Government affair management method for real estate

Abstract

The invention relates to a government affair management method for real estate, and aims to improve the efficiency and accuracy of real estate data processing. Comprising the following steps: carrying out structured processing on real estate text data, establishing mapping pairs between table names and classification types and between field names and text vocabularies, carrying out training based on a table field matching model, and outputting a mapping total vector; a recurrent neural network is adopted to learn complex features and context information, and field matching and clustering are realized through cosine similarity, so that the matching precision and intelligence are improved; according to the scheme, through automatic selection of UART ports and read-write operation of an internal bus, high efficiency and real-time performance of data communication are realized; the platform comprises government affair management equipment and an upper computer, has a selectable level conversion function, and adjusts a communication signal according to an input instruction to ensure the stability of data transmission; the real estate government affair management system improves automation, intelligence and adaptability of real estate government affair management, and has high practical value.

Description

Technical Field

[0001] The present invention relates to the technical field of government administration, and in particular to a government administration method for real estate. Background Art

[0002] With the rapid development of society, the government departments need to formulate and implement more and more administrative affairs every year. According to the different problems that need to be solved each time, the government departments need to collect a large amount of data for analysis and judgment, and finally make corresponding administrative decisions.

[0003] However, with the increase in government affairs every year, many tasks even involve repeated data collection, which greatly wastes human resources. At the same time, each time a corresponding government decision is made, a lot of time is spent on re-analyzing basic data, summarizing rules, and deriving methods for government decision-making. This will undoubtedly increase the time cost of making government decisions and there is a problem of low efficiency. Therefore, there is an urgent need for a government affairs management method and system based on big data, which can assist decision-making through big data technology to solve the problem of waste of human resources each time data is collected and low efficiency in making government decisions. Summary of the invention

[0004] In order to achieve the above-mentioned purpose of improving work efficiency, the present invention provides the following technical solution: a government affairs management method for real estate, comprising the following steps:

[0005] S1. Structural processing is performed on the text data of real estate, and mapping pairs are established between table names and classification types, and between field names and words in the text data;

[0006] S2, using the text data of the real estate after structured processing as the first text data, using the first mapping pair and the second mapping pair in the first text data to train the table field matching model, and outputting a total mapping vector;

[0007] S3. Match the field name with the text data, including:

[0008] S31. After receiving the host computer instruction, the debugging state machine is triggered to actively read the instruction, determine whether it is a valid instruction header, and extract the field names of all master data entity type definition tables;

[0009] S32, inputting the second text data into a table field matching model, and generating a first field name sequence from the field names of the master data entity type definition table obtained by outputting the table field matching model;

[0010] S33, generating a second field name sequence from the extracted field names of the master data entity type definition table, placing the first field name sequence and the second field name sequence into a clustering space, clustering the first field name sequence with the second field name sequence as the clustering center, and generating a plurality of first field name clustering sets;

[0011] S34, marking a target text classification type for the second text data associated with the first field name in the first field name clustering set, wherein the marked target text classification type is the target text classification type mapped by the master data entity type definition table associated with the second field name sequence in the first field name clustering set;

[0012] S4. Determine the transmission line TX and the receiving line RX: Determine the transmission line TX and the receiving line RX according to the data transmission and receiving capabilities of the UART port of the government affairs management device;

[0013] S5. Perform a send operation or a write operation based on the internal bus read and write operation.

[0014] Furthermore, in step S12, an independence assumption is made based on naive Bayes classification, assuming that each feature word in the text is independent under a given category, and the formula is as follows:

[0015] Given a text D and a category C i , Bayes’ theorem can be expressed as:

[0016]

[0017] Among them, P(D|C i ) is in Category C i The likelihood probability of the following text D, P(C i ) is Category C i The prior probability of , P(D) is the marginal probability of text D.

[0018] Further, in step S32, training the table field matching model specifically means that: the parameter of the table field matching model is cosine similarity, the cosine similarity of each mapping pair is obtained by calculation, and the calculated cosine similarity is assigned as the parameter of the table field matching model, and the calculation formula satisfies:

[0019]

[0020] Among them, V i , W i , W i and Q i are the table name, field name, classification type and high-dimensional vector of words in the text data of the ith master data entity definition table; S(V i , U i) indicates V i and U i The cosine similarity of T(W i , Q i ) indicates W i and Q i The cosine similarity of V i ·U j and W i Q i represents the dot product of vectors, |V i |、|U j |、|W i |and|Q i | represents the norm of a vector.

[0021] Furthermore, step S1 specifically includes:

[0022] S11, determining the independent classification type of the real estate text data, extracting the master data entity type definition table, the master data entity type definition table including the table name and the field name;

[0023] S12. Establishing mapping pairs: establishing a first mapping pair between the table name and the classification type, and establishing a second mapping pair between the field name and the word in the text data.

[0024] Furthermore, step S4 specifically includes:

[0025] S41, determining a UART port capable of sending data to the government administration device as a transmission line TX of the government administration device, and determining another UART port as a receiving line RX;

[0026] S42. If an access command is sent to one of the UART ports of the government administration device and a response is received from another port, the port is determined as a receiving line RX and the other port is determined as a transmitting line TX.

[0027] Furthermore, step S5 specifically includes:

[0028] S51. After receiving the operation instruction from the host computer, if it is judged to be a read operation, the state machine initiates an internal bus read operation. The single address mode of the read operation reads the internal bus once, and sends the read data out through the transmit line TX; if it is judged to be a write operation, the state machine initiates an internal bus write operation. The single address mode of the write operation reads the internal bus once, and writes the data received by the receive line RX into the memory of the state machine;

[0029] S52, after completing an internal bus read operation and sending the read data, perform address accumulation, continue to initiate an internal bus read operation, and continue to send the read data through the transmission line TX;

[0030] S53. If the length of the data read from the internal bus reaches the number of operations multiplied by the number of 1s in the byte valid information, the read operation is stopped, and the state machine returns to the initial state to wait for new instructions.

[0031] Furthermore, the UART interface of the government administration device has a selectable level conversion function, and the government administration device has the function of actively outputting UART data and adjusting the receiving line RX and the transmitting line TX signals;

[0032] The government administration equipment responds according to the input instructions, and adjusts the receiving line RX and the transmitting line TX signals according to the user input instructions to make the UART communication normal.

[0033] Furthermore, before step S51, it also includes: determining whether the input terminal for reading UART data is connected to the output terminal having a command output function, and when the determination result is "yes", executing subsequent steps, and when the determination result is "no", not executing subsequent steps.

[0034] Furthermore, in step S32, the table field matching model uses a recurrent neural network to learn complex features and context information through a multi-layer neural network. The basic form of the recurrent neural network includes:

[0035] h t =σ(W hx x t +W ht-1 h t-1 +b t )

[0036] y t =softmax(W ht h t +b y )

[0037] Among them, x t Represents the word vector in the text data of the tth order; W hx Represents the initial weight of the word vector in the text data; W ht-1 and W ht Respectively represent the cycle weights of the model at the t-1th order and the tth order; b t represents the bias at the tth order; h t-1 and h t Respectively represent the hidden state functions of the t-1th and tth order, y t is the predicted output; b y Represents the output bias; σ is the activation function, and softmax represents the multi-task classification output of the calculation results.

[0038] The present invention also provides a government affairs management platform for real estate, which is used to execute the government affairs management method for real estate, and is characterized in that the government affairs management platform includes: government affairs management equipment and a host computer, and the government affairs management equipment and the host computer have a UART interface.

[0039] Beneficial Effects

[0040] The beneficial effects of this technical solution are mainly reflected in the efficiency, accuracy and automation of real estate government administration. The specific results are summarized as follows:

[0041] 1. Improve the efficiency of structured processing of real estate text data: By performing structured processing on real estate text data, mapping pairs are established between table names and classification types, and between field names and text vocabulary, thereby simplifying the data processing process, being able to quickly convert complex text data into structured data information, and improving the efficiency of data management.

[0042] 2. Improve the accuracy of table field matching: Using the table field matching model, the table fields are matched and clustered by calculating the cosine similarity, achieving high-precision matching of field names and text data, ensuring the accurate correspondence between different fields and text content, and effectively improving the accuracy and reliability of field matching.

[0043] 3. Enhance the intelligence of data processing: Introduce a recurrent neural network (RNN) to train the table field matching model, and use a multi-layer neural network to learn complex features and contextual information, so that the model can capture deeper semantic relationships in real estate texts, thereby improving the intelligence and generalization ability of the matching model.

[0044] 4. Achieve efficient UART communication and internal bus operation: The solution automatically selects the transmit and receive lines through the UART port, and combines the read and write operations of the internal bus to achieve efficient data communication and processing. This process has the function of automatically adjusting the TX and RX signals, ensuring the stability and real-time performance of data transmission, and enhancing the communication efficiency of the government management platform.

[0045] 5. Improve the flexibility and automation of the system: This method can automatically respond to and adjust the communication of government management equipment according to different input instructions, and can automatically judge and select the sending / receiving lines during the data communication process, thereby reducing manual intervention and realizing the automation and intelligence of the government management process, which significantly improves the flexibility and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of the steps of the government administration method for real estate proposed by the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 As shown, a government administration method for real estate includes the following steps:

[0049] S1. Structural processing is performed on the text data of real estate, and mapping pairs are established between table names and classification types, and between field names and words in the text data;

[0050] S2, using the text data of the real estate after structured processing as the first text data, using the first mapping pair and the second mapping pair in the first text data to train the table field matching model, and outputting a total mapping vector;

[0051] S3. Match the field name with the text data, including:

[0052] S31. After receiving the host computer instruction, the debugging state machine is triggered to actively read the instruction, determine whether it is a valid instruction header, and extract the field names of all master data entity type definition tables;

[0053] S32, inputting the second text data into a table field matching model, and generating a first field name sequence from the field names of the master data entity type definition table obtained by outputting the table field matching model;

[0054] S33, generating a second field name sequence from the extracted field names of the master data entity type definition table, placing the first field name sequence and the second field name sequence into a clustering space, clustering the first field name sequence with the second field name sequence as the clustering center, and generating a plurality of first field name clustering sets;

[0055] S34, marking a target text classification type for the second text data associated with the first field name in the first field name clustering set, wherein the marked target text classification type is the target text classification type mapped by the master data entity type definition table associated with the second field name sequence in the first field name clustering set;

[0056] S4. Determine the transmission line TX and the receiving line RX: Determine the transmission line TX and the receiving line RX according to the data transmission and receiving capabilities of the UART port of the government affairs management device;

[0057] S5. Perform a send operation or a write operation based on the internal bus read and write operation.

[0058] In step S12, an independence assumption is made based on the naive Bayes classification, assuming that each feature word in the text is independent under a given category. The formula is as follows:

[0059] Given a text D and a category C i , Bayes’ theorem can be expressed as:

[0060]

[0061] Among them, P(D|C i ) is in Category C i The likelihood probability of the following text D, P(C i ) is Category C i The prior probability of , P(D) is the marginal probability of text D.

[0062] In step S32, training the table field matching model specifically means that the parameter of the table field matching model is cosine similarity, the cosine similarity of each mapping pair is obtained by calculation, and the calculated cosine similarity is assigned as the parameter of the table field matching model, and the calculation formula satisfies:

[0063]

[0064] Among them, V i , W i , W i and Q i are the table name, field name, classification type and high-dimensional vector of words in the text data of the ith master data entity definition table; S(V i , U i ) indicates V i and U i The cosine similarity of T(W i , Q i ) indicates W i and Q i The cosine similarity of V i ·U j and W i Q i represents the dot product of vectors, |V i |、|U j |、|W i |and|Q i | represents the norm of a vector.

[0065] Step S1 specifically includes:

[0066] S11, determining the independent classification type of the real estate text data, extracting the master data entity type definition table, the master data entity type definition table including the table name and the field name;

[0067] S12. Establishing mapping pairs: establishing a first mapping pair between the table name and the classification type, and establishing a second mapping pair between the field name and the word in the text data.

[0068] Step S4 specifically includes:

[0069] S41, determining a UART port capable of sending data to the government administration device as a transmission line TX of the government administration device, and determining another UART port as a receiving line RX;

[0070] S42. If an access command is sent to one of the UART ports of the government administration device and a response is received from another port, the port is determined as a receiving line RX and the other port is determined as a transmitting line TX.

[0071] Step S5 specifically includes:

[0072] S51. After receiving the operation instruction from the host computer, if it is judged to be a read operation, the state machine initiates an internal bus read operation. The single address mode of the read operation reads the internal bus once, and sends the read data out through the transmit line TX; if it is judged to be a write operation, the state machine initiates an internal bus write operation. The single address mode of the write operation reads the internal bus once, and writes the data received by the receive line RX into the memory of the state machine;

[0073] S52, after completing an internal bus read operation and sending the read data, perform address accumulation, continue to initiate an internal bus read operation, and continue to send the read data through the transmission line TX;

[0074] S53. If the length of the data read from the internal bus reaches the number of operations multiplied by the number of 1s in the byte valid information, the read operation is stopped, and the state machine returns to the initial state to wait for new instructions.

[0075] The UART interface of the government administration device has a selectable level conversion function, and the government administration device has the function of actively outputting UART data and adjusting the receiving line RX and the transmitting line TX signals;

[0076] The government administration equipment responds according to the input instructions, and adjusts the receiving line RX and the transmitting line TX signals according to the user input instructions to make the UART communication normal.

[0077] Before step S51, it also includes: judging whether the input terminal for reading UART data is connected to the output terminal with command output function, and when the judgment result is "yes", executing the subsequent steps, and when the judgment result is "no", not executing the subsequent steps.

[0078] In step S32, the table field matching model uses a recurrent neural network to learn complex features and context information through a multi-layer neural network. The basic form of the recurrent neural network includes:

[0079] h t =σ(W hx x t +W ht-1 h t-1 +b t )

[0080] y t =softmax(W ht h t +b y )

[0081] Among them, x t Represents the word vector in the text data of the tth order; W hx Represents the initial weight of the word vector in the text data; W ht-1 and W ht Respectively represent the cycle weights of the model at the t-1th order and the tth order; b t represents the bias at the tth order; h t-1 and h t Respectively represent the hidden state functions of the t-1th and tth order, y t is the predicted output; b y Represents the output bias; σ is the activation function, and softmax represents the multi-task classification output of the calculation results.

[0082] In this embodiment, a government affairs management platform for real estate is provided, which is used to execute the government affairs management method, including: real estate registration management module: the platform supports registration functions such as basic information entry and ownership information entry of real estate such as land and houses, and ensures the accuracy and completeness of real estate information through a standardized registration process;

[0083] Query and statistics module: provides multiple query methods, such as query by address, query by right holder, query by real estate type, etc., so that users can quickly find the required information. At the same time, it has powerful statistical functions and can generate various reports and charts to provide data support for decision-making; Ownership management module: records the ownership information of real estate in detail, including right holders, co-ownership, mortgage status, etc. Through the ownership management function, you can clearly understand the ownership status of real estate and provide a basis for property rights transactions and dispute resolution;

[0084] Spatial information management module: Based on GIS technology, the platform can display the spatial location and surrounding environment of real estate, providing users with a more intuitive management view. At the same time, through spatial analysis functions, the value and potential risks of real estate can be evaluated;

[0085] Business process management module: supports customized business processes, such as transfer registration, change registration, cancellation registration, etc. Through process management, it can improve work efficiency and reduce operational errors;

[0086] Archives management module: electronic management of real estate-related archives, convenient for users to check and call at any time. Through the archives management function, long-term preservation and traceability of real estate information can be achieved;

[0087] Early warning and monitoring module: set early warning conditions to monitor the ownership changes and mortgage status of real estate in real time. Once the early warning conditions are triggered, the system will automatically remind relevant personnel to handle the situation.

[0088] System management and maintenance module: provides comprehensive permission management functions to ensure data security. At the same time, it supports data backup and recovery to prevent data loss.

[0089] In this embodiment, the UART interface provides an optional level conversion function, and the government management device can actively output UART data. The algorithm directly recognizes and adjusts the receiving line RX and the transmitting line TX signals to make the UART communication normal. The government management device needs to input instructions to respond. The user input is assigned to the algorithm module, and the algorithm starts to recognize and adjust the receiving line RX and the transmitting line TX signals to make the UART communication normal.

[0090] In this embodiment, when the algorithm module is connected to the UART interface of the government administration device, the algorithm module can automatically detect the access of the peripheral device and start triggering the automatic identification algorithm.

[0091] In this embodiment, the output terminal of the bus write data is connected to the input terminal of the end judgment, and the output terminal of the end judgment is connected to the input terminal of the UART read data.

[0092] In this embodiment, the device responds only after the user inputs an instruction, the start signal output end is connected to the input end of the UART interface, the output end of the UART interface is connected to the input end for reading the data received by the UART, and the output end of the UART interface is connected to the start input end.

[0093] In this embodiment, the output terminal for reading the data received by UART is connected to the input terminal for whether there is a command, the output terminal for whether there is a command is connected to the input terminal for start, and the output terminal for whether there is a command is connected to the input terminal for UART read address.

[0094] In this embodiment, the output end of the UART read address is connected to the input end of the UART read operation number, the output end of the UART read operation number is connected to the input end of the UART read data, and the output end of the UART read data is connected to the input end of the bus write data.

[0095] In this embodiment, the read operation indicates that the host computer reads the specified address information inside the integrated circuit chip, and the write operation indicates that the host computer writes data to the specified address inside the integrated circuit chip. The instruction header is followed by a 32-bit internal bus address, which is the starting address in the case of an address increment operation.

[0096] When in use, first set the UART signal level to a high level. Then, when this algorithm module is connected to the UART interface of the government management device, the algorithm module can automatically detect the access of the peripheral device, and then start to trigger the automatic identification algorithm, by detecting which of the two data lines of the current device will send data. If one of the data lines has sent data, then this data line is the sending line TX, and the other data line is the receiving line RX. If the current device requires the user to input instructions before responding, then this algorithm module will test the two data lines of the government management device with the input instructions. If a response is received, this data line is the receiving line RX signal, and the other data line is the sending line T. X signal, and then the algorithm automatically corrects the UART receiving line RX and sending line TX interface, and the UART interface communicates normally. Expand the scope of application: expand the application scope of the platform to more fields, such as approval matters in the fields of social security, civil affairs, transportation, etc., and improve the service capabilities of smart government affairs. Introduce new technologies: combine new technologies such as artificial intelligence and big data to improve the intelligence level of the platform and achieve more accurate data analysis and prediction. Strengthen security protection: strengthen the platform's security protection measures to ensure the security and integrity of data and prevent data leakage and tampering. Promote use: increase the publicity and promotion of the platform, improve the awareness and usage rate of the platform among enterprises and the public, and promote the digital and intelligent development of real estate government management.

[0097] To sum up, the government administration method and platform for real estate first opens the UART receiving end. When the algorithm module is connected to the UART interface of the government administration device, the algorithm module can automatically detect the access of the peripheral device, and then start to trigger the automatic identification algorithm, by detecting which of the two data lines of the current device has sending data. If one of the data lines has sending data, then this data line is the sending line TX, and the other data line is the receiving line RX. If the current device requires the user to input instructions before it responds, then this algorithm module will test the two data lines of the government administration device with the input instructions respectively. If a response is obtained, this data line is the receiving line RX signal, and the other has a response. The data line obtained is the transmission line TX signal, and then the algorithm automatically corrects the UART receiving line RX and the transmission line TX interface, the UART interface communicates normally, and then turns off the UART interrupt function, the debugging state machine in the integrated circuit chip enters the waiting state, waiting for the host computer instruction, and then when the UART receiving end of the integrated circuit chip receives the host computer instruction, the UART receiving FIFO state changes, triggering the debugging state machine to actively read the instruction, and determine whether it is a valid instruction header. If it is a valid instruction header, start operating the read address and data, and continue to read the subsequent data of the UART receiving end. If it is not a valid command, the data will be discarded and the receiving end will continue to be monitored. The subsequent data includes: operation address, The number of operations and the bus data width byte valid information. If the instruction is judged to be a write operation, the debug state machine initiates an internal bus write operation. The single-address mode of the state machine write operation only writes the internal bus once. The address increment mode will also accumulate the address and continue to write the subsequent data to the accumulated address. If the length of the written data is equal to the number of operations multiplied by the number of 1s in the byte valid information, the write operation is stopped, the state machine returns to the initial state, and waits for new instructions. If it is judged to be a read operation, the debug state machine will initiate an internal bus read operation. The single-address mode of the read operation will only read the internal bus once and send the read data through UART. The address increment mode will also accumulate the address and continue to initiate internal bus read operations. , and sends the read data through UART. If the length of the data read from the internal bus is equal to the number of operations multiplied by the number of 1s in the valid information of the byte, the read operation is stopped, the state machine returns to the initial state, and waits for new instructions. Through the command word and the operation flow of steps S1 to S4, the device register mounted on the internal bus of the embedded system can be edited to realize the debugging of the internal bus path of the embedded system. The specific process of editing the device register mounted on the internal bus of the embedded system is: the host computer sends a read command to start the debugging state machine to initiate a bus read operation, read the current value of the device register and return it to the host computer, or the host computer sends a write command to start the debugging state machine to initiate a bus write operation, and write the data sent by the host computer into the device register.

[0098] It should be noted that, in this document, relational terms such as first and second, etc. are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0099] Although 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 present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for government administration of real estate, characterized in that: The steps include: S1. Structural processing of the text data of real estate: establishing mapping pairs between table names and classification types, and between field names and words in the text data; S2, using the structured real estate text data to train the table field matching model: using the structured real estate text data as the first text data, using the first mapping pair and the second mapping pair in the first text data to train the table field matching model, and outputting a total mapping vector; S3. Match the field name with the text data; specifically including: S31. After receiving the host computer instruction, the debugging state machine is triggered to actively read the instruction, determine whether it is a valid instruction header, and extract the field names of all master data entity type definition tables; S32, inputting the second text data into a table field matching model, and generating a first field name sequence from the field names of the master data entity type definition table obtained by outputting the table field matching model; S33, generating a second field name sequence from the extracted field names of the master data entity type definition table, placing the first field name sequence and the second field name sequence into a clustering space, clustering the first field name sequence with the second field name sequence as the clustering center, and generating a plurality of first field name clustering sets; S34, marking a target text classification type for the second text data associated with the first field name in the first field name clustering set, wherein the marked target text classification type is the target text classification type mapped by the master data entity type definition table associated with the second field name sequence in the first field name clustering set; S4. Determine the transmission line TX and the receiving line RX of the government administration device: determine the transmission line TX and the receiving line RX according to the data transmission and receiving capabilities of the UART port of the government administration device; S5. Perform a send operation or a write operation based on the internal bus read and write operation.

2. A method for government administration of real estate according to claim 1, characterized in that: In step S12, an independence assumption is made based on the naive Bayes classification, assuming that each feature word in the text is independent under a given category. The formula is as follows: Given a text D and a category C i , Bayes’ theorem can be expressed as: Among them, P(D|C i ) is in Category C i The likelihood probability of the following text D, P(C i ) is Category C i The prior probability of , P(D) is the marginal probability of text D.

3. A method for government administration of real estate according to claim 2, characterized in that: In step S32, training the table field matching model specifically means that the parameter of the table field matching model is cosine similarity, the cosine similarity of each mapping pair is obtained by calculation, and the calculated cosine similarity is assigned as the parameter of the table field matching model, and the calculation formula satisfies: Among them, V i , W i , W i and Q i are the table name, field name, classification type and high-dimensional vector of words in the text data of the ith master data entity definition table; S(V i , U i ) indicates V i and U i The cosine similarity of T(W i , Q i ) indicates W i and Q i The cosine similarity of V i ·U j and W i Q i represents the dot product of vectors, |V i |、|U j |、|W i |and|Q i | represents the norm of a vector.

4. A method for government administration of real estate according to claim 1, characterized in that: Step S1 specifically includes: S11, determining the independent classification type of the real estate text data, extracting the master data entity type definition table, the master data entity type definition table including the table name and the field name; S12. Establishing mapping pairs: establishing a first mapping pair between the table name and the classification type, and establishing a second mapping pair between the field name and the word in the text data.

5. A method for government administration of real estate according to claim 1, characterized in that: Step S4 specifically includes: S41, determining a UART port capable of sending data to the government administration device as a transmission line TX of the government administration device, and determining another UART port as a receiving line RX; S42. If an access command is sent to one of the UART ports of the government administration device and a response is received from another port, the port is determined as a receiving line RX and the other port is determined as a transmitting line TX.

6. A method for government administration of real estate according to claim 1, characterized in that: Step S5 specifically includes: S51. After receiving the operation instruction from the host computer, if it is judged to be a read operation, the state machine initiates an internal bus read operation. The single address mode of the read operation reads the internal bus once, and sends the read data out through the transmit line TX; if it is judged to be a write operation, the state machine initiates an internal bus write operation. The single address mode of the write operation reads the internal bus once, and writes the data received by the receive line RX into the memory of the state machine; S52, after completing an internal bus read operation and sending the read data, perform address accumulation, continue to initiate an internal bus read operation, and continue to send the read data through the transmission line TX; S53. If the length of the data read from the internal bus reaches the number of operations multiplied by the number of 1s in the byte valid information, the read operation is stopped, and the state machine returns to the initial state to wait for new instructions.

7. A method for government administration of real estate according to claim 1, characterized in that: The UART interface of the government administration device has a selectable level conversion function, and the government administration device has the function of actively outputting UART data and adjusting the receiving line RX and the transmitting line TX signals; The government administration equipment responds according to the input instructions, and adjusts the receiving line RX and the transmitting line TX signals according to the user input instructions to make the UART communication normal.

8. A method for government administration of real estate according to claim 1, characterized in that: Before step S51, it also includes: judging whether the input terminal for reading UART data is connected to the output terminal with command output function, and when the judgment result is "yes", executing the subsequent steps, and when the judgment result is "no", not executing the subsequent steps.

9. A method for government administration of real estate according to claim 1, characterized in that: In step S32, the table field matching model uses a recurrent neural network to learn complex features and context information through a multi-layer neural network. The basic form of the recurrent neural network includes: h t =σ(W hx x t +W ht-1 h t-1 +b t ) y t =Softmax(W ht h t +b y ) Among them, x t Represents the word vector in the text data of the tth order; W hx Represents the initial weight of the word vector in the text data; W ht-1 and W ht Respectively represent the cycle weights of the model at the t-1th order and the tth order; b t represents the bias at the tth order; h t-1 and h t Respectively represent the hidden state functions of the t-1th and tth order, y t is the predicted output; b y Represents the output bias; σ is the activation function, Softma x Indicates multi-task classification output of the calculation results.

Images