A file retrieval method, system, product and readable storage medium

Abstract

A file retrieval method, system, product and readable storage medium, which relate to the field of file retrieval. The method includes: after initial matching based on retrieval terms in the field of water conservancy engineering, obtaining a professional term dictionary and a business rule template from a water conservancy engineering corpus, using the professional term dictionary to perform word frequency statistics and keyword extraction, matching the extracted business keywords with standard terms to obtain professional feature items, and then substituting the professional feature items into the business rule template for rule matching to determine the engineering stage, technical indicators and construction technology of engineering technical documents, and generating file business tags. Select corresponding combinations of business tags according to the professional role type of the user for reorganization and display. Implementing this method can improve the relevance between the retrieval results and the user.

Description

Technical Field

[0001] This application relates to the field of file retrieval, and particularly to a file retrieval method, system, product, and readable storage medium. Background Art

[0002] With the vigorous development of the water conservancy project construction industry, a large number of engineering technical documents have been generated in different stages such as planning, design, construction, and operation and maintenance of various water conservancy projects. These documents not only record the whole process of project construction, but also provide important technical references and experience accumulation for subsequent project construction. Therefore, establishing an efficient file retrieval system is of great significance for the water conservancy project field.

[0003] The current file retrieval systems mainly adopt a keyword-based retrieval method. By establishing a document index library, the retrieval terms input by users are character-matched with the keywords in the documents, so as to return a list of documents containing relevant keywords. At the same time, the system classifies the documents according to basic attributes such as time and type, which is convenient for users to browse and filter by category.

[0004] However, due to the strong professionalism of water conservancy projects, the technical documents contain a large number of professional terms and industry terms, and there are differences in the expression methods among different design units and construction units. As a result, the same technical content may be described differently by different user identities (such as designers, construction workers, etc.). This makes it difficult to accurately understand the professional content of the documents solely relying on the keyword matching retrieval method, resulting in a low relevance between the retrieval results and users. Users with different identities need to spend a lot of time on manual screening and judgment of the retrieval results. Summary of the Invention

[0005] This application provides a file retrieval method, system, product, and readable storage medium for improving the relevance between the retrieval results and users.

[0006] In a first aspect, the present application provides an archive retrieval method, which is applied to an archive retrieval system. The method includes: responding to a user's archive retrieval request, which includes retrieval terms for the water conservancy engineering specialty; performing a match in an archive database based on the retrieval terms for the water conservancy engineering specialty to obtain an initial retrieval result, which includes multiple engineering technical documents; obtaining a professional term dictionary and a business rule template from a water conservancy engineering corpus, where the professional term dictionary includes standard terms and business attributes of water conservancy engineering, and the business rule template includes determination rules for engineering phases, extraction rules for technical indicators, and identification rules for construction techniques; using the professional term dictionary to perform word frequency statistics and keyword extraction on the text content in the initial retrieval result, matching the obtained business keywords with the standard terms in the professional term dictionary to obtain professional feature items of the engineering technical documents; substituting the professional feature items into the business rule template for rule matching, and determining the engineering phase, technical indicators, and construction technique of the engineering technical document based on the matching result of the rule matching to generate archive business tags; selecting a corresponding combination of business tags from the archive business tags according to the professional role type of the user, and reorganizing the initial retrieval result according to a preset display template to obtain an adaptable retrieval result.

[0007] By adopting the above technical solution, after performing an initial match based on the retrieval terms for the water conservancy engineering specialty, a professional term dictionary and a business rule template are obtained from a water conservancy engineering corpus. The professional term dictionary is used for word frequency statistics and keyword extraction, and the extracted business keywords are matched with the standard terms to obtain professional feature items. Then, the professional feature items are substituted into the business rule template for rule matching to determine the engineering phase, technical indicators, and construction technique of the engineering technical document, and generate archive business tags. According to the professional role type of the user, a corresponding combination of business tags is selected for reorganization and display, realizing the professional analysis and personalized presentation of the retrieval result, improving the accuracy and applicability of the retrieval of water conservancy engineering technical documents, and reducing the workload of the user for screening and judging the retrieval result.

[0008] In some embodiments in combination with some embodiments of the first aspect, the steps of using the specialized term dictionary to perform word frequency statistics and keyword extraction on the text content in the initial retrieval results, and matching the obtained business keywords with the standard terms in the specialized term dictionary to obtain the professional feature items of the engineering technical document specifically include: performing word segmentation on the text content of each engineering technical document in the initial retrieval results, counting the occurrence frequencies of each word in the engineering technical document, and selecting the words with frequencies exceeding the preset frequency threshold as the candidate keywords of the engineering technical document; screening out the words that appear in the specialized term dictionary from the candidate keywords as the business keywords of the engineering technical document; performing string matching on the business keywords of the engineering technical document and the standard terms in the specialized term dictionary one by one, and when the matching degree between the business keyword and the standard term exceeds the preset matching threshold, taking the standard term and the corresponding business attribute as the professional feature item of the engineering technical document.

[0009] By adopting the above technical solution, candidate keywords are obtained through word segmentation and statistics of the document content, and then through screening by the specialized term dictionary and matching with the standard terms, the corresponding relationship between the document content and the standard terms is established. Through the method of multi-level screening and threshold control, the extraction accuracy of the professional feature items is ensured, laying a reliable foundation for subsequent rule matching.

[0010] In some embodiments in combination with some embodiments of the first aspect, the steps of substituting the professional feature items into the business rule template for rule matching, and determining the engineering stage, technical indicators and construction technology of the engineering technical document based on the matching results of the rule matching to generate archive business labels specifically include: classifying the professional feature items of the engineering technical document according to the business attributes; inputting the professional feature items with engineering stage attributes in the engineering technical document into the engineering stage determination rule to determine the engineering stage to which the engineering technical document belongs; inputting the professional feature items with technical indicator attributes in the engineering technical document into the technical indicator extraction rule to extract the numerical and descriptive technical indicators of the engineering technical document; inputting the professional feature items with construction technology attributes in the engineering technical document into the construction technology identification rule to identify the main construction technology and process parameters of the engineering technical document; combining the determined engineering stage, extracted technical indicators and identified construction technology of the engineering technical document to generate the business label of the engineering technical document.

[0011] By adopting the above technical solution, after classifying the professional feature items according to the business attributes, they are respectively input into the corresponding determination rules, the engineering stage, technical indicators and construction technology information are extracted and combined to generate business labels. Through the rule-based and modular processing method, the multi-dimensional analysis of the professional content of the engineering technical document is realized, enhancing the professional value of the retrieval results.

[0012] In some embodiments in combination with some embodiments of the first aspect, the step of selecting a corresponding business label combination from the file business labels according to the professional role type of the user and reorganizing the initial search results according to a preset display template to obtain an adaptable search result specifically includes: obtaining the professional role type of the user, and determining the business label type corresponding to the professional role type according to the role-label mapping relationship; screening out the business label subset of the engineering technical document corresponding to the business label type from the file business labels; reorganizing and typesetting the engineering stage, technical indicators, and construction process information in the business label subset according to the preset display template to obtain a reorganized technical document; and integrating the reorganized technical document with the basic information of the engineering technical document to obtain an adaptable search result.

[0013] By adopting the above technical solution, the corresponding business label type is obtained according to the professional role type of the user, the matching business label subset is screened out, the professional information content is reorganized according to the preset display template, and the reorganized professional information is integrated with the document basic information, forming differentiated search results for different professional roles, making the search display content more in line with the professional needs of the user and improving the reading efficiency of the search results.

[0014] In some embodiments in combination with some embodiments of the first aspect, after the step of selecting a corresponding business label combination from the file business labels according to the professional role type of the user and reorganizing the search results according to a preset display template to obtain an adaptable search result, the method further includes: counting the water conservancy project professional search terms with the top preset numerical values in the usage frequency within the preset usage time of the user; classifying the water conservancy project professional search terms according to the dimensions of engineering stage, technical indicators, and construction process to form a search term recommendation list for the user; and when the user issues a new file search request, displaying the water conservancy project professional search terms in the search term recommendation list with the highest matching degree to the currently input characters.

[0015] By adopting the above technical solution, the usage frequency of the user's historical search terms is counted, classified and sorted according to the dimensions of engineering stage, technical indicators, and construction process, and the recommended search terms with the highest matching degree are displayed when the user inputs a new search request, establishing a search term recommendation mechanism based on the user's usage habits and improving the input accuracy of professional search terms and the search operation efficiency.

[0016] In some embodiments in combination with some embodiments of the first aspect, after the step of selecting a corresponding business label combination from the file business labels according to the professional role type of the user and reorganizing the search results according to a preset display template to obtain an adaptable search result, the method further includes: extracting the chart page number information included in the engineering technical document; and marking the chart page number information in the title of the engineering technical document in the adaptable search result.

[0017] By adopting the above technical solution, the page number information of the charts in the engineering technical documents is extracted, and the page number information is marked in the document title, so that the retrieved results directly display the position distribution of the charts, establishing a quick positioning association between the retrieved results and the document content, and optimizing the chart search efficiency of the engineering technical documents.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of marking the chart page number information in the document title in the adaptability retrieval result, the method further includes: when the user clicks on the chart page number information, jumping to the corresponding page position of the engineering technical document.

[0019] By adopting the above technical solution, a page jump function is added to the chart page number information displayed in the title. When the user clicks on the page number information, it directly locates to the corresponding page position of the document, organically combining the information display of the retrieved results with the access to the document content, realizing a quick navigation from the retrieval interface to the specific chart content, and enhancing the retrieval and browsing experience of the engineering technical documents.

[0020] In a second aspect, an embodiment of the present application provides an archive retrieval system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to enable the archive retrieval system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product containing instructions. When the above computer program product runs on the archive retrieval system, it enables the above archive retrieval system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions. When the above instructions run on the archive retrieval system, it enables the above archive retrieval system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the archive retrieval system provided in the second aspect above, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0025] 1. After initial matching based on the retrieval terms of water conservancy engineering, this application obtains a professional term dictionary and a business rule template from the water conservancy engineering corpus, uses the professional term dictionary for word frequency statistics and keyword extraction, matches the extracted business keywords with standard terms to obtain professional feature items, and then substitutes the professional feature items into the business rule template for rule matching to determine the engineering stage, technical indicators, and construction technology of the engineering technical document, and generates archive business tags. By selecting the corresponding business tag combination according to the user's professional role type for recombination display, it realizes the professional analysis and personalized presentation of retrieval results, improves the accuracy and applicability of water conservancy engineering technical document retrieval, and reduces the workload of users in screening and judging retrieval results.

[0026] 2. This application classifies and organizes the usage frequencies of the user's historical retrieval terms by dimension of engineering stage, technical indicators, and construction technology, and displays the recommended retrieval terms with the highest matching degree when the user enters a new retrieval request, establishing a retrieval term recommendation mechanism based on the user's usage habits, which improves the input accuracy of professional retrieval terms and the retrieval operation efficiency.

[0027] 3. This application adds a page jump function to the chart page number information displayed in the title. When the user clicks on the page number information, it directly locates to the corresponding page position in the document, organically combines the information display of the retrieval result with the access to the document content, realizes the fast navigation from the retrieval interface to the specific chart content, and improves the retrieval and browsing experience of the engineering technical document. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flowchart of a file retrieval method in an embodiment of this application;

[0029] Figure 2 is another flowchart of a file retrieval method in an embodiment of this application;

[0030] Figure 3 is a schematic structural diagram of an entity device of a file retrieval system in an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The terms used in the following embodiments of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the specification of this application, the singular forms "a", "one", "the above", "the", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations including one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and should not be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0033] For ease of understanding, the method provided in this embodiment will be described in terms of its process below. Please refer to Figure 1 , which is a schematic flowchart of a process of the file retrieval method in the embodiments of the present application.

[0034] S101. In response to a user's file retrieval request, the file retrieval request includes retrieval terms for the water conservancy engineering specialty.

[0035] Among them, the file retrieval request refers to a query operation request initiated by the user in the retrieval system; the response represents the process of the system processing and giving feedback to the user request; the retrieval terms for the water conservancy engineering specialty are used to represent professional terms or keywords input by the user related to the water conservancy engineering field, such as "flood discharge sluice design", "concrete strength", "diversion construction", etc.; "including" means that the retrieval terms must be a core element existing in the retrieval request.

[0036] The retrieval system executes this step when the user inputs retrieval terms through the retrieval interface and triggers the retrieval action. Specifically, after the system receives the user's retrieval operation, it first verifies the validity of the retrieval request to confirm that the request includes retrieval terms. Then, it preprocesses the retrieval terms, including removing extra spaces, unifying the case format, identifying the professional attributes of the retrieval terms, etc. The system encapsulates the preprocessed retrieval terms into a standard retrieval request format to prepare for subsequent matching retrieval.

[0037] In some embodiments, the response processing of the retrieval request can be implemented in various ways: Optionally, a retrieval input box is set to receive user input, the retrieval terms are obtained through form submission, the legality of the input content is verified, and the retrieval request is generated after standardizing the retrieval terms; Optionally, a retrieval term recommendation list is provided for the user to select. After the user clicks on a recommended term, it is automatically filled into the input box, triggering a retrieval event, and the system captures the event and constructs a retrieval request. It can be understood that other ways can also be used to implement the response processing of the retrieval request, such as voice input, scanning input, etc., which are not limited herein.

[0038] S102. Based on the retrieval terms for the water conservancy engineering specialty, perform matching in the file database to obtain an initial retrieval result, and the initial retrieval result includes multiple engineering technical documents.

[0039] Among them, the archival database refers to a data storage system that stores engineering technical documents and their metadata; matching refers to the process of comparing and searching the search terms with the document content in the database; the initial search results are used to represent the original document set returned by the matching process; engineering technical documents refer to professional documents such as design documents, construction plans, technical specifications, etc. related to water conservancy project construction.

[0040] This step is performed after the system completes the search request processing and is used to find relevant documents from the archive. Specifically, the system uses the pre-processed search terms as matching conditions to perform a full-text search in the archive database. The search scope includes fields such as document title, text content, and metadata. The system extracts all document information that matches the search terms, sorts them by relevance, and forms an initial search result set.

[0041] In some embodiments, database matching retrieval can be implemented in a variety of ways: optionally, using inverted index technology to establish a document retrieval index, searching the index after segmenting the search terms, counting the word frequency and position information, calculating the document relevance score, sorting by score and returning the matching documents; optionally, using a vector space model, converting the search terms and document content into vector representations, calculating the vector cosine similarity, setting a similarity threshold to filter matching documents, and generating a retrieval result list. It is understandable that other methods can also be used to implement database matching retrieval, such as semantic retrieval, fuzzy matching and other technologies, which are not limited here.

[0042] S103, obtaining a professional terminology dictionary and a business rule template from a water conservancy project corpus, wherein the professional terminology dictionary includes standard terms and business attributes of water conservancy projects, and the business rule template includes determination rules for project stages, extraction rules for technical indicators, and identification rules for construction processes.

[0043] Among them, the water conservancy engineering corpus refers to a professional knowledge base containing a large amount of text data in the field of water conservancy engineering; the professional terminology dictionary is used to represent the vocabulary that includes standardized terms and their attribute information in the field of water conservancy engineering; the business rule template refers to a set of rules used to identify and process the content of professional documents; standard terms represent professional terms that are unified by industry specifications; business attributes refer to characteristic information such as professional classification and usage scenarios of terms; judgment rules are used to represent conditional logic for identifying engineering stages; extraction rules represent pattern matching rules for obtaining technical indicators; and identification rules refer to classification rules for determining construction process types.

[0044] This step is executed during system initialization or when updating the knowledge base, and is used to construct the basic support data for document processing. Specifically, the system connects to the water conservancy project corpus, reads the term entries and attribute definitions in the professional term dictionary, and loads various business rule templates. It performs a structured processing on the term dictionary to establish a term index and an attribute mapping relationship. At the same time, it parses the business rule templates into executable rule expressions to prepare for subsequent document feature recognition.

[0045] In some embodiments, professional knowledge acquisition can be achieved in multiple ways: Optionally, access the corpus through direct database connection, execute query statements to obtain dictionary data, parse the data structure to construct an in-memory dictionary, load the rule configuration file to parse the rule template, and establish a rule execution engine; Optionally, use knowledge graph technology to construct a term system, extract term nodes and relationship edges in the graph to construct a dictionary, generate business rule templates based on the inference rules of the graph, and construct a rule inference mechanism. It can be understood that other methods can also be used to achieve the acquisition and loading of professional knowledge, such as technical solutions like distributed caching and real-time synchronization, which are not limited here.

[0046] S104. Use this professional term dictionary to perform word frequency statistics and keyword extraction on the text content in the initial retrieval result, and match the obtained business keywords with the standard terms in the professional term dictionary to obtain the professional feature items of this engineering technical document.

[0047] Among them, word frequency statistics refers to the process of calculating the number of occurrences of each word in a document; keyword extraction is used to represent the operation of identifying important words from the text; business keywords refer to important terms related to water conservancy project business; standard term matching represents the process of corresponding the extracted keywords with dictionary terms; professional feature items are used to represent the professional attribute identifiers of a document.

[0048] This step is executed after obtaining the initial retrieval result and is used to deeply analyze the document content to extract professional features. Specifically, the system first performs word segmentation on the document text, counts the occurrence frequency of each word, and extracts keywords based on word frequency and position information. Then it matches the extracted keywords with the professional term dictionary to determine the canonical term form and business attributes of each keyword. Finally, it integrates the matching results to form a set of feature items reflecting the professional features of the document.

[0049] In some embodiments, document feature extraction can be achieved in various ways: Optionally, the TF-IDF algorithm is used to calculate the importance of words, and words with importance exceeding a threshold are selected as keywords. The standard terms are obtained by querying the term dictionary, and the business attributes of the terms are extracted as feature items to construct the feature vector of the document. Optionally, a deep learning model is used for text feature learning, the pre-trained model is used to extract text semantic features, the features are mapped to the term space for standardization, and the feature item descriptions are generated based on the attributes of the terms. It can be understood that other methods can also be used to achieve the extraction and standardization of document features, such as topic models, semantic similarity and other analysis methods, which are not limited here.

[0050] S105. Substitute the professional feature item into the business rule template for rule matching, and determine the project phase, technical indicators and construction technology of the engineering technical document based on the matching result of the rule matching, and generate an archive business label.

[0051] Among them, rule matching refers to the process of comparing and validating feature items with business rules; the matching result is used to represent the judgment data obtained after rule verification; the project phase represents different periods of project construction, such as feasibility study, preliminary design, construction drawing design, etc.; technical indicators refer to the key parameters and performance requirements in engineering construction; construction technology is used to represent specific construction methods and technical measures; the archive business label represents the business attribute classification identifier of the document.

[0052] This step is executed after the professional feature extraction is completed and is used to deeply analyze the business features of the document. Specifically, the system takes the extracted professional feature items as input and substitutes them into the business rule template for rule calculation. First, the project phase determination rule is applied to determine the project phase based on the time series features and professional terms in the document. Then, the technical indicator extraction rule is used to identify the parameter values and performance descriptions in the document. Next, through the construction technology identification rule, the process flow and technical solutions in the document are analyzed. Finally, these analysis results are integrated to form the business label set of the document.

[0053] In some embodiments, rule matching and label generation can be achieved in various ways: Optionally, the rule engine technology is used to process business rules, the feature items are converted into fact data, rule reasoning is performed to obtain the matching result, the rule output variables are extracted to generate label attributes, the label data structure is assembled, and the business label set is output. Optionally, a decision tree model is used for rule determination, the feature items are mapped to the decision tree nodes, the classification result is obtained by traversing the decision path, the label information of the leaf nodes is extracted, the label attributes are merged, and the complete business label is generated. It can be understood that other methods can also be used to achieve rule matching and label generation, such as fuzzy reasoning, neural network and other technical solutions, which are not limited here.

[0054] S106. Select the corresponding business label combination from the file business labels according to the professional role type of the user, and reorganize the initial search results according to the preset display template to obtain the adaptable search results.

[0055] Among them, the professional role type refers to the business position or professional function classification of the user; the business label combination represents the subset of labels filtered according to the user role; the preset display template is used to represent the layout and display rules of the search results; reorganization refers to the process of reorganizing and formatting the search results according to the template; the adaptable search results represent the final search results after personalized processing.

[0056] This step is executed after generating the file business labels and is used to achieve personalized display of the search results. Specifically, the system first obtains the professional role information of the current user, and filters relevant labels from the file business labels according to the business concerns of the role. Then, it loads the display template corresponding to the user role to determine the display structure and format requirements of the search results. Finally, it reorganizes the initial search results according to the template rules, adjusts the content layout and display order, and highlights the business information that the user is concerned about.

[0057] In some embodiments, the adaptable display of the search results can be achieved in multiple ways: Optionally, configure label filtering rules based on the user role, execute rule filtering to obtain the label subset, read the display template associated with the role, parse the template structure, reorganize the search data, apply style rendering, and output personalized results; Optionally, use template engine technology to process the display logic, map the user role to the label selector, filter the matching business labels, load the template file, replace the template variables, render the page content, and generate the adaptable results. It can be understood that other methods can also be used to achieve the adaptable display of the search results, such as technical solutions like dynamic layout and adaptive rendering, which are not limited here.

[0058] Next, a further and more specific process description of the method provided in this embodiment will be given. Please refer to Figure 2 , which is another process schematic diagram of the file retrieval method in the embodiment of the present application.

[0059] S201. In response to the user's file retrieval request, the file retrieval request contains retrieval terms related to the water conservancy engineering major.

[0060] The file retrieval request is a document query operation initiated by the user through the file retrieval system interface, and contains parameters such as retrieval terms and retrieval scope. The retrieval terms related to the water conservancy engineering major refer to professional terms related to the water conservancy engineering field, such as "dam design", "flood discharge sluice", "seepage analysis", etc.

[0061] The specific implementation process of this step is as follows: When the user enters a search term in the system search box and clicks the search button, the system captures this search behavior, parses the entered search term, and determines whether it belongs to the category of water conservancy engineering professional vocabulary. The system encapsulates the search request into a standard format, including information such as the search term, search time, and user ID, and passes it to the background search module for processing. For example, when the user enters "Design Specification of Flood Discharge Sluice", the system will identify this search term as a professional search term in water conservancy engineering.

[0062] S202. Perform a match in the archive database based on this water conservancy engineering professional search term to obtain an initial search result, which contains multiple engineering technical documents.

[0063] The archive database is a structured data warehouse for storing engineering technical documents, including information such as document content and metadata. The initial search result refers to the original document set matched in the database according to the search term. Engineering technical documents include various types of technical literature such as design drawings, construction plans, and acceptance reports.

[0064] The specific implementation of this step is as follows: The system performs a fuzzy match in the full text content and metadata fields of the documents in the archive database using the search term. The matching algorithm adopts the inverted index method, and after segmenting the document content, it establishes a term-document mapping relationship. The system returns all relevant documents containing the search term as the initial result set. For example, searching for "Design Specification of Flood Discharge Sluice" will return design documents, specification documents, etc. containing these keywords.

[0065] S203. Obtain a professional term dictionary and business rule template from the water conservancy engineering corpus. The professional term dictionary contains standard terms and business attributes in water conservancy engineering, and the business rule template contains determination rules for engineering stages, extraction rules for technical indicators, and identification rules for construction techniques.

[0066] The water conservancy engineering corpus is a knowledge base containing knowledge such as professional vocabulary and business rules. The professional term dictionary stores standardized professional terms and their attribute information. The business rule template defines a set of rules for document classification and information extraction.

[0067] The specific implementation of this step is as follows: The system calls the professional term dictionary and rule template from the corpus. The term dictionary stores terms and their hierarchical relationships, business attributes, etc. in a tree structure. The rule template uses a formal language to define determination rules, including rule elements such as keyword features and numerical ranges. For example, the term dictionary stores attributes such as the type and parameters of "Flood Discharge Sluice", and the rule template defines how to identify engineering stages and extract technical indicators from documents.

[0068] S204. Perform word segmentation on the text content of each engineering and technical document in the initial search results, count the occurrence frequency of each word in the engineering and technical document, and select the words with a frequency exceeding the preset frequency threshold as the candidate keywords for the engineering and technical document;

[0069] Word segmentation refers to the process of splitting continuous text content into individual words, and a word segmentation algorithm based on dictionary and statistics is adopted. The occurrence frequency refers to the number of times a word repeats in a document. The preset frequency threshold is a pre-set word frequency judgment criterion, such as set to 0.1% of the total number of words in the document. Candidate keywords refer to a set of words with a relatively high frequency.

[0070] The specific implementation process is as follows: First, preprocess the engineering and technical document, including removing punctuation marks, special characters, etc. Then use the word segmentation algorithm to split the text. For example, for the sentence "The flood discharge sluice construction adopts a reinforced concrete structure", the word segmentation results in "flood discharge sluice / construction / adopts / reinforced concrete / structure". Next, count the occurrence times of each word, establish a word frequency statistics table, and record the words and their frequencies. Finally, compare the word frequency with the preset threshold, and screen out the words with a frequency higher than the threshold. For example, if the total number of words in the document is 1000 and the threshold is set to 0.1%, then the words with an occurrence times exceeding 1 are selected as candidate keywords.

[0071] S205. Screen out the words that appear in the professional term dictionary from the candidate keywords as the business keywords of the engineering and technical document;

[0072] The professional term dictionary is a standard term set in the field of water conservancy projects, containing standardized professional vocabulary. Business keywords are words with professional meanings screened out from candidate keywords.

[0073] The specific implementation process is as follows: Compare the candidate keywords obtained in step S204 with the professional term dictionary. Adopt an exact matching method to check whether each candidate keyword exists in the term dictionary. If it exists, mark the word as a business keyword. For example, the candidate keywords "flood discharge sluice" and "reinforced concrete" have corresponding entries in the term dictionary, so they are determined as business keywords, while ordinary words such as "adopts" are filtered out.

[0074] S206. Perform string matching one by one between the business keywords of the engineering and technical document and the standard terms in the professional term dictionary. When the matching degree between the business keyword and the standard term exceeds the preset matching threshold, take the standard term and the corresponding business attribute as the professional feature items of the engineering and technical document.

[0075] String matching is the process of calculating the similarity between two text passages. The matching degree is a quantitative indicator of similarity, with a value range of 0 - 1. The preset matching threshold is the standard value for determining the matching degree of words, such as 0.8. Professional feature items include standard terms and their business attribute information.

[0076] The specific implementation process is as follows: For each business keyword, calculate the string similarity with the standard terms in the term dictionary. Using the edit distance algorithm, divide the minimum number of operations required to convert two words by the length of the longer word to obtain a normalized similarity value. When the similarity is greater than the preset threshold, extract the business attributes of the corresponding standard term to form professional feature items. For example, the similarity between the business keyword "reinforced concrete" and the standard term "reinforced concrete" is 0.85, exceeding the threshold of 0.8, then extract the material type, strength, etc. of "reinforced concrete" as feature items.

[0077] S207. Classify the professional feature items of the engineering technical document according to their business attributes;

[0078] Business attributes are the classification identifiers of professional feature items, including three categories: engineering stage, technical indicators, and construction technology. The classification of professional feature items is the process of classifying and organizing feature items with the same business attributes.

[0079] The specific implementation process is as follows: The system reads the business attribute fields of each professional feature item and establishes a mapping table of attribute - feature item. Group the feature items according to their business attributes to form an engineering stage feature set, a technical indicator feature set, and a construction technology feature set. For example, the business attribute of the feature item "preliminary design" is the engineering stage and is classified into the engineering stage feature set; the business attribute of the feature item "concrete strength C30" is the technical indicator and is classified into the technical indicator feature set.

[0080] S208. Input the professional feature items with engineering stage attributes in the engineering technical document into the engineering stage determination rule to determine the engineering stage to which the engineering technical document belongs;

[0081] The engineering stage is the division of the water conservancy project construction process, including stages such as feasibility study, preliminary design, construction drawing design, construction, and acceptance. The engineering stage determination rule defines the characteristic words and determination methods for different stages.

[0082] The specific implementation process is as follows: Match the feature items in the engineering stage feature set with the determination rule. The determination rule adopts a decision tree structure, and the nodes contain the combination conditions of feature items and the stage determination results. The system checks in turn whether the feature items meet the rule conditions to determine the stage to which the document belongs. For example, if the document contains feature items such as "preliminary design" and "design specification", which meet the determination conditions of the preliminary design stage, then it is determined that the document belongs to the preliminary design stage.

[0083] S209. Input the professional feature items with technical index attributes in the engineering technical document into the technical index extraction rule to extract the numerical and descriptive technical indexes of the engineering technical document;

[0084] Technical indexes include numerical indexes (such as dimensions, strength) and descriptive indexes (such as material type, construction requirements). The technical index extraction rule defines the identification mode and extraction method of the indexes.

[0085] The specific implementation process is as follows: Apply the extraction rule to the feature items in the technical index feature set. For numerical indexes, use regular expressions to match numbers and units. For example, extract the numerical value "30" and unit "MPa" from "Concrete strength C30". Descriptive indexes are extracted through keyword positioning and context analysis. For example, extract the material type "reinforced concrete" from "Adopt reinforced concrete structure". The extracted indexes are stored separately according to numerical and descriptive types to form the technical index set of the document. For complex professional terms, use predefined templates for parsing. For example, parse "C30P8" into strength grade "C30" and impermeability grade "P8".

[0086] S210. Input the professional feature items with construction process attributes in the engineering technical document into the construction process identification rule to identify the main construction processes and process parameters of the engineering technical document.

[0087] Construction process is the specific construction method and technical measure adopted in engineering construction. The main construction process refers to the core construction method adopted in the project. Process parameter is the specific technical requirement and control index of the construction process. The construction process identification rule defines the identification method of process features and the parameter extraction method.

[0088] The specific implementation process is as follows: Apply the identification rule to parse the feature items in the construction process feature set. First, identify the main construction processes through keyword matching, such as process action words like "vibrating", "spraying", "pouring", etc. Then extract the parameter information related to the process, including process requirements and control values. For example, identify the "vibrating" process and time parameter "30 seconds" from "The vibrating time of concrete is not less than 30 seconds"; identify the "segmented pouring" process and thickness parameter "50 cm" from "Adopt three-layer segmented pouring, and the thickness of each layer is controlled within 50 cm". For complex process descriptions, use predefined semantic templates for parsing to extract the process flow and parameter combination.

[0089] S211. Combine the engineering stages determined in the engineering technical document, the extracted technical indexes, and the identified construction processes to generate the business labels of the engineering technical document.

[0090] Business tags are professional attribute identifiers for engineering and technical documents, containing information in three dimensions: engineering stage, technical indicators, and construction technology. Tag generation is the process of integrating professional information from different dimensions to form structured tags.

[0091] The specific implementation process is as follows: Organize tag data in JSON format, including three main fields: engineering stage, technical indicators, and construction technology. The engineering stage field stores the construction stage to which the document belongs; the technical indicators field stores indicator information separately according to numerical and descriptive types; the construction technology field includes the main technology and its parameters. For example, the generated tag format is: {"Engineering Stage": "Construction Stage", "Technical Indicators": {"Numerical": ["Concrete Strength: 30MPa"], "Descriptive": ["Material: Reinforced Concrete"]}, "Construction Technology": {"Technology Name": "Segmental Pouring", "Technology Parameters": ["Layer Thickness: 50cm"]}}.

[0092] S212. Obtain the professional role type of the user, and determine the corresponding business tag type for this professional role type according to the role-tag mapping relationship.

[0093] The professional role type refers to the business positions of users in water conservancy projects, such as designers, constructors, supervisors, etc. The role-tag mapping relationship defines the business tag types that different roles are concerned about. The business tag type is the classification of professional information in business tags.

[0094] The specific implementation process is as follows: First, obtain the role information of the user from the user management system. Then query the role-tag mapping table to obtain the tag types that this role is concerned about. The mapping table adopts a matrix structure, where the rows represent role types and the columns represent tag types. The matrix element value of 1 indicates concern, and 0 indicates no concern. For example, designers focus on technical indicators, and the value corresponding to the technical indicators column is 1; constructors focus on construction technology, and the value corresponding to the construction technology column is 1. Determine the combination of tag types to be displayed according to the mapping relationship.

[0095] S213. Screen out the subset of business tags of this engineering and technical document corresponding to this business tag type from the archive business tags.

[0096] The business tag subset is a set of specific type tags screened out from the complete business tags. Tag screening is the process of extracting relevant information from business tags according to the user role requirements.

[0097] The specific implementation process is as follows: Read the business label types determined in step S212 and traverse the JSON data structure of the complete business labels. For each label type, extract the label information of the corresponding field. For example, if designers are concerned about technical indicators, extract the content of the "technical indicators" field from the business labels; if they are also concerned about the engineering stage, additionally extract the content of the "engineering stage" field. The extracted label information maintains the original data structure and hierarchical relationship to form a label subset for a specific role. The extraction process uses JSON parsing and field filtering operations to ensure the integrity and structure of the data.

[0098] S214. Reorganize and typeset the engineering stage, technical indicators, and construction technology information in the business label subset according to a preset display template to obtain a reorganized technical document.

[0099] The preset display template is a template file that defines the layout and display format of label information. Reorganizing and typesetting is the process of reorganizing label information according to the format specified by the template. The reorganized technical document is a collection of professional information that has been formatted.

[0100] The specific implementation process is as follows: Substitute the content of the label subset into a preset HTML template. The template defines the display positions, font styles, typesetting formats, etc. of different types of information. The engineering stage information is displayed at the head of the document in a prominent title style. The technical indicators are classified and displayed as numerical and descriptive types, presenting the indicator names and values in a table form. The construction technology is presented in a list form, showing the technology names and parameter information. For example, the display template for technical indicators is: Index Name Index Value , substitute the concrete strength of 30 MPa and generate the corresponding table row.

[0101] S215. Integrate the reorganized technical document with the basic information of the engineering technical document to obtain an adaptable retrieval result.

[0102] The basic information includes metadata such as document number, name, date, etc. The adaptable retrieval result is a complete document view after integrating the reorganized professional information with the basic information.

[0103] The specific implementation process is as follows: First, obtain the basic information fields of the engineering technical document from the archive database. Use a predefined document structure template to place the basic information in a fixed position in the document view. Then insert the content of the reorganized professional information into the corresponding display area. The integrated document structure includes: a document title area (displaying the document name and number), a basic information area (displaying the creation date, author, etc.), and a professional information area (displaying the reorganized label content). The finally generated adaptable retrieval result is in HTML format, facilitating display on a Web interface. The integration process ensures the integrity of the information and the rationality of the layout, making the retrieval result convenient for users to read and use.

[0104] In some embodiments, the following steps may further be included:

[0105] Count the water conservancy engineering professional search terms whose usage frequencies rank among the top preset values within the preset usage time of the user.

[0106] In this step, the preset usage time is the time interval of the statistical range, such as the recent 30 days. The usage frequency is the number of times the search term is used. The preset value is the selected frequency ranking threshold, such as the top 10. The water conservancy engineering professional search term is the professional term search term used by the user in the past.

[0107] The specific implementation process is as follows: Query the search records of the specified user within the preset time period from the search log database. Extract the search term field for each search record and establish a search term-frequency mapping table. Use a counter to count the number of times each search term appears and sort them in descending order of frequency. Select the top N search terms, where N is the preset ranking threshold. For example, when counting the search terms used by the user in the recent 30 days: "flood discharge sluice design" is used 15 times, "concrete strength" is used 12 times, and select the top 10 search terms with the highest frequencies to form a high-frequency search term set.

[0108] Classify the water conservancy engineering professional search terms according to the dimensions of engineering stage, technical indicators, and construction technology to form a search term recommendation list for the user.

[0109] In this step, the search term classification is the process of classifying high-frequency search terms according to professional attributes. The search term recommendation list is a structured set of search terms after classification and collation.

[0110] The specific implementation process is as follows: Identify the attributes of each search term in the high-frequency search term set. Adopt predefined classification rules, including feature word lists for each dimension. Match the search term with the feature words to determine the dimension to which it belongs. For example, search terms containing "design" and "planning" are classified into the engineering stage dimension; search terms containing "strength" and "size" are classified into the technical indicator dimension; search terms containing "construction" and "technology" are classified into the construction technology dimension. The classification results are stored in a tree structure, with each dimension as a first-level node and the subordinate nodes as the corresponding search term lists.

[0111] When the user adds a file search request, display the water conservancy engineering professional search terms in the search term recommendation list with the highest matching degree to the current input characters.

[0112] In this step, the current input characters are the text entered by the user in the search box. The matching degree is the similarity between the input characters and the recommended search terms.

[0113] The specific implementation process is as follows: Listen for the user's input event in the search box and obtain the input character sequence. Calculate the string similarity between the input characters and the search terms in the recommendation list. Use the character-level edit distance algorithm to calculate the minimum number of operations required to convert the input characters into the search terms. Select the top N search terms with the highest similarity as candidate recommended terms. Display the list of candidate recommended terms below the search box, sorted in descending order of similarity. When the user continues to input, update the recommendation list in real time. For example, when the user inputs "flood discharge", the system matches relevant search terms such as "flood discharge sluice design" and "flood discharge scale" from the recommendation list and displays them to the user.

[0114] Extract the page number information of the charts and tables contained in this engineering technical document.

[0115] In this step, the page number information of the charts and tables is the page number position where the drawings and tables are located in the document. The title of the engineering technical document is the name and basic identification information of the document. Page number annotation is the process of displaying the chart position information in the document title.

[0116] The specific implementation process is as follows: First, parse the table of contents structure and content index of the document, and extract the table of contents items containing the keywords "Figure" and "Table". Extract the page number values for each table of contents item to form a chart-page number mapping table. Use the regular expression matching pattern to identify the page number information of the charts and tables in the document. Sort and organize the extracted page number information by drawings and tables to generate a formatted page number list, such as "Figures: pages 3, 5, 7; Tables: pages 10, 12". Add page number annotation after the document title, enclosing the page number information in parentheses, such as "Flood Discharge Sluice Design Specification (Figures: pages 3, 5, 7; Tables: pages 10, 12)", integrating the page number information with the title to facilitate users to quickly understand the chart position distribution.

[0117] Annotate the page number information of the charts and tables in the title of this engineering technical document in the adaptable search results.

[0118] In this step, the adaptable search results are the reorganized document display views. Annotation is the process of adding information in a specific format to a specified location. Annotation within the title refers to the display method of integrating additional information at the document title position.

[0119] The specific implementation process is as follows: Set the display position of page number information in the document title area of the adaptability retrieval results. Use the combination format of "title text + parentheses + page number information" for marking to ensure that the page number information is clearly distinguishable from the original title. The specific marking method of page number information is: Add a small parenthesis after the title text, and list the page numbers in the parentheses according to the categories of "Figure:" and "Table:", separated by commas between the page numbers, and separated by semicolons between the categories. For example, expand the title of "Flood Discharge Sluice Design Specification" to "Flood Discharge Sluice Design Specification (Figure: Pages 3, 5, 7; Table: Pages 10, 12)". When marking, maintain the primary and secondary relationship of the title, make the page number information a supplementary explanation of the title, and at the same time highlight the interactive characteristics of the page number through styles such as font size, color, and spacing. To improve the recognition of information, the page number digits are in a highlighted style, and a mouse hover effect is added to prompt clickable jump. The entire marking process needs to ensure that it does not affect the clarity of the original title and the integrity of the document structure.

[0120] When the user clicks on the page number information of the chart, it jumps to the corresponding page position of the engineering technical document.

[0121] In this step, the page number jump is an interactive operation that locates to the specified page of the document after clicking on the page number information. The page number position is the specific page number in the document.

[0122] The specific implementation process is as follows: Add a hyperlink mark to the page number information in the title, and the link address contains a page number parameter. Set an anchor mark at the corresponding page position in the document content area to establish the correspondence between the page number and the document position. When the user clicks on the page number link, a page positioning event is triggered, and the document content area is scrolled to the corresponding page position. The clickable characteristics of the page number information are highlighted through style settings to enhance the user interaction experience. Add a smooth scrolling effect during the page jump process to make the positioning process more natural. The entire interactive process does not require the user to manually turn the page, improving the efficiency of document browsing.

[0123] The following gives a specific example of the method in this embodiment:

[0124] Construction worker Zhang needs to query the concrete construction process parameters of a certain reservoir flood discharge sluice at the construction site. He opens the file retrieval system and enters the professional retrieval term of "flood discharge sluice concrete construction" in the retrieval box.

[0125] The system responds to the retrieval request, performs a full-text match in the file database with "flood discharge sluice concrete construction" as the keyword, and returns 15 engineering technical documents containing relevant content as the initial retrieval results, including various types of documents such as design specifications, construction organization designs, and construction plans.

[0126] The system calls professional resources from the water conservancy project corpus. The professional term dictionary contains standard terms such as "flood discharge sluice", "concrete strength", "vibration compaction", "curing", etc., as well as their business attributes such as the engineering parts they belong to, material properties, and construction techniques. The business rule templates include: rules for determining the engineering stage through keywords such as "design", "construction", "acceptance", etc. in the document; rules for extracting technical indicators through formats such as "strength grade C30", "impermeability grade P8", etc.; rules for identifying construction techniques through parameters such as "vibration compaction time", "curing period", etc.

[0127] Word segmentation is performed on each document in the initial retrieval results, and the word frequencies are counted. Taking a "Concrete Construction Plan for Flood Discharge Sluice" as an example, the system identifies that the high-frequency words include "concrete" appearing 95 times, "vibration compaction" appearing 42 times, "curing" appearing 38 times, etc. These high-frequency words are matched with the professional term dictionary to determine that "concrete" belongs to the material category of terms, and "vibration compaction" and "curing" belong to the construction technique category of terms, forming the professional feature items of this document. For this document with a total of 5000 words, the system sets the word frequency threshold to 0.1% of the total number of words, that is, words appearing more than 5 times are selected as candidate keywords. When these candidate keywords are matched with the professional term dictionary, those with a matching degree exceeding 0.8 are determined as standard terms. For example, the matching degree of "vibration compaction" and "vibratory compaction" is 0.85, so the standard term "vibration compaction" and its construction technique attributes are selected.

[0128] The professional feature items are classified according to the three dimensions of engineering stage, technical indicators, and construction techniques, and then substituted into the business rule template for matching. The engineering stage rule identifies feature words such as "construction plan" and "construction steps", and determines that this document belongs to the construction stage; the technical indicator rule extracts numerical indicators such as "concrete strength grade C30" and "impermeability grade P8"; the construction technique rule identifies specific process parameters such as "using a high-frequency vibrator for vibration compaction, vibration compaction time 30 - 40 seconds" and "using film covering for curing, curing time not less than 14 days", and generates the archival business labels for this document accordingly.

[0129] The system identifies Xiao Zhang's role as a construction worker, and determines that the content related to construction techniques needs to be highlighted according to the role-label mapping relationship. Construction technique parameters such as vibration compaction method, time requirements, curing method, and curing period are screened out from the archival business labels. The system loads the display template dedicated to construction workers, and displays these process parameters in a table form at a prominent position on the page, including columns such as process name, parameter value, and control requirements, and generates an adaptable retrieval result after integrating with the basic information of the document and presents it to Xiao Zhang.

[0130] Meanwhile, the system analysis found that the frequently used search terms by Xiao Zhang within the past 30 days included "concrete vibration" and "concrete curing", etc. These terms were sorted into a recommended list according to the construction technology dimension. When he entered "hun", the system would prompt these high-frequency search terms for quick selection.

[0131] The system also marked the page numbers where the documents contained construction technology diagrams in the search results, such as "(Figure: Pages 15, 16)". The page numbers supported click operations. After Xiao Zhang clicked, he could directly jump to view the specific diagrams of vibration, curing and other technologies.

[0132] Through this intelligent search and display method, Xiao Zhang can quickly and accurately obtain the concrete construction technology parameters required at the construction site, which fully reflects the practical application value of this application in improving the management efficiency of water conservancy project archives.

[0133] The following describes the archive retrieval system in the embodiment of the present invention from the perspective of hardware processing. Please refer to Figure 3 which is a schematic structural diagram of an entity device of the archive retrieval system in the embodiment of this application.

[0134] It should be noted that Figure 3 the structure of the shown archive retrieval system is only an example and should not bring any restrictions to the functions and usage scope of the embodiments of the present invention.

[0135] As Figure 3 shown, the archive retrieval system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 302 or the program loaded from the storage section 308 into the Random Access Memory (RAM) 303, such as executing the method described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other through a bus 304. The Input / Output (I / O) interface 305 is also connected to the bus 304.

[0136] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. The drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0137] Specifically, according to an embodiment of the present invention, the processes described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are executed.

[0138] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or combined with an instruction execution system, device, or apparatus.

[0139] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the block may occur in a different order from that marked in the accompanying drawings.

[0140] Specifically, the file retrieval system of this embodiment includes a processor and a memory. A computer program is stored on the memory. When the computer program is executed by the processor, it implements the file retrieval method provided in the above-mentioned embodiment.

[0141] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the file retrieval system described in the above-mentioned embodiment; or it may exist separately and not be assembled into the file retrieval system. The above storage medium carries one or more computer programs. When the one or more computer programs are executed by a processor of the file retrieval system, the file retrieval system is enabled to implement the file retrieval method provided in the above-mentioned embodiment.

[0142] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present application.

[0143] As used in the above embodiments, depending on the context, the term "when..." may be interpreted to mean "if...", or "after...", or "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" may be interpreted to mean "if determining...", or "in response to determining...", or "when detecting (the stated condition or event)", or "in response to detecting (the stated condition or event)".

[0144] Those of ordinary skill in the art can understand all or part of the processes in the methods of the above embodiments. These processes can be completed by relevant hardware instructed by a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it may include the processes of the above method embodiments. The foregoing storage medium includes: various media such as ROM or random access memory RAM, magnetic disk, or optical disk that can store program codes.

Claims

1. A file retrieval method, characterized in that: Applied to an archive retrieval system, the method comprises: Responding to a user's file search request, wherein the file search request includes a search term for water conservancy engineering; Based on the water conservancy engineering professional search terms, matching is performed in the archive database to obtain initial search results, wherein the initial search results include multiple engineering technology documents; Obtaining a professional terminology dictionary and a business rule template from a water conservancy project corpus, wherein the professional terminology dictionary includes standard terms and business attributes of water conservancy projects, and the business rule template includes a determination rule for a project stage, an extraction rule for technical indicators, and an identification rule for a construction process; Using the professional term dictionary to perform word frequency statistics and keyword extraction on the text content in the initial search results, matching the obtained business keywords with the standard terms in the professional term dictionary to obtain the professional feature items of the engineering technical document; Substituting the professional feature item into the business rule template for rule matching, determining the engineering phase, technical indicators and construction process of the engineering technical document based on the matching result of the rule matching, and generating an archive business tag; Acquire the professional role type of the user, and determine the business tag type corresponding to the professional role type according to the role-tag mapping relationship; Filtering out a subset of business tags of the engineering technical documents corresponding to the business tag type from the archive business tags; Reorganize and typeset the engineering phases, technical indicators, and construction process information in the business tag subset according to a preset display template to obtain a reorganized technical document; The basic information of the recombinant technology document and the engineering technology document are integrated to obtain an adaptive search result.

2. The method according to claim 1, characterized in that The step of using the professional term dictionary to perform word frequency statistics and keyword extraction on the text content in the initial search results, matching the obtained business keywords with the standard terms in the professional term dictionary, and obtaining the professional feature items of the engineering technical document specifically includes: Performing word segmentation processing on the text content of each engineering technical document in the initial search results, counting the frequency of occurrence of each word in the engineering technical document, and selecting words whose frequency exceeds a preset frequency threshold as candidate keywords for the engineering technical document; Screening out words that appear in the professional term dictionary from the candidate keywords as business keywords of the engineering technical document; The business keywords of the engineering technical document are string matched with the standard terms in the professional terminology dictionary one by one. When the matching degree between the business keywords and the standard terms exceeds a preset matching threshold, the standard terms and corresponding business attributes are used as professional feature items of the engineering technical document.

3. The method according to claim 1, characterized in that: Substituting the professional feature item into the business rule template for rule matching, determining the engineering phase, technical indicators and construction process of the engineering technical document based on the matching result of the rule matching, and generating an archive business tag specifically includes: Classifying the professional feature items of the engineering technical documents according to business attributes; Inputting the professional feature items with engineering stage attributes in the engineering technical document into the engineering stage determination rule to determine the engineering stage to which the engineering technical document belongs; Inputting the professional feature items with technical indicator attributes in the engineering technical document into the technical indicator extraction rule to extract the numerical and descriptive technical indicators of the engineering technical document; Inputting the professional feature items with construction process attributes in the engineering technical document into the construction process identification rule to identify the main construction processes and process parameters of the engineering technical document; The engineering phase determined in the engineering technical document, the extracted technical indicators and the identified construction process are combined to generate a business label for the engineering technical document.

4. The method according to claim 1, characterized in that: After the step of selecting a corresponding business tag combination from the archive business tags according to the professional role type of the user, and reorganizing the search results according to a preset display template to obtain an adaptive search result, the method further includes: Count the water conservancy engineering professional search terms with the highest frequency of use by the user within the preset usage time and the highest preset value; Classifying the water conservancy engineering professional search terms according to the dimensions of engineering stage, technical indicators and construction technology to form a search term recommendation list for the user; When the user adds a new archive search request, the water conservancy engineering professional search terms in the recommended list of search terms with the highest matching degree with the current input characters are displayed.

5. The method according to claim 1, characterized in that After the step of selecting a corresponding business tag combination from the archive business tags according to the professional role type of the user, and reorganizing the search results according to a preset display template to obtain an adaptive search result, the method further includes: Extracting the diagram page number information contained in the engineering technical document; In the adaptability search result, the diagram page number information is marked in the title of the engineering technical document.

6. The method according to claim 5, characterized in that After the step of marking the diagram page number information in the document title in the adaptability search result, the method further includes: When the user clicks on the page number information of the chart, it jumps to the page number position corresponding to the engineering technical document.

7. A file retrieval system, characterized in that: The archive retrieval system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the archive retrieval system to execute the method described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on an archive retrieval system, the archive retrieval system is caused to execute the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is run on an archive retrieval system, the archive retrieval system is caused to execute the method according to any one of claims 1 to 6.

Images