurality of voice sample data, respectively;

[0021] The voice consultation data is matched with all voice sample data in the voice-text database to obtain voice sample data with the highest matching degree with the voice consultation data;

[0022] The text data corresponding to the voice sample data with the highest matching degree is extracted from the voice-text database, and the text data is taken as target text data to obtain text format consultation data matched with the voice consultation data.

[0023] In one possible design, the sentence conversion model includes a data generation module, a data storage module, a data input module, a data conversion module, and a data output module;

[0024] The data generation module is configured to regularly crawl natural sentences of specified theme data from the Internet, and analyze the natural sentences to obtain standardized data corresponding to the natural sentences.

[0025] The data storage module is configured to store the natural language sentence and the standardized data corresponding to the natural language sentence.

[0026] The data input module is configured to receive the consultation data and input the consultation data as query information into the data conversion module.

[0027] The data conversion module is configured to receive the consultation data, search for a natural language sentence with the highest similarity to the consultation data in the data storage module, extract standardized data corresponding to the natural language sentence with the highest similarity in the data storage module, and output the extracted standardized data through the data output module.

[0028] In one possible design, the standardized data is parsed to obtain a logical expression corresponding to the standardized data, including:

[0029] The standardized data is subjected to word decomposition processing to obtain all words in the standardized data.

[0030] The all words are subjected to classification processing to map each word in the all words to a corresponding word meaning class; and word meaning classes corresponding to all words in the standardized data constitute a word class sequence.

[0031] The word class sequence is subjected to word meaning chunk analysis to obtain a word meaning chunk sequence corresponding to the word class sequence.

[0032] The word meaning chunk sequence is subjected to analysis processing to obtain a logical expression framework and a logical expression fragment data corresponding to the standardized data.

[0033] The logical expression framework and the logical expression fragment data are subjected to merging processing to obtain a logical expression corresponding to the standardized data.

[0034] In one possible design, the word meaning chunk sequence is subjected to analysis processing to obtain a logical expression framework and a logical expression fragment data corresponding to the standardized data, including:

[0035] The word meaning chunk sequence is subjected to statistical parsing processing by using an HMM model to obtain a linear symbol sequence.

[0036] The linear symbol sequence is subjected to decomposition to obtain a plurality of logical expression fragment data.

[0037] The linear symbol sequence is subjected to data conversion to obtain a logical expression framework corresponding to the standardized data.

[0038] In a possible design, the consultation result corresponding to the logical expression is retrieved from a preset question and answer database according to the logical expression, including:

[0039] The logical expression is compared with each question and answer data in the question and answer database, to obtain a similarity of the logical expression and question data in each question and answer data;

[0040] Question and answer data with the highest similarity to the logical expression are extracted, and answer data in the question and answer data is output, and the answer data is the consultation result corresponding to the logical expression.

[0041] In a possible design, the similarity of the logical expression and question data in any question and answer data is:

[0042]

[0043] In the formula, D is question data in current question and answer data; s i is a word in the question data D; P is a current logical expression; p j is a word in the logical expression; W i is a weight of the word s i ; and C(s i , p j ) is a relevance score of the words s i and p j .

[0044] In a second aspect, a semantic analysis system is provided, configured to implement the semantic analysis method according to any one of the preceding aspects; the semantic analysis system includes a consultation data input module, a sentence conversion module, a data analysis module and a consultation result output module which are sequentially connected in communication; wherein,

[0045] The consultation data input module is configured to receive consultation data input by a user, and input the consultation data into the sentence conversion module;

[0046] The sentence conversion module is configured to perform sentence conversion processing on the consultation data input into a preset sentence conversion model, to obtain standardized data corresponding to the consultation data;

[0047] The data analysis module is configured to perform analysis processing on the standardized data, to obtain a logical expression corresponding to the standardized data;

[0048] The consultation result output module is configured to retrieve a consultation result corresponding to the logical expression from a preset question and answer database according to the logical expression.

[0049] In a third aspect, an electronic device is provided, including:

[0050] a memory for storing computer program instructions; and

[0051] a processor for executing the computer program instructions to accomplish the operations of the semantic parsing method according to any one of the above.

[0052] In a fourth aspect, a computer readable storage medium is provided for storing computer readable computer program instructions configured to perform the operations of the semantic parsing method according to any one of the above when executed. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 is a flowchart of a semantic parsing method in an embodiment;

[0054] Figure 2 is a module block diagram of a sentence conversion model in an embodiment;

[0055] Figure 3 is a module block diagram of a semantic parsing system in an embodiment;

[0056] Figure 4 is a module block diagram of an electronic device in an embodiment. DETAILED DESCRIPTION

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the present application will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the drawings is only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings. It should be noted that the description of these embodiments is used to help understand the present application, but does not constitute a limitation on the present application.

[0058] Embodiment 1:

[0059] The present embodiment discloses a semantic parsing method, which can be executed by a computer device or a virtual machine with certain computing resources, such as a personal computer, a smart phone, a personal digital assistant, a wearable device, or a virtual machine.

[0060] As shown in Figure 1 , a semantic parsing method can include but is not limited to the following steps:

[0061] S1. receiving user input consultation data; it should be noted that the consultation data includes but is not limited to the following categories: querying the location information of a specified scenic spot, querying the path information from the current location to a specified scenic spot, querying the scenic spots that meet the specified conditions, querying the historical background information of a specified scenic spot, etc.

[0062] In this embodiment, the consultation data is in the form of text; the consultation data input by the user includes:

[0063] The voice consultation data input by the user; it should be noted that the voice consultation data can be input by the user through a microphone or other device, which is not limited here.

[0064] Convert the voice consultation data into text format consultation data;

[0065] In which, the voice consultation data is converted into text format consultation data, including:

[0066] Build a voice text database; the voice text database includes a plurality of voice sample data, and text data corresponding to the plurality of voice sample data respectively;

[0067] Match the voice consultation data with all voice sample data in the voice text database to obtain the voice sample data with the highest matching degree with the voice consultation data;

[0068] Extract the text data corresponding to the voice sample data with the highest matching degree from the voice text database, and take the text data as the target text data, to obtain the text format consultation data matched with the voice consultation data; it should be understood that the target text data obtained finally is the text format consultation data.

[0069] S2. Input the consultation data into a preset sentence conversion model for sentence conversion processing to obtain standardized data corresponding to the consultation data;

[0070] In this embodiment, the sentence conversion model is used to convert the consultation data into standardized data; as shown in Figure 2 The sentence conversion model includes a data generation module, a data storage module, a data input module, a data conversion module and a data output module;

[0071] The data generation module is used to periodically crawl natural sentences of specified theme data from the Internet, and analyze the natural sentences to obtain standardized data corresponding to the natural sentences;

[0072] The data storage module is used to store the natural sentences and the standardized data corresponding to the natural sentences;

[0073] The data input module is used to receive the consultation data and input the consultation data as query information into the data conversion module;

[0074] The data conversion module is configured to receive the consultation data, retrieve a natural sentence with the highest similarity to the consultation data from the data storage module, extract standardized data corresponding to the natural sentence with the highest similarity from the data storage module, and output the extracted standardized data through the data output module.

[0075] It should be noted that the standardized data is a statement data with a specified logical form. The initially received user input consultation data may have a logical disorder problem, which is not conducive to subsequent computer analysis processing. As an example, the user input consultation data is "how far is xx scenic spot", and the corresponding standardized data is "the distance of xx scenic spot from me". In this embodiment, the user input consultation data is converted into standardized data, which is convenient for computer recognition processing and improves the generalization ability of subsequent computer analysis processing.

[0076] S3. Analyzing the standardized data to obtain a logical expression corresponding to the standardized data; it should be noted that the logical expression is a structured query language, such as SQL language, which can be directly executed by a computer program to retrieve corresponding results from a preset database.

[0077] Specifically, in step S3, the standardized data is analyzed to obtain a logical expression corresponding to the standardized data, including:

[0078] S301. Performing word decomposition processing on the standardized data to obtain all words in the standardized data;

[0079] S302. Classifying all words to map each word in the all words to a corresponding word meaning class; the word meaning classes corresponding to all words in the standardized data constitute a word class sequence; it should be noted that when classifying words in this embodiment, the semantics in the standardized data is used, such as "far" and "near", which have different word meanings but the same semantic function in the standardized data, so they are classified as one class of data; the word meaning classes corresponding to all words in the standardized data are sorted to obtain a word class sequence.

[0080] S303. Performing word meaning chunk analysis on the word class sequence to obtain a word meaning chunk sequence corresponding to the word class sequence; it should be noted that the word meaning chunk refers to the smallest component that can represent a certain meaning, which includes a combination of multiple words, such as "xx scenic spot distance from me", "xx scenic spot background story", etc.

[0081] S304. Analyzing the word meaning chunk sequence to obtain a logical expression framework and a logical expression fragment data corresponding to the standardized data.

[0082] Specifically, the word sense chunk sequence is analyzed to obtain a logical expression framework corresponding to the standardized data and logical expression fragment data, including:

[0083] A1. The word sense chunk sequence is statistically analyzed using an HMM model to obtain a linear symbol sequence.

[0084] A2. The linear symbol sequence is decomposed to obtain a plurality of logical expression fragment data.

[0085] A3. The linear symbol sequence is data-converted to obtain a logical expression framework corresponding to the standardized data. It should be noted that when the linear symbol sequence is data-converted, all symbol data in the linear symbol sequence is first annotated, and the positions of the symbol data are obtained, and then based on the annotated symbol data and their positions, the linear symbol sequence is converted to obtain a logical expression framework corresponding to the linear symbol sequence, i.e., a logical expression framework corresponding to the standardized data.

[0086] Specifically, the HMM (Hidden Markov Model) model is a directed graph model mainly used for time series data modeling. Using the HMM model to analyze the standardized data can facilitate in-depth semantic level analysis of the standardized data, and has high robustness.

[0087] S305. The logical expression framework and the logical expression fragment data are merged to obtain a logical expression corresponding to the standardized data. It should be noted that when the logical expression framework and the logical expression fragment data are merged, the logical expression fragment data is taken as a child node of each node in the logical expression framework, i.e., a logical expression corresponding to the standardized data is obtained.

[0088] S4. According to the logical expression, a consultation result corresponding to the logical expression is retrieved from a pre-set question and answer database.

[0089] In step S4, according to the logical expression, a consultation result corresponding to the logical expression is retrieved from a pre-set question and answer database, including:

[0090] S401. The logical expression is compared with each question and answer data in the question and answer database to obtain a similarity between the logical expression and question data in each question and answer data.

[0091] Specifically, in the present embodiment, the similarity between the logical expression and the question data in any question and answer data is:

[0092]

[0093] wherein, D is the question data in the current question and answer data; s i is the word in the question data D; P is the current logical expression; p j is the word in the logical expression; W i is the weight of the word s i ; C(s i , p j ) is the correlation score of the words s i and p j .

[0094] S402. Extract the question and answer data with the highest similarity to the logical expression, and output the answer data in the question and answer data, which is the consultation result corresponding to the logical expression.

[0095] S5. Convert the consultation result into audio data and output.

[0096] In the embodiment, the question and answer database includes a plurality of question data corresponding to the current scenic spot and answer data corresponding to the question data, so as to quickly obtain the consultation result by the user. Of course, the corresponding result can also be searched on the Internet according to the logical expression, but the time-consuming is longer by using this way, and the result obtained by searching is more complicated, which often needs to be further screened by the user, resulting in poor user experience.

[0097] Semantic analysis is the process of converting natural language into a formal computer understandable, accepted and executed meaning representation or logical form. The meaning representation language is a formal representation language with a definite context-free grammar to ensure that each meaning representation has a unique parse tree, and each production in the grammar is treated as a semantic concept.

[0098] The embodiment can realize online tour, has strong interaction with the user, and can facilitate the user to quickly obtain the consultation result, and is beneficial to improve the convenience of the user in the scenic tour process. In addition, the accuracy of the obtained consultation result is high. Specifically, in the implementation process of the embodiment, the sentence conversion processing and analysis processing can be performed based on the consultation data input by the user to obtain the corresponding logical expression, and the corresponding consultation result is retrieved from the preset question and answer database and output in the form of audio. In this process, since the consultation data input by the user is analyzed in advance by inputting it into the preset sentence conversion model for sentence conversion processing to realize the standardization of the consultation data, the accuracy of subsequent semantic analysis and result output is improved, which is beneficial to improve the user experience.

[0099] Embodiment 2:

[0100] The embodiment discloses a semantic analysis system for implementing the semantic analysis method in embodiment 1; as shown in Figure 3 The semantic analysis system comprises, which are sequentially connected in communication, a consultation data input module, a sentence conversion module, a data analysis module and a consultation result output module; wherein,

[0101] The consultation data input module is configured to receive the consultation data input by a user and input the consultation data into the sentence conversion module;

[0102] The sentence conversion module is configured to input the consultation data into a preset sentence conversion model for sentence conversion processing to obtain standardized data corresponding to the consultation data;

[0103] The data analysis module is configured to analyze the standardized data to obtain a logical expression corresponding to the standardized data;

[0104] The consultation result output module is configured to retrieve a consultation result corresponding to the logical expression from a preset question and answer database according to the logical expression.

[0105] Embodiment 3:

[0106] On the basis of embodiment 1 or 2, the embodiment discloses an electronic device, which can be a smart phone, a tablet computer, a notebook computer or a desktop computer, etc. The electronic device can be referred to as a terminal, a portable terminal, a desktop terminal, etc., as shown in Figure 4 The electronic device comprises:

[0107] a memory configured to store computer program instructions; and

[0108] a processor configured to execute the computer program instructions to complete the operations of the semantic analysis method according to any one of embodiments 1.

[0109] In particular, the processor 301 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 can be implemented in the form of at least one of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), a PLA (Programmable Logic Array). The processor 301 can also include a main processor and a coprocessor, the main processor being a processor for processing data in an awake state, also referred to as a CPU (Central Processing Unit), and the coprocessor being a low-power processor for processing data in a standby state. In some embodiments, the processor 301 can be integrated with a GPU (Graphics Processing Unit) that is responsible for rendering and drawing of content to be displayed by the display screen.

[0110] The memory 302 can include one or more computer-readable storage media that can be non-transitory. The memory 302 can also include a high-speed random access memory, and a nonvolatile memory such as one or more disk storage devices, flash storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 302 is used to store at least one instruction for being executed by the processor 301 to implement the semantic parsing method provided in Embodiment 1 of the present application.

[0111] In some embodiments, the terminal can also optionally include a communication interface 303 and at least one peripheral device. The processor 301, the memory 302, and the communication interface 303 can be connected through a bus or a signal line. Each peripheral device can be connected to the communication interface 303 through a bus, a signal line, or a circuit board. In particular, the peripheral device includes at least one of a radio frequency circuit 304, a display screen 305, and a power supply 306.

[0112] The communication interface 303 can be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 301 and the memory 302. In some embodiments, the processor 301, the memory 302, and the communication interface 303 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 301, the memory 302, and the communication interface 303 can be implemented on a separate chip or circuit board, and the present embodiment does not limit this.

[0113] The radio frequency (RF) circuit 304 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 304 communicates with communication networks and other communication devices via electromagnetic signals.

[0114] Display screen 305 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof.

[0115] Power supply 306 is used to supply power to various components in electronic devices.

[0116] Example 4:

[0117] Based on any one of the embodiments 1 to 3, this embodiment discloses a computer-readable storage medium for storing computer-readable computer program instructions, which are configured to perform operations as described in the semantic parsing method in embodiment 1 at runtime.

[0118] Obviously, those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, the present invention is not limited to any particular hardware and software combination.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0120] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A semantic parsing method, characterized in that: include: Receive user-input consultation data; The consultation data is input into a preset statement conversion model for statement conversion processing to obtain standardized data corresponding to the consultation data. The standardized data is parsed to obtain the logical expression corresponding to the standardized data; Based on the logical expression, the corresponding consultation result is retrieved from a preset question-and-answer database; The consultation results are converted into audio data and output. The standardized data is parsed to obtain the corresponding logical expression, including: The standardized data is subjected to word decomposition processing to obtain all the words in the standardized data; All the words are classified so that each word is mapped to a corresponding semantic class; the semantic classes corresponding to all words in the standardized data constitute a word class sequence. Perform semantic chunking analysis on the word class sequence to obtain the semantic chunking sequence corresponding to the word class sequence; The semantic chunk sequence is analyzed and processed to obtain the logical expression framework and logical expression fragment data corresponding to the standardized data; The logical expression framework and the logical expression fragment data are merged to obtain the logical expression corresponding to the standardized data; The semantic chunk sequence is analyzed and processed to obtain the logical expression framework and logical expression fragment data corresponding to the standardized data, including: The semantic chunk sequence is statistically analyzed using an Hidden Markov Model (HMM) to obtain a linear symbol sequence. The linear symbol sequence is decomposed to obtain multiple logical expression fragment data; The linear symbol sequence is transformed to obtain the logical expression framework corresponding to the standardized data; Based on the logical expression, the corresponding consultation results are retrieved from a preset question-and-answer database, including: The logical expression is compared with each question and answer data in the question and answer database to obtain the similarity between the logical expression and the question data in each question and answer data; Extract the question-and-answer data with the highest similarity to the logical expression, and output the answer data from the question-and-answer data. The answer data is the consultation result corresponding to the logical expression. The similarity between the logical expression and the question data in any question-and-answer dataset is: ； In the formula, D This refers to the question data in the current question and answer data; s i For question data D Words in the text; P This is the current logical expression; p j For words in a logical expression; For words s i The weights; For words s i and p j The relevance score.

2. The semantic parsing method according to claim 1, characterized in that: The consultation data is in text format; Receive user-input consultation data, including: Receive voice consultation data input by the user; Convert the voice consultation data into text-format consultation data; The process of converting the voice consultation data into text-format consultation data includes: Construct a speech-text database; the speech-text database includes multiple speech sample data and text data corresponding to the multiple speech sample data respectively; The voice consultation data is matched with all voice sample data in the voice text database to obtain the voice sample data with the highest matching degree with the voice consultation data. Extract the text data corresponding to the voice sample data with the highest matching degree from the voice-text database, and use this text data as the target text data to obtain consultation data in text format that matches the voice consultation data.

3. The semantic parsing method according to claim 1, characterized in that: The statement conversion model includes a data generation module, a data storage module, a data input module, a data conversion module, and a data output module; The data generation module is used to periodically crawl natural sentences of data on a specified topic from the Internet, and parse the natural sentences to obtain standardized data corresponding to the natural sentences. The data storage module is used to store the natural language statement and the standardized data corresponding to the natural language statement; The data input module is used to receive the consultation data and input the consultation data as query information into the data conversion module; The data conversion module is used to receive the consultation data, retrieve the natural sentence with the highest similarity to the consultation data in the data storage module, extract the standardized data corresponding to the natural sentence with the highest similarity in the data storage module, and then output the extracted standardized data through the data output module.

4. A semantic parsing system, characterized in that: This system is used to implement the semantic parsing method as described in any one of claims 1 to 3; the semantic parsing system includes a consultation data input module, a statement conversion module, a data parsing module, and a consultation result output module, which are sequentially connected in communication; wherein... The consultation data input module is used to receive consultation data input by the user and input the consultation data into the statement conversion module; The statement conversion module is used to input the consultation data into a preset statement conversion model for statement conversion processing to obtain standardized data corresponding to the consultation data. The data parsing module is used to parse and process the standardized data to obtain the logical expression corresponding to the standardized data; The consultation result output module is used to retrieve the consultation result corresponding to the logical expression from a preset question and answer database based on the logical expression.

5. An electronic device, characterized in that: include: Memory is used to store computer program instructions; as well as, A processor for executing the computer program instructions to perform the semantic parsing method as described in any one of claims 1 to 3.

6. A computer-readable storage medium for storing computer-readable computer program instructions, characterized in that: The computer program instructions are configured to perform the semantic parsing method as described in any one of claims 1 to 3 at runtime.

Images