Urban and rural construction digital archives signature management system

Abstract

The present application relates to the field of intelligent management, and specifically to a digital archive signature management system for urban and rural construction. It uses natural language processing technology based on deep learning to perform semantic analysis on the search conditions input by the user, dig out the semantic feature expression of the search conditions, and extract alternative urban and rural construction digital archives from the archives, and perform keyword recognition and semantic embedding coding on them to capture the key semantic information of the alternative urban and rural construction digital archives, and then based on the cross-domain query matching characteristics between the search conditions and the alternative urban and rural construction digital archives, intelligently identify whether the alternative urban and rural construction digital archives match the search conditions, which can effectively improve the accuracy and relevance of urban and rural construction digital archive retrieval, and provide users with more convenient and accurate archive query services.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and specifically to a digital archive signature management system for urban and rural construction. Background Art

[0002] In the field of urban and rural construction, a large amount of archival materials need to be managed and retrieved efficiently and safely. The establishment of a digital archive system has greatly improved the efficiency and security of archive management, and also provided more convenient and efficient information support for the planning, design, construction and maintenance of urban and rural construction. However, in the process of archive utilization, how to efficiently retrieve relevant materials and ensure the accuracy and security of retrieval results is still an important challenge facing the current digital archive system.

[0003] Traditional archive retrieval methods mainly rely on keyword matching technology. Although they can meet the user's query needs to a certain extent, due to the lack of in-depth semantic understanding of query intent and digital archives, they can often only return results that literally match the query keywords, resulting in low accuracy and relevance of retrieval results, making it difficult to meet users' needs for efficient and accurate retrieval of urban and rural construction digital archives. In addition, urban and rural construction digital archives often contain a large amount of professional terms and industry specification information, which makes retrieval methods based on simple keyword matching even more difficult to cope with.

[0004] Therefore, an optimized urban and rural construction digital archive retrieval solution is needed to solve the above technical problems. Summary of the invention

[0005] This application is made in consideration of the above problems. One purpose of this application is to provide a digital archive signature management system for urban and rural construction, which includes:

[0006] A user search information acquisition module is used to obtain search conditions input by the user;

[0007] A retrieval condition semantic encoding module, used for semantically encoding the retrieval condition to obtain a retrieval condition semantic encoding feature vector;

[0008] The alternative digital archive extraction module is used to extract the alternative urban and rural construction digital archives from the archives;

[0009] A digital archive keyword semantic coding module is used to perform semantic embedding coding on the keywords in the candidate urban and rural construction digital archives to obtain a set of semantic embedding coding feature vectors of the candidate urban and rural construction archive keywords;

[0010] The alternative digital archive matching analysis module is used to determine whether to return the alternative urban and rural construction digital archive as a query result based on the cross-domain query matching characteristics between the semantic coding feature vector of the retrieval condition and the set of semantic embedding coding feature vectors of the alternative urban and rural construction archive keywords.

[0011] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the search condition semantic coding module is used to:

[0012] The retrieval condition is input into a retrieval condition semantic encoder based on the Bert model for semantic encoding to obtain the retrieval condition semantic encoding feature vector.

[0013] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the digital archive keyword semantic encoding module includes:

[0014] A keyword extraction unit, used to extract keywords of the candidate urban and rural construction digital archives to obtain a set of candidate urban and rural construction archive keywords;

[0015] The keyword semantic embedding coding unit is used to use the semantic embedding matrix to perform semantic embedding coding on each candidate urban and rural construction archives keyword in the set of candidate urban and rural construction archives keywords to obtain a set of semantic embedding coding feature vectors of the candidate urban and rural construction archives keywords.

[0016] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the keyword extraction unit includes:

[0017] A text extraction subunit, used for extracting text parts from the candidate urban and rural construction digital archives to obtain the text parts of the candidate urban and rural construction archives;

[0018] The document keyword extraction subunit is used to input the text part of the candidate urban and rural construction archives into the document keyword extractor based on the AIGC model to obtain a set of candidate urban and rural construction archives keywords.

[0019] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the candidate digital archive matching analysis module includes:

[0020] A cross-domain query matching analysis unit, used for inputting the set of the retrieval condition semantic encoding feature vector and the candidate urban and rural construction archives keyword semantic embedding encoding feature vector into a cross-domain attention gate query encoder based on key matrix local association enhancement to obtain a retrieval condition-candidate urban and rural construction archives cross-domain query matching feature vector;

[0021] The retrieval result generating unit is used to input the retrieval condition-alternative urban and rural construction archive cross-domain query matching feature vector into the classifier-based retrieval result generator to obtain the retrieval result, and the retrieval result is used to indicate whether to return the alternative urban and rural construction digital archive as the query result.

[0022] For example, according to the urban and rural construction digital archive signature management system of an embodiment of the present application, the cross-domain query matching analysis unit includes:

[0023] A linear transformation subunit is used to perform a linear transformation on each candidate urban and rural construction archives keyword semantic embedding coding feature vector in the set of candidate urban and rural construction archives keyword semantic embedding coding feature vectors to obtain an initial key matrix, wherein each row vector in the initial key matrix is ​​a candidate urban and rural construction archives keyword semantic embedding coding feature vector after each linear transformation;

[0024] A feature distribution difference measurement subunit, used for calculating the feature distribution difference energy coefficients between the retrieval condition semantic encoding feature vector and each row vector in the initial key matrix to obtain a set of feature distribution difference energy coefficients;

[0025] A key matrix updating subunit, configured to compare each feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients with a predetermined threshold value, and in response to the feature distribution difference energy coefficient being less than the predetermined threshold value, using a mean vector between a row vector corresponding to the feature distribution difference energy coefficient and the retrieval condition semantic encoding feature vector as an updated row vector to obtain an updated key matrix;

[0026] A value vector calculation subunit, used for processing the retrieval condition semantic encoding feature vector using a value embedding matrix to obtain a value vector;

[0027] A cross-domain query subunit is used to input the retrieval condition semantic encoding feature vector as a query vector, each row vector in the update key matrix as a key vector, and the value vector into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of retrieval condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vectors;

[0028] The global matching aggregation subunit is used to calculate the positional mean vector of the sequence of the search condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vector to obtain the search condition-alternative urban and rural construction archives cross-domain query matching feature vector.

[0029] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the feature distribution difference measurement subunit includes:

[0030] A difference feature extraction secondary subunit is used to extract the difference features between the retrieval condition semantic coding feature vector and each row vector in the initial key matrix to obtain a set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors;

[0031] The energy coefficient calculation secondary sub-unit is used to calculate the energy coefficient of each retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector in the set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors to obtain the set of feature distribution difference energy coefficients.

[0032] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the difference feature extraction secondary subunit is used to:

[0033] Calculate the position point addition, position point subtraction and position point multiplication between the retrieval condition semantic encoding feature vector and the row vector to obtain a first matching fusion result, a second matching fusion result and a third matching fusion result;

[0034] After the first matching fusion result, the second matching fusion result and the third matching fusion result are cascaded into a multi-scale semantic matching feature vector of retrieval conditions-alternative urban and rural construction archives keywords, one-dimensional convolution processing and maximum pooling processing are performed on them to obtain the semantic association difference feature vector of the retrieval conditions-alternative urban and rural construction archives keywords.

[0035] For example, according to the urban and rural construction digital archive signature management system of the embodiment of the present application, the feature distribution difference measurement subunit is used to:

[0036] Calculate the weighted sum between the feature variance and the bias parameter of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector to obtain the first intermediate term, and calculate the square of the difference between the maximum eigenvalue and the feature mean of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector, the variance of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector, and the bias term to obtain the second intermediate term;

[0037] The first intermediate term is divided by the second intermediate term to obtain the characteristic distribution difference energy coefficient.

[0038] The present application has at least the following technical effects: compared with the prior art, the urban and rural construction digital archive signature management system of the embodiment of the present application adopts natural language processing technology based on deep learning to perform semantic analysis on the search conditions input by the user, and mines the semantic feature expression of the search conditions. At the same time, it extracts alternative urban and rural construction digital archives from the archives, and performs keyword recognition and semantic embedding coding on them to capture the key semantic information of the alternative urban and rural construction digital archives, and then based on the cross-domain query matching features between the search conditions and the alternative urban and rural construction digital archives, it intelligently identifies whether the alternative urban and rural construction digital archives matches the search conditions, which can effectively improve the accuracy and relevance of urban and rural construction digital archive retrieval, and provide users with more convenient and accurate archive query services. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0040] Figure 1 It shows a schematic diagram of the application architecture of the urban and rural construction digital archive signature management system in the embodiment of the present application;

[0041] Figure 2 It shows a schematic diagram of the structure of the urban and rural construction digital archive signature management system in the embodiment of the present application;

[0042] Figure 3 It shows a schematic diagram of the structure of the digital archive keyword semantic coding module of the urban and rural construction digital archive signature management system in the embodiment of the present application;

[0043] Figure 4 A schematic diagram of the structure of an alternative digital archive matching analysis module of the urban and rural construction digital archive signature management system in an embodiment of the present application is shown;

[0044] Figure 5 A flowchart of a method for managing digital archive signatures for urban and rural construction in an embodiment of the present application is shown;

[0045] Figure 6 It shows the application scenario diagram of the urban and rural construction digital archive signature management system in the embodiment of the present application;

[0046] Figure 7 A schematic diagram of archive retrieval of the urban and rural construction digital archive signature management system in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0048] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0049] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0050] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0051] Figure 1 A schematic diagram of the application architecture of the urban and rural construction digital archive signature management system in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0052] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of local area network (LAN), metropolitan area network (MAN), wide area network (WAN), mobile, wired or wireless network, private network or virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0053] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0054] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the urban and rural construction digital archive signature management system in the embodiment of the present application.

[0055] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the urban and rural construction digital archive signature management methods in the embodiments of the present application according to the obtained program instructions.

[0056] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0057] The following is a non-restrictive description of the urban and rural construction digital archive signature management system provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of this application.

[0058] In response to the above technical problems, the technical concept of the present application is to use natural language processing technology based on deep learning to perform semantic analysis on the search conditions input by the user, to explore the semantic feature expression of the search conditions, and at the same time extract alternative urban and rural construction digital archives from the archives, and perform keyword recognition and semantic embedding coding on them to capture the key semantic information of the alternative urban and rural construction digital archives, and then based on the cross-domain query matching features between the search conditions and the alternative urban and rural construction digital archives, intelligently identify whether the alternative urban and rural construction digital archives matches the search conditions, which can effectively improve the accuracy and relevance of urban and rural construction digital archive retrieval, and provide users with more convenient and accurate archive query services.

[0059] Based on this, Figure 2 The structural schematic diagram of the urban and rural construction digital archive signature management system 800 in the embodiment of the present application is shown. The urban and rural construction digital archive signature management system 800 includes: a user search information acquisition module 810, which is used to obtain the search conditions input by the user; a search condition semantic coding module 820, which is used to semantically encode the search conditions to obtain the search condition semantic coding feature vector; an alternative digital archive extraction module 830, which is used to extract alternative urban and rural construction digital archives from the archives; a digital archive keyword semantic coding module 840, which is used to semantically embed the keywords in the alternative urban and rural construction digital archives to obtain a set of alternative urban and rural construction archive keyword semantic embedding coding feature vectors; an alternative digital archive matching analysis module 850, which is used to determine whether to return the alternative urban and rural construction digital archive as a query result based on the cross-domain query matching feature between the search condition semantic coding feature vector and the set of alternative urban and rural construction archive keyword semantic embedding coding feature vectors.

[0060] Specifically, in the technical solution of the present application, first, the search conditions input by the user are obtained. It should be understood that the search conditions input by the user directly reflect their query intentions and needs, and are the starting point of the entire search process. In the technical solution of the present application, in order to deeply understand the semantic meaning of the search conditions, the Bert model is used as a search condition semantic encoder to perform contextual semantic encoding on the search conditions, so as to utilize the bidirectional contextual semantic understanding ability of the Bert model to capture the contextual semantic dependency relationship between the words in the search conditions, fully understand the user's query intention, and generate a search condition semantic encoding feature vector with high semantic expression ability.

[0061] Correspondingly, the retrieval condition semantic encoding module 820 is used to: input the retrieval condition into a retrieval condition semantic encoder based on the Bert model for semantic encoding to obtain the retrieval condition semantic encoding feature vector.

[0062] Next, extract the candidate urban and rural construction digital archives from the archives. It should be understood that urban and rural construction digital archives usually exist in the form of structured or unstructured data, such as design plans, construction logs, acceptance reports, etc. Among them, the text part in the archives usually contains the core information of the archives, such as project name, construction location, construction content, time node, etc., which is the main target of user retrieval and query. Therefore, in the technical solution of the present application, first, extract the text part from the candidate urban and rural construction digital archives to filter out non-critical images, tables and other information, obtain a text representation with a unified format, and thus obtain the text part of the candidate urban and rural construction archives.

[0063] Then, in order to further narrow the scope of subsequent semantic matching and query, and reduce unnecessary calculation and time consumption. This application further uses a document keyword extractor based on the AIGC model to extract keywords from the text part of the candidate urban and rural construction archives. Among them, the AIGC model is based on natural language processing technology, which can deeply understand the text content of the document, identify the key concepts and themes therein, and thus generate a set of candidate urban and rural construction archive keywords. Compared with the traditional keyword extraction method based on word frequency statistics, the AIGC model can better consider the context and topic relevance, and improve the accuracy and representativeness of keyword extraction.

[0064] Similarly, in order to convert the text representation of each candidate urban and rural construction archives keyword into a continuous high-dimensional semantic space feature representation, the present application further uses a semantic embedding matrix to perform semantic embedding encoding on each candidate urban and rural construction archives keyword in the set of candidate urban and rural construction archives keywords to capture the semantic meaning of each keyword and generate a set of semantic embedding encoding feature vectors of the candidate urban and rural construction archives keywords, so that each keyword has a comparable representation in the high-dimensional semantic space. In an embodiment of the present application, the semantic embedding matrix is ​​generated by a pre-trained Word2Vec model.

[0065] Accordingly, if Figure 3 As shown, the digital archive keyword semantic coding module 840 includes: a keyword extraction unit 841, which is used to extract the keywords of the alternative urban and rural construction digital archives to obtain a set of alternative urban and rural construction archive keywords; a keyword semantic embedding coding unit 842, which is used to use a semantic embedding matrix to perform semantic embedding coding on each alternative urban and rural construction archive keyword in the set of alternative urban and rural construction archive keywords to obtain a set of semantic embedding coding feature vectors of the alternative urban and rural construction archive keyword.

[0066] Among them, the keyword extraction unit 841 includes: a text extraction subunit, used to extract the text part from the alternative urban and rural construction digital archives to obtain the alternative urban and rural construction archive text part; a document keyword extraction subunit, used to input the alternative urban and rural construction archive text part into a document keyword extractor based on the AIGC model to obtain a set of alternative urban and rural construction archive keywords.

[0067] Furthermore, considering that in traditional feature query matching, usually only the correlation between the query vector and each key vector is paid attention to, while the semantic correlation between each key vector is ignored. Based on this, in the technical solution of the present application, a cross-domain attention gate query encoder based on the local correlation enhancement of the key matrix is ​​introduced to match and analyze the set of the semantic encoding feature vector of the retrieval condition and the semantic embedding encoding feature vector of the candidate urban and rural construction archives keywords, and the implicit correlation information contained between the semantic embedding encoding feature vectors of each candidate urban and rural construction archives keyword is used to improve the accuracy of the cross-domain query matching encoding. Specifically, the cross-domain attention gate query encoder first linearly transforms each candidate urban and rural construction archives keyword semantic embedding encoding feature vector to adjust its feature representation dimension so that it is aligned with the feature dimension of the retrieval condition semantic encoding feature vector to construct an initial key matrix. Then, the retrieval condition semantic encoding feature vector is used as a common soft anchor reference point, and the local correlation strength contained between each key vector is revealed by calculating the feature distribution difference energy coefficient between the retrieval condition semantic encoding feature vector and each key vector in the initial key matrix. Secondly, for the key vector corresponding to the feature distribution difference energy coefficient less than the predetermined threshold, use the mean vector of the key vector and the retrieval condition semantic encoding feature vector to update its feature representation, so as to further strengthen the semantic feature representation of archive keywords that are more similar to the retrieval condition semantics, generate an updated key matrix, and improve the pertinence and matching quality of the key matrix. Furthermore, based on the retrieval condition semantic encoding feature vector, construct a query vector and a value vector, and use each row vector in the updated key matrix as a key vector. Through the self-attention calculation of the Transformer structure, cross-domain information integration is achieved to obtain a cross-domain query attention vector between the retrieval condition and each archive keyword. Finally, by calculating the mean of each cross-domain query attention vector by position, the matching information of all keywords is integrated to generate a retrieval condition-alternative urban and rural construction archive cross-domain query matching feature vector.

[0068] Next, the search condition-alternative urban and rural construction archives cross-domain query matching feature vector is input into the classifier-based search result generator to obtain the search result, and the search result is used to indicate whether to return the alternative urban and rural construction digital archives as the query result. It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0069] Accordingly, if Figure 4 As shown, the alternative digital archive matching analysis module 850 includes: a cross-domain query matching analysis unit 851, which is used to input the set of the retrieval condition semantic encoding feature vector and the alternative urban and rural construction archive keyword semantic embedding encoding feature vector into a cross-domain attention gate query encoder based on key matrix local association enhancement to obtain a retrieval condition-alternative urban and rural construction archive cross-domain query matching feature vector; a retrieval result generation unit 852, which is used to input the retrieval condition-alternative urban and rural construction archive cross-domain query matching feature vector into a classifier-based retrieval result generator to obtain a retrieval result, and the retrieval result is used to indicate whether to return the alternative urban and rural construction digital archive as a query result.

[0070] Among them, the cross-domain query matching analysis unit 851 includes: a linear transformation subunit, which is used to perform linear transformation on each alternative urban and rural construction archives keyword semantic embedding coding feature vector in the set of alternative urban and rural construction archives keyword semantic embedding coding feature vectors to obtain an initial key matrix, wherein each row vector in the initial key matrix is ​​the alternative urban and rural construction archives keyword semantic embedding coding feature vector after each linear transformation; a feature distribution difference measurement subunit, which is used to calculate the feature distribution difference energy coefficient between the retrieval condition semantic coding feature vector and each row vector in the initial key matrix to obtain a set of feature distribution difference energy coefficients; a key matrix updating subunit, which is used to compare each feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients with a predetermined threshold, and in response to the feature distribution difference energy coefficient being less than the predetermined threshold, using the feature distribution difference energy coefficient as the retrieval condition semantic coding feature vector. The row vector corresponding to the distribution difference energy coefficient and the mean vector between the retrieval condition semantic coding feature vector are used as the updated row vector to obtain the updated key matrix; the value vector calculation subunit is used to use the value embedding matrix to process the retrieval condition semantic coding feature vector to obtain the value vector; the cross-domain query subunit is used to input the retrieval condition semantic coding feature vector as the query vector, each row vector in the updated key matrix as the key vector and the value vector into the cross-domain attention gate query module based on the Transformer structure to obtain the sequence of retrieval condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vector; the global matching aggregation subunit is used to calculate the positional mean vector of the sequence of retrieval condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vector to obtain the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector.

[0071] Specifically, the feature distribution difference measurement subunit includes: a difference feature extraction secondary subunit, which is used to respectively extract the difference features between the retrieval condition semantic encoding feature vector and each row vector in the initial key matrix to obtain a set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors; an energy coefficient calculation secondary subunit, which is used to calculate the energy coefficient of each retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector in the set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors to obtain the set of feature distribution difference energy coefficients.

[0072] More specifically, the difference feature extraction secondary subunit is used to: calculate the point-by-point addition, point-by-point subtraction and point-by-point multiplication between the retrieval condition semantic encoding feature vector and the row vector to obtain a first matching fusion result, a second matching fusion result and a third matching fusion result; after cascading the first matching fusion result, the second matching fusion result and the third matching fusion result into a retrieval condition-alternative urban and rural construction archives keyword multi-scale semantic matching feature vector, perform one-dimensional convolution processing and maximum value pooling processing on it to obtain the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector.

[0073] Specifically, the feature distribution difference measurement subunit is used to: calculate the weighted sum between the feature variance and the bias parameter of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector to obtain a first intermediate term, and calculate the square of the difference between the maximum eigenvalue and its feature mean of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector, the variance of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector and the bias term to obtain a second intermediate term; calculate the first intermediate term divided by the second intermediate term to obtain the feature distribution difference energy coefficient.

[0074] In one example, the cross-domain query matching analysis unit 851 is used to: process the set of the retrieval condition semantic encoding feature vector and the candidate urban and rural construction archives keyword semantic embedding encoding feature vector using the following cross-domain query matching formula to obtain the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, wherein the cross-domain query matching formula is:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] in, represents the set of semantic embedding coding feature vectors of the candidate urban and rural construction archive keywords, , , and They are respectively the first, second, and third in the set of semantic embedding coding feature vectors of the candidate urban and rural construction archive keywords. and The semantic embedding encoding feature vectors of candidate urban and rural construction archive keywords, The value of is the number of feature vectors in the set of the candidate urban and rural construction archives keyword semantic embedding coding feature vectors, is the linear transformation matrix, is the bias matrix, represents the initial bond matrix, , , and are the first, second, and and row vectors, i.e., the semantic embedding encoding feature vectors of candidate urban and rural construction archive keywords after linear transformation, is a linear transformation, represents the semantic encoding feature vector of the retrieval condition, represents dot product, Indicates point addition, Indicates point reduction, Indicates cascade operation, represents the maximum pooling operation, represents a one-dimensional convolution operation, Indicates Search conditions - alternative urban and rural construction archives keyword semantic association difference feature vector, , and Respectively represent the The feature variance, maximum eigenvalue and feature mean of the semantic association difference feature vector of the search conditions - alternative urban and rural construction archive keywords, is the bias parameter, is the first in the set of characteristic distribution difference energy coefficients The characteristic distribution difference energy coefficient, is the predetermined threshold, represents the updated key matrix, represents the first row vectors, is the value embedding matrix, is the bias vector, is a value vector, is the normalized exponential function, The value of is the length of the row vector in the update key matrix, represents the transpose of a vector, It is Search conditions - alternative urban and rural construction archives keyword semantic features cross-domain query attention vector, Represents the search condition-alternative urban and rural construction archives cross-domain query matching feature vector.

[0085] Preferably, considering that the set of the retrieval condition semantic coding feature vector and the set of the alternative urban and rural construction archive keyword semantic embedding coding feature vector respectively represent the semantic coding features of the retrieval condition input by the user and the keyword semantic coding features of the alternative urban and rural construction digital archives, when performing cross-domain feature gated query with key matrix local correlation enhancement, due to the local correlation enhancement cross-domain query differences caused by the differences in text coding feature distribution and semantic selectivity, it will lead to semantic query aggregation fine-grained alignment conflicts, thereby causing the loss of aggregated key information, affecting the expression effect of the retrieval condition-alternative urban and rural construction archive cross-domain query matching feature vector.

[0086] Based on this, in the preferred example, when the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector is input into the classifier-based retrieval result generator to obtain the retrieval result, the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector is optimized. Specifically, the retrieval result generating unit 852 includes: an optimization subunit, which is used to optimize the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector to obtain an optimized retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector; a classification subunit, which is used to input the optimized retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector into the classifier-based retrieval result generator to obtain the retrieval result.

[0087] Among them, the optimization subunit is used to: cluster all feature values ​​of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector based on the distance between feature values, such as the difference absolute value distance, and determine the ratio of the feature number of cluster features to the feature number of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector; arrange the cluster features into a retrieval condition-alternative urban and rural construction archives cross-domain query matching cluster vector, and divide the second norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching cluster vector by the second norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector to obtain a first retrieval condition-alternative urban and rural construction archives cross-domain query matching cluster related weight value; divide the first power value of the first norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching cluster vector with the ratio value as the exponent by the second power value of the first norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector with the ratio value as the exponent to obtain a second retrieval condition-alternative urban Township construction archives cross-domain query matching clustering related weight values; for each eigenvalue of the search condition-alternative urban and rural construction archives cross-domain query matching clustering vector, multiply it by the inverse of the difference between the first and second search conditions-alternative urban and rural construction archives cross-domain query matching clustering related weight values ​​to obtain the optimized eigenvalue of the search condition-alternative urban and rural construction archives cross-domain query matching clustering vector; for each eigenvalue outside the cluster in the search condition-alternative urban and rural construction archives cross-domain query matching feature vector, multiply it by the inverse of the sum of the first and second search conditions-alternative urban and rural construction archives cross-domain query matching clustering related weight values ​​to obtain the optimized out-of-class eigenvalue of the search condition-alternative urban and rural construction archives cross-domain query matching feature vector; the optimized eigenvalue of the search condition-alternative urban and rural construction archives cross-domain query matching clustering vector and the optimized out-of-class eigenvalue of the search condition-alternative urban and rural construction archives cross-domain query matching feature vector form the optimized search condition-alternative urban and rural construction archives cross-domain query matching feature vector.

[0088] Specifically, in one example, the optimization subunit is used to optimize the search condition-alternative urban and rural construction archives cross-domain query matching feature vector using the following optimization formula to obtain the optimized search condition-alternative urban and rural construction archives cross-domain query matching feature vector; wherein the optimization formula is:

[0089]

[0090]

[0091] in, is the search condition-selected urban and rural construction archives cross-domain query matching feature vector, is the number of features of the search condition-selected urban and rural construction archives cross-domain query matching feature vector, is the search condition - alternative urban and rural construction archives cross-domain query matching clustering vector, is the number of features of the search condition-selected urban and rural construction archives cross-domain query matching cluster vector, is the ratio value, represents the clustering feature set corresponding to the clustering vector of the search condition-alternative urban and rural construction archives cross-domain query matching, and Represents the binary norm and the uninorm of the vector respectively. Power, is the feature value of each position in the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, It is the feature value of each position in the optimized retrieval condition - the alternative urban and rural construction archive cross-domain query matching feature vector.

[0092] Here, in order to avoid the loss of key suffix semantic information relative to the original feature set as a whole due to aggregation conflict caused by the fact that the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector is based on aggregated features, the clustering ratio of the number of features of the retrieval condition-alternative urban and rural construction archives cross-domain query matching clustering vector relative to the number of features of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector is used as a decision function to perform an adversarial decision on the set absolute representation of the retrieval condition-alternative urban and rural construction archives cross-domain query matching clustering vector and the one-norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, and respectively compare it with the retrieval condition-alternative urban and rural construction archives cross-domain query matching clustering vector. Positive and negative interactions are performed with the clustering intrinsic conflict representation of the two norm of the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector to construct a solid alignment guardrail with the original feature set as a whole based on the aggregated features of the optimized retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, thereby achieving the mitigation of the intention of harmful information loss based on the aggregated risk transferability of the optimized retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, improving the expression effect of the optimized retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector, thereby improving the accuracy of the retrieval result obtained by inputting the retrieval condition-alternative urban and rural construction archives cross-domain query matching feature vector into the classifier-based retrieval result generator.

[0093] Based on the above embodiments, see Figure 5 As shown in FIG. 1 , it is a flow chart of a method for managing the signature of urban and rural construction digital archives in an embodiment of the present application. For example, the method for managing the signature of urban and rural construction digital archives can be executed by a server, which can be Figure 1The server 100 shown in FIG. Figure 5 As shown, the urban and rural construction digital archive signature management method according to the embodiment of the present application includes the following steps: S510, obtaining the search conditions input by the user; S520, semantically encoding the search conditions to obtain the search condition semantic encoding feature vector; S530, extracting the alternative urban and rural construction digital archives from the archive; S540, semantically embedding the keywords in the alternative urban and rural construction digital archives to obtain a set of alternative urban and rural construction archive keyword semantic embedding encoding feature vectors; S550, based on the cross-domain query matching characteristics between the search condition semantic encoding feature vector and the set of alternative urban and rural construction archive keyword semantic embedding encoding feature vectors, determining whether to return the alternative urban and rural construction digital archive as a query result.

[0094] Specifically, the specific operations of each step in the above-mentioned urban and rural construction digital archive signature management method have been referred to above. Figures 2 to 4 It has been introduced in detail in the description of the urban and rural construction digital archive signature management system 800, and therefore, its repeated description will be omitted.

[0095] Figure 6 FIG. 1 is an application scenario diagram of the urban and rural construction digital archive signature management system according to an embodiment of the present application. Figure 6 As shown, in this application scenario, first, the search conditions entered by the user are obtained (for example, Figure 6 D1 shown in the figure) and extract alternative urban and rural construction digital archives from the archives (for example, Figure 6 Then, the search condition and the candidate urban and rural construction digital archive are input into a server that is deployed with an urban and rural construction digital archive signature management algorithm (for example, Figure 6 In S) shown in , the server can use the urban and rural construction digital archive signature management algorithm to process the search conditions and the alternative urban and rural construction digital archives to obtain a search result indicating whether to return the alternative urban and rural construction digital archives as a query result.

[0096] It should be understood that the system supports various forms of use of physical and digital archive resources, including archive retrieval, archive browsing, archive borrowing, feedback on utilization effects, archive compilation and research, open use settings, copy management and archive statistics. Figure 7As shown, the system can provide a friendly search page similar to "Baidu style", and can provide different search strategies and methods for catalogs and full texts. It supports retrieving archives by attributes, and has multiple search methods such as precise search, fuzzy search, advanced combination search, screening search, related search, in-depth secondary search, full-text search, catalog search, etc. Among them, full-text search is based on archive entry information and text content, with a larger search range and more accuracy; catalog search is suitable for professionals familiar with archive business, which can quickly locate the target and save time. In the search results, keywords are marked in red, and the specific information of the entry can be viewed and collected, and provided for use according to the authority. If it exceeds the scope of authority, an authorization application must be submitted and approved before it can be provided for use.

[0097] The system supports browsing of electronic files or other digital resources in common formats, authorizes downloading functions according to permissions, and provides playback of multimedia files in common formats.

[0098] The system supports the borrowing application, approval and authorization of electronic archives, and has the functions of borrowing reservation, reminder and return of physical archives. The approval process must comply with the archive utilization management system, and the borrowing content must comply with security and confidentiality regulations.

[0099] According to the borrowing process, the borrower submits a borrowing application form. After approval, the administrator can register the file borrowing, as well as subsequent operations such as return and renewal. For borrowing forms that are about to expire or have expired, the administrator can remind and urge them to return, and the borrower can receive corresponding reminders through emails, messages, etc. Administrators are supported to configure borrowing rules for different types of files: for example, whether to allow overdue borrowers to continue borrowing, the system default number of days that can be borrowed, etc.

[0100] Applications for borrowing archives are reviewed and approved separately according to cross-departmental and cross-archive requirements. The applicant's archive administrator submits a request for physical borrowing or electronic original use to the designated unit. After receiving the application, the provider's archive administrator selects the corresponding archive item in the system and provides it for use after approval by the relevant leaders of the unit: for borrowing physical archives, the provider's archive administrator needs to associate the application form to create a borrowing form, register and review it, and then provide the physical archive to the applicant's archive administrator; for the use of electronic archives, the applicant's archive administrator can attach the download content to the application form and set a watermark mark.

[0101] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiment of the present application. In some possible implementations, the electronic device in the embodiment of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the urban and rural construction digital archive signature management method in the above embodiment when executing the program.

[0102] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0103] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the urban and rural construction digital archive signature management method according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0104] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the urban and rural construction digital archive signature management method according to the embodiment of the present application.

[0105] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0106] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0107] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0108] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0109] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A digital archive signature management system for urban and rural construction, characterized in that: include: A user search information acquisition module is used to obtain search conditions input by the user; A retrieval condition semantic encoding module, used for semantically encoding the retrieval condition to obtain a retrieval condition semantic encoding feature vector; The alternative digital archive extraction module is used to extract the alternative urban and rural construction digital archives from the archives; A digital archive keyword semantic coding module is used to perform semantic embedding coding on the keywords in the candidate urban and rural construction digital archives to obtain a set of semantic embedding coding feature vectors of the candidate urban and rural construction archive keywords; The candidate digital archive matching analysis module is used to determine whether to return the candidate urban and rural construction digital archive as a query result based on the cross-domain query matching characteristics between the retrieval condition semantic coding feature vector and the set of candidate urban and rural construction archive keyword semantic embedding coding feature vectors; The retrieval condition semantic encoding module is used to: Inputting the search condition into a search condition semantic encoder based on the Bert model for semantic encoding to obtain a search condition semantic encoding feature vector; The digital archive keyword semantic encoding module includes: A keyword extraction unit, used to extract keywords of the candidate urban and rural construction digital archives to obtain a set of candidate urban and rural construction archive keywords; A keyword semantic embedding coding unit, used for performing semantic embedding coding on each candidate urban and rural construction archives keyword in the set of candidate urban and rural construction archives keywords using a semantic embedding matrix to obtain a set of semantic embedding coding feature vectors of the candidate urban and rural construction archives keywords; The keyword extraction unit comprises: A text extraction subunit, used for extracting text parts from the candidate urban and rural construction digital archives to obtain the text parts of the candidate urban and rural construction archives; A document keyword extraction subunit, used for inputting the text part of the candidate urban and rural construction archives into a document keyword extractor based on the AIGC model to obtain a set of candidate urban and rural construction archives keywords; The candidate digital archive matching analysis module includes: A cross-domain query matching analysis unit, used for inputting the set of the retrieval condition semantic encoding feature vector and the candidate urban and rural construction archives keyword semantic embedding encoding feature vector into a cross-domain attention gate query encoder based on key matrix local association enhancement to obtain a retrieval condition-candidate urban and rural construction archives cross-domain query matching feature vector; A search result generating unit, used for inputting the search condition-candidate urban and rural construction archives cross-domain query matching feature vector into a classifier-based search result generator to obtain a search result, wherein the search result is used to indicate whether to return the candidate urban and rural construction digital archives as a query result; The cross-domain query matching analysis unit includes: A linear transformation subunit is used to perform a linear transformation on each candidate urban and rural construction archives keyword semantic embedding coding feature vector in the set of candidate urban and rural construction archives keyword semantic embedding coding feature vectors to obtain an initial key matrix, wherein each row vector in the initial key matrix is ​​a candidate urban and rural construction archives keyword semantic embedding coding feature vector after each linear transformation; A feature distribution difference measurement subunit, used for calculating the feature distribution difference energy coefficients between the retrieval condition semantic encoding feature vector and each row vector in the initial key matrix to obtain a set of feature distribution difference energy coefficients; A key matrix updating subunit, configured to compare each feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients with a predetermined threshold value, and in response to the feature distribution difference energy coefficient being less than the predetermined threshold value, using a mean vector between a row vector corresponding to the feature distribution difference energy coefficient and the retrieval condition semantic encoding feature vector as an updated row vector to obtain an updated key matrix; A value vector calculation subunit, used for processing the retrieval condition semantic encoding feature vector using a value embedding matrix to obtain a value vector; A cross-domain query subunit is used to input the retrieval condition semantic encoding feature vector as a query vector, each row vector in the update key matrix as a key vector, and the value vector into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of retrieval condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vectors; A global matching aggregation subunit, used to calculate the position mean vector of the sequence of the search condition-alternative urban and rural construction archives keyword semantic feature cross-domain query attention vector to obtain the search condition-alternative urban and rural construction archives cross-domain query matching feature vector; The set of the retrieval condition semantic coding feature vector and the candidate urban and rural construction archives keyword semantic embedding coding feature vector is processed by the following cross-domain query matching formula to obtain the retrieval condition-candidate urban and rural construction archives cross-domain query matching feature vector, wherein the cross-domain query matching formula is: in, represents the set of semantic embedding coding feature vectors of the candidate urban and rural construction archive keywords, , , and They are the first, second, and third in the set of semantic embedding encoding feature vectors of the candidate urban and rural construction archive keywords. and The semantic embedding encoding feature vectors of candidate urban and rural construction archive keywords, The value of is the number of feature vectors in the set of the candidate urban and rural construction archives keyword semantic embedding coding feature vectors, is the linear transformation matrix, is the bias matrix, represents the initial bond matrix, , , and are the first, second, and and row vectors, i.e., the semantic embedding encoding feature vectors of candidate urban and rural construction archive keywords after linear transformation, is a linear transformation, represents the semantic encoding feature vector of the retrieval condition, represents dot product, Indicates point addition, Indicates point reduction, Indicates cascade operation, represents the maximum pooling operation, represents a one-dimensional convolution operation, Indicates Search conditions - alternative urban and rural construction archives keyword semantic association difference feature vector, , and Respectively represent the The feature variance, maximum eigenvalue and feature mean of the semantic association difference feature vector of the search conditions - alternative urban and rural construction archive keywords, is the bias parameter, is the first in the set of characteristic distribution difference energy coefficients The characteristic distribution difference energy coefficient, is the predetermined threshold, represents the updated key matrix, represents the first row vectors, is the value embedding matrix, is the bias vector, is a value vector, is the normalized exponential function, The value of is the length of the row vector in the update key matrix, represents the transpose of a vector, It is Search conditions - alternative urban and rural construction archives keyword semantic features cross-domain query attention vector, Represents the search condition-alternative urban and rural construction archives cross-domain query matching feature vector.

2. The urban and rural construction digital archive signature management system according to claim 1 is characterized in that: The feature distribution difference measurement subunit includes: A difference feature extraction secondary subunit is used to extract the difference features between the retrieval condition semantic coding feature vector and each row vector in the initial key matrix to obtain a set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors; The energy coefficient calculation secondary sub-unit is used to calculate the energy coefficient of each retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector in the set of retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vectors to obtain the set of feature distribution difference energy coefficients.

3. The urban and rural construction digital archive signature management system according to claim 2 is characterized in that: The difference feature extraction secondary subunit is used for: Calculate the position point addition, position point subtraction and position point multiplication between the retrieval condition semantic encoding feature vector and the row vector to obtain a first matching fusion result, a second matching fusion result and a third matching fusion result; After the first matching fusion result, the second matching fusion result and the third matching fusion result are cascaded into a multi-scale semantic matching feature vector of retrieval conditions-alternative urban and rural construction archives keywords, one-dimensional convolution processing and maximum pooling processing are performed on them to obtain the semantic association difference feature vector of the retrieval conditions-alternative urban and rural construction archives keywords.

4. The urban and rural construction digital archive signature management system according to claim 3 is characterized in that: The feature distribution difference measurement subunit is used to: Calculate the weighted sum of the feature variance and the bias parameter of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector to obtain the first intermediate term, and calculate the square of the difference between the maximum eigenvalue and the feature mean of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector, the variance of the retrieval condition-alternative urban and rural construction archives keyword semantic association difference feature vector, and the bias term to obtain the second intermediate term; The first intermediate term is divided by the second intermediate term to obtain the characteristic distribution difference energy coefficient.

Images