Operation tunnel disease case library construction and rapid retrieval method and system

Abstract

The invention discloses an operation tunnel disease case library construction and rapid retrieval method and system, and the method and system achieve the efficient collection, arrangement, storage and retrieval of operation tunnel disease cases through employing an advanced data processing method and a knowledge graph construction and retrieval technology. Data mining and ontology modeling technologies are utilized to carry out deep processing and classification on collected data, an operation tunnel disease case library based on a knowledge graph is constructed, intelligent retrieval of disease knowledge is realized in combination with an NLP natural language processing technology, and powerful technical support is provided for diagnosis, processing and prevention of operation tunnel diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel engineering, and in particular to a method and system for constructing and quickly retrieving a disease case library of an operating tunnel. Background Art

[0002] With the advancement of infrastructure construction, the number and scale of tunnel projects have increased, but operating tunnels have cracks, water leakage and other defects due to natural and human factors. These defects reduce the safety and durability of tunnel structures, affect their use functions, and may even cause safety accidents that threaten people's lives and property. For example, crack expansion can cause tunnel structure damage, water leakage can accelerate material aging and damage, and severe lining deformation can cause tunnel collapse.

[0003] Traditional disease records are mostly paper documents, which are inconvenient to store and manage data, and difficult to conduct effective data analysis and mining. In terms of disease retrieval, traditional methods mainly rely on manual review of archives and empirical judgment, with extremely low retrieval efficiency, and are unable to quickly and accurately find similar disease cases to provide reference for the treatment of current diseases.

[0004] Knowledge graph technology is a key research direction of artificial intelligence. It constructs a knowledge network by representing entities and relationships in a structured manner. It can assist in diagnosis and treatment in medicine and in risk assessment in finance. Although it has not been widely used in the field of tunnel engineering, it has broad prospects. It can be used for the management of tunnel defects in operation, and can achieve efficient organization and association of defect cases, laying a technical foundation for retrieval, which is of great significance. Summary of the invention

[0005] The present invention provides a method and system for constructing and quickly retrieving a disease case library of an operating tunnel, so as to solve the problems of inconvenient storage and management of traditional disease data and low disease retrieval efficiency.

[0006] According to the first aspect, an embodiment provides a method for constructing and quickly retrieving a case library of operational tunnel defects, the method comprising:

[0007] Collect the disease case data of operating tunnels and pre-process the disease case data;

[0008] According to the types and characteristics of tunnel diseases, the disease case data is classified and labeled, and features are extracted;

[0009] Based on the processed disease case data, a tunnel disease knowledge graph is constructed by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating;

[0010] Using natural language processing technology, we perform semantic analysis and understanding on the query statements entered by users, convert the query statements into query paths in the knowledge graph, and then perform path search and matching in the knowledge graph, return semantically consistent retrieval results, and display the retrieval results to users.

[0011] Furthermore, data on operational tunnel disease cases are collected, including:

[0012] The disease case data includes disease type, disease location, disease occurrence time, disease severity, and tunnel environmental parameters.

[0013] Furthermore, the disease case data is preprocessed, including:

[0014] Clean the collected data to remove duplicate data, erroneous data and noise data.

[0015] Furthermore, according to the types and characteristics of tunnel diseases, the disease case data is classified and labeled, and feature extraction is performed, including:

[0016] Data classification and labeling: Establish a disease classification system to classify diseases; for each type of disease, further subdivide the specific type and degree; when labeling data, use standardized labeling methods and terminology to ensure data consistency and comprehensibility;

[0017] Feature extraction: Use machine learning algorithms to extract key information that can reflect the characteristics of tunnel diseases from raw data.

[0018] Furthermore, based on the processed disease case data, a tunnel disease knowledge graph is constructed by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating, including:

[0019] Ontology modeling: Based on the domain knowledge and data characteristics of tunnel diseases, an ontology model of the tunnel disease knowledge graph is constructed. The ontology model defines the concepts, attributes, relationships and classification system of tunnel diseases.

[0020] Knowledge extraction: extract knowledge from processed data and fill it into the knowledge graph; knowledge extraction includes entity recognition, relationship extraction and attribute extraction; entity recognition refers to identifying entities related to tunnel diseases from text data; relationship extraction refers to determining the relationship between entities; attribute extraction refers to extracting attribute information of entities; knowledge extraction adopts a method that combines natural language processing technology and image recognition technology;

[0021] Knowledge fusion: Integrate knowledge extracted from different sources, eliminate duplicate and contradictory information, and build a unified knowledge graph. Knowledge fusion uses knowledge graph fusion algorithm and ontology mapping technology to achieve semantic alignment and integration between different knowledge sources.

[0022] Knowledge storage and update: Store the constructed knowledge graph in the database and create an index for retrieval; use graph database technology to store the knowledge graph; establish an update mechanism for the knowledge graph to update the data in the knowledge graph regularly or in real time.

[0023] Furthermore, the knowledge extraction specifically includes:

[0024] The K-means algorithm combined with the 3σ principle is used to extract knowledge from the processed data, and manual review and verification are combined to ensure that the extracted knowledge is accurate;

[0025] If the quantitative indicators in the tunnel characteristics are directly extracted, the obtained data is not knowledge and cannot be used for feature calculation in the later stage. The K-means algorithm is used to group the data according to the degree of aggregation in the data set, and the data with similar indicator ranges are classified to facilitate knowledge extraction;

[0026] The principles and steps of data aggregation through K-merns are as follows:

[0027] First, randomly select K samples as the initial cluster centers, then calculate the distance from each sample to each cluster center, and assign the sample to the cluster with the closest distance; then recalculate the center of each cluster, that is, the mean of all samples in the cluster; repeat the corresponding process until the cluster center no longer changes or the preset number of iterations is reached;

[0028] If there are abnormal data that are too large or too small in the aggregated data, in order to remove the abnormal values ​​in the aggregation results, the 3σ principle is used to eliminate the abnormal values ​​with large deviations in the grouped data and complete the setting of the upper and lower limits of each group.

[0029] Furthermore, the knowledge storage and updating specifically includes:

[0030] Data storage includes:

[0031] Create nodes: Use Cypher query language to create nodes;

[0032] Create relationship: establish the relationship between nodes;

[0033] Create index: Create an index on a specific attribute;

[0034] The integrated knowledge is stored in the Neo4j graph database and can be easily retrieved and queried.

[0035] Furthermore, path search and matching are performed in the knowledge graph to return semantically consistent search results, including:

[0036] The core algorithms of semantic search include:

[0037] TF-IDF: Calculate the frequency of keywords in disease knowledge and multiply it by the frequency in the disease knowledge set;

[0038] TF-IDF t,d =TF t,d ×IDF t

[0039] That is, the TF-IDF value of term t in document d is equal to the term frequency IDF t and Inverse Document Frequency TF t，d The product of

[0040] By calculating the TF-IDF value, it is possible to determine which keywords are important keywords for the disease, thereby improving the precision and accuracy of the search in semantic search.

[0041] Furthermore, the search results are displayed to the user, specifically including:

[0042] Design a friendly user interface to present the search results to users in an intuitive and clear manner; use a variety of forms including charts and images to display information including the location, characteristics, and treatment methods of the disease, making it easy for users to view and understand.

[0043] According to the second aspect, an embodiment provides a system for constructing and quickly retrieving a case library of operational tunnel defects, the system comprising:

[0044] Data collection module, used to collect and pre-process the disease case data of operating tunnels;

[0045] The data processing module is used to classify and label the tunnel disease case data according to the type and characteristics of the tunnel disease, and to extract features;

[0046] The knowledge graph construction module is used to construct a tunnel disease knowledge graph based on the processed disease case data by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating;

[0047] The retrieval and display module uses natural language processing technology to perform semantic analysis and understanding on the query statements input by users, convert the query statements into query paths in the knowledge graph, and then perform path search and matching in the knowledge graph, return semantically consistent retrieval results, and display the retrieval results to users.

[0048] According to a third aspect, an embodiment provides an electronic device, the device comprising: a processor and a memory;

[0049] The memory is used to store one or more program instructions;

[0050] The processor is used to run one or more program instructions to execute the steps of a method for constructing and quickly retrieving a case library of operating tunnel defects as described in any of the above items.

[0051] According to the fourth aspect, an embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a method for constructing and quickly retrieving a case library of operating tunnel defects as described in any one of the above items are implemented.

[0052] The present invention provides a method and system for constructing and quickly retrieving a case library of operational tunnel diseases, which has the following beneficial effects:

[0053] 1. Intelligent retrieval based on knowledge graph: Using knowledge graph technology, tunnel disease cases are associated and integrated to achieve fast and accurate retrieval; the retrieval algorithm based on keywords and semantics can meet the different retrieval needs of users and improve retrieval efficiency and accuracy.

[0054] 2. Good scalability and compatibility: The technical solution of the present invention has good scalability, and can add disease case data according to actual needs to expand the content and function of the knowledge graph; at the same time, it can be integrated with other tunnel management systems to achieve data sharing and collaborative work.

[0055] 3. Provide decision-making support for tunnel disease management: Through the analysis and mining of tunnel disease cases, provide scientific decision-making basis for tunnel maintenance, repair and management; help managers to timely identify potential disease risks, formulate reasonable treatment plans, and improve the safety and reliability of tunnels. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of a method for constructing and quickly retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0057] Figure 2 A software implementation architecture diagram of a method for constructing and retrieving a case library of operational tunnel defects provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of K-means aggregation in a method for constructing and retrieving a case library of operational tunnel defects provided by an embodiment of the present invention;

[0059] Figure 4The data aggregation and processing effects before and after the data aggregation in a method for constructing and retrieving a case library of operational tunnel defects provided by an embodiment of the present invention;

[0060] Figure 5 A schematic diagram of the 3σ principle in a method for constructing and retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0061] Figure 6 A neo4j graph data storage result in a method for constructing and retrieving a case library of operational tunnel defects provided by an embodiment of the present invention;

[0062] Figure 7 A schematic diagram of the relationship between the storage of disease case data in a method for constructing and retrieving a disease case library for an operating tunnel provided by an embodiment of the present invention;

[0063] Figure 8 A disease case library management page of a method for constructing and retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0064] Fig. 9 A disease retrieval page of a method for constructing and retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0065] Fig.10 A disease knowledge graph display page for a method for constructing and retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0066] Fig.11 A graph search result page of a method for constructing and retrieving a case library of operational tunnel defects provided by an embodiment of the present invention;

[0067] Fig.12 A disease search result details page of a method for constructing and retrieving a disease case library of an operating tunnel provided by an embodiment of the present invention;

[0068] Fig.13 A schematic diagram of the logical structure of a system for constructing and quickly retrieving a disease case library for operating tunnels provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present invention better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present invention are not shown or described in the specification, this is to avoid the core part of the present invention being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0070] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0071] The first embodiment of the present invention provides a method for constructing and quickly retrieving a case library of operational tunnel diseases. Figure 1 Provide detailed explanation.

[0072] like Figure 2 As shown, the system architecture design: The present invention adopts a layered architecture to design the software system, including a data processing layer, a knowledge graph construction layer, a retrieval service layer and a user interface layer. The data processing layer cleans, classifies and annotates the collected data; the knowledge graph construction layer constructs and maintains the knowledge graph; the retrieval service layer implements the retrieval algorithm; and the user interface layer provides a user-friendly operation interface.

[0073] like Figure 1 As shown, in step S100, operating tunnel disease case data is collected and pre-processed.

[0074] (1) Data collection: Data on operational tunnel disease cases are collected through records of the disease treatment process at the project site and the Internet, including tunnel type information, disease information, common causes of diseases, disease treatment measures, and related diseases caused when the disease occurs.

[0075] Information on the types of defects includes common cracks (including longitudinal cracks, transverse cracks, mesh cracks and other cracks of different forms), water leakage (such as point leakage, line leakage, surface leakage, etc.), lining spalling (small piece spalling, large area spalling, etc.) and other detailed records of various types of defects such as tunnel deformation, anchor failure, etc.

[0076] The defect location data is accurate to the specific mileage pile number of the tunnel, the location in the tunnel cross section (top, side wall, bottom, etc.) and the location range in the longitudinal section.

[0077] The time of disease occurrence is recorded, including year, month, day and even accurate to the hour, so as to analyze the development trend of the disease over time.

[0078] Assessment data of the severity of the disease, such as the width of the crack (the actual width value measured by high-precision measuring tools), the flow rate of leaking water (data obtained by professional flow monitoring equipment), the area of ​​lining peeling (the area value obtained by using image recognition technology or actual on-site measurement), etc.

[0079] Tunnel environmental parameters, such as the geological conditions of the area where the tunnel is located (including rock type, geological structure, soil properties, etc.) and climatic conditions (meteorological data such as temperature, humidity, rainfall, etc.).

[0080] (2) Data cleaning: Clean the collected data to remove duplicate data, erroneous data, and noise data.

[0081] like Figure 1 As shown, in step S200, the disease case data is classified and labeled according to the type and characteristics of the tunnel disease, and feature extraction is performed.

[0082] (1) Data classification and labeling: Data are classified and labeled according to the types and characteristics of tunnel defects. A defect classification system is established to classify defects into cracks, water leakage, lining deformation, etc. For each type of defect, its specific type and degree are further subdivided. For example, crack defects can be divided into longitudinal cracks, transverse cracks, mesh cracks, etc., and their degree is evaluated based on the width and length of the cracks. When labeling data, standardized labeling methods and terminology are used to ensure data consistency and comprehensibility.

[0083] (2) Feature extraction: Use machine learning algorithms to extract key information that can reflect the characteristics of tunnel diseases from the original data. For example, for quantitative indicators, the K-means algorithm is used to group them according to the degree of data concentration (aggregation), and data with similar indicator ranges are classified to facilitate feature extraction.

[0084] like Figure 1As shown, in step S300, based on the processed disease case data, a tunnel disease knowledge graph is constructed by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating.

[0085] (1) Ontology modeling: Based on the domain knowledge and data characteristics of tunnel diseases, an ontology model of the tunnel disease knowledge graph is constructed. The ontology model defines the concepts, attributes, relationships, and classification system of tunnel diseases. For example, it defines entity concepts such as "tunnel", "disease", "occurrence location", and "treatment method", as well as the relationships between them such as "occurrence", "cause", and "treatment method". Ontology modeling uses professional ontology modeling tools and methods to ensure the accuracy and completeness of the model.

[0086] Organize experts and technicians in the field of tunnel engineering to jointly develop an ontology model for the knowledge graph of tunnel diseases. The ontology model should cover all relevant concepts, attributes and relationships of tunnel diseases and comply with industry standards and specifications.

[0087] (2) Knowledge extraction: Extract knowledge from the processed data and fill it into the knowledge graph. Knowledge extraction includes entity recognition, relationship extraction, and attribute extraction. Entity recognition refers to identifying entities related to tunnel diseases from text data, such as cracks, water leakage points, etc.; relationship extraction refers to determining the relationship between entities, such as "cracks are located on the tunnel wall" and "water leakage causes lining corrosion", etc.; attribute extraction refers to extracting attribute information of entities, such as the length, width, and occurrence time of cracks. Knowledge extraction uses a combination of natural language processing technology and image recognition technology to improve the accuracy and efficiency of extraction.

[0088] In this embodiment, the K-means algorithm is combined with the 3σ principle to extract knowledge from the processed data. At the same time, manual review and verification are combined to ensure that the extracted knowledge is accurate.

[0089] If the quantitative indicators in the tunnel characteristics are directly extracted, the obtained data is not knowledge and cannot be used for feature calculation in the later stage, because it is difficult to match the quantitative indicators between cases. Therefore, it is particularly important to mine the data value of the quantitative indicators. The K-means algorithm groups the data according to the degree of concentration (aggregation) of the data set, and classifies the data with similar indicator ranges to facilitate knowledge extraction.

[0090] The principles and steps of data aggregation through K-merns are as follows:

[0091] First, randomly select K samples as the initial cluster centers, then calculate the distance from each sample to each cluster center, and assign the sample to the cluster with the closest distance. Next, recalculate the center of each cluster, that is, the mean of all samples in the cluster. Repeat the above process until the cluster center no longer changes or the preset number of iterations is reached.

[0092] 1. Select Manhattan distance to calculate the distance metric

[0093]

[0094] 2. Cluster center update

[0095] Let the kth cluster be C k , which contains a collection of samples The number of samples in this cluster is n k .

[0096] Calculate the center of the kth cluster

[0097]

[0098] 3. Algorithm stop condition judgment

[0099] Stop when the cluster center no longer changes. For all cluster centers If in two consecutive iterations, Where t represents the number of iterations, then the algorithm stops.

[0100] The algorithm stops when the preset number of iterations reaches the preset maximum value T.

[0101] The K-means algorithm optimizes the clustering results through continuous iteration, so that the final clustering can meet the goal of making the samples in each cluster as similar as possible, and the samples in different clusters as different as possible. The K-means data aggregation principle and data processing before and after are as follows: Figure 3 , Figure 4 shown.

[0102] There may be abnormal data that is too large or too small in the aggregated data. In order to remove the outliers in the aggregation results, the 3σ principle is used to eliminate the outliers with large deviations in the grouped data and complete the setting of the upper and lower limits of each group.

[0103] Figure 5 The percentage shown in is the probability that the data falls within the interval, which is given by Figure 5 It can be seen that within the range of plus or minus one times sigma, the probability of this interval is the largest, reaching 34.1%, while the probability of the interval outside the range of plus or minus three times sigma is the smallest, only 0.1%. Therefore, 3sigma is usually used as the dividing node.

[0104] (3) Knowledge fusion: The knowledge extracted from different sources is integrated to eliminate duplicate and contradictory information and build a unified knowledge graph. Knowledge fusion uses knowledge graph fusion algorithm and ontology mapping technology to achieve semantic alignment and integration between different knowledge sources. For example, information on the same disease is integrated to ensure that the information in the knowledge graph is complete and accurate.

[0105] (4) Knowledge storage and update: The constructed knowledge graph is stored in the database and indexed for easy retrieval. Graph database technologies, such as Neo4j and JanusGraph, are used to store the knowledge graph. At the same time, a knowledge graph update mechanism is established to update the data in the knowledge graph regularly or in real time to ensure the timeliness and accuracy of the knowledge.

[0106] The extracted knowledge is integrated to eliminate duplicate and contradictory information. The integrated knowledge is stored in a graph database and indexed for easy retrieval. Neo4j is used to store the data.

[0107] Data storage example:

[0108] 1. Create a node: Use the Cypher query language to create a node. For example, to create a node representing "tunnel disease", you can use the following statement:

[0109] CREATE(:TunnelDisease{name:'Tunnel disease'}).

[0110] 2. Create a relationship: Establish a relationship between nodes. For example, to create a "belongs to" relationship between a "Disease Type" node and a "Tunnel Disease" node, you can use the following statement:

[0111] MATCH(a:TunnelDisease),(b:DiseaseType{name:'crack'})CREATE(a)-[:BELONGS_TO]->(b).

[0112] 3. Create an index: To improve retrieval efficiency, you can create an index on a specific attribute. For example, to create an index on the "name" attribute of the "tunnel disease" node, you can use the following statement:

[0113] CREATE INDEX ON:TunnelDisease(name).

[0114] Through the above steps, the integrated knowledge can be stored in the Neo4j graph database, and can be easily retrieved and queried. Figure 6 shown.

[0115] Knowledge graph update and maintenance: Establish a knowledge graph update mechanism to update the data in the knowledge graph regularly or in real time. As new disease cases and treatment methods continue to emerge during tunnel operation, this information should be added to the knowledge graph in a timely manner to maintain the timeliness and accuracy of the knowledge graph.

[0116] In the knowledge graph constructed in this embodiment, the disease case data storage relationship is as follows: Figure 7 As shown in the figure, the disease case library management page is as follows Figure 8 As shown in the figure, the disease search page is as follows Fig. 9 As shown in the figure, the disease knowledge graph display page is as follows Fig.10 As shown in the figure, the graph search result page is as follows Fig.11 As shown in the figure, the disease search result details page is as follows Fig.12 shown.

[0117] like Figure 1 As shown, in step S100, natural language processing technology is used to perform semantic analysis and understanding on the query statement input by the user, and the query statement is converted into a query path in the knowledge graph. Then, path search and matching are performed in the knowledge graph, and semantically consistent retrieval results are returned and displayed to the user.

[0118] (1) Retrieval algorithm design: Utilize the semantic information of the knowledge graph to implement semantic-based retrieval. Use natural language processing technology to perform semantic analysis and understanding on the query statement entered by the user, convert it into a query path in the knowledge graph, and then perform path search and matching in the knowledge graph. For example, if the user enters "find cracks with a length greater than 1 meter at the top of the tunnel and their treatment methods", the system can understand the user's semantic needs and find the qualified crack entities and related treatment method information in the knowledge graph.

[0119] The core algorithms of semantic search include:

[0120] TF-IDF (Term Frequency-Inverse Document Frequency): Calculate the frequency of keywords in disease knowledge and multiply it by the frequency in the disease knowledge set. TF-IDF can help us capture important keywords in disease knowledge, thereby improving search accuracy.

[0121] TF-IDF t,d =TF t,d ×IDF t

[0122] That is, the TF-IDF value of term t in document d is equal to the term frequency and the inverse document frequency TF t，d The product of .

[0123] By calculating the TF-IDF value, we can determine which keywords are important for diseases, thereby improving the precision and accuracy of searches in semantic searches.

[0124] (2) Display of search results: Display the search results to users in an intuitive and clear manner. Use charts, images and other forms to display information such as the location, characteristics, causes, and treatment methods of the disease. For example, for crack-type diseases, the location and length of the cracks can be marked on the 3D model of the tunnel, and pictures of the cracks and related treatment suggestions can be displayed at the same time.

[0125] The construction and rapid retrieval technology of the operating tunnel disease case library of the present invention has important practical application value and broad development prospects, and will have a positive impact on the development of the tunnel engineering field. It has the following innovative features:

[0126] (1) Comprehensive application of multiple technologies

[0127] The present invention integrates multiple advanced technologies such as data mining technology and knowledge graph technology and applies them to the field of operating tunnel disease management, realizing an integrated solution for data collection, processing, storage and retrieval.

[0128] (2) Innovation of intelligent retrieval algorithms based on knowledge graphs

[0129] A keyword and semantic-based retrieval algorithm is designed to make full use of the semantic information of the knowledge graph to realize intelligent retrieval function. The retrieval algorithm is efficient and accurate and can meet the needs of practical applications.

[0130] (3) Data-driven innovation in tunnel disease management model

[0131] It has changed the traditional tunnel disease management model that relies on experience and manual inspections, and adopted a data-driven approach to achieve accurate management and prediction of tunnel diseases through the analysis and mining of large amounts of disease data.

[0132] (4) Innovation in user-friendly interactive interface design

[0133] An intuitive and clear user interface is designed to facilitate users to query, analyze and manage data. The search results are displayed in a variety of visual ways such as charts and images to improve the user experience.

[0134] Corresponding to the above-disclosed method for constructing and quickly retrieving a case library of operational tunnel diseases, the embodiment of the present invention further discloses a system for constructing and quickly retrieving a case library of operational tunnel diseases, such as Fig.13 As shown, it specifically includes:

[0135] Data collection module, used to collect and pre-process the disease case data of operating tunnels;

[0136] The data processing module is used to classify and label the tunnel disease case data according to the type and characteristics of the tunnel disease, and to extract features;

[0137] The knowledge graph construction module is used to construct a tunnel disease knowledge graph based on the processed disease case data by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating;

[0138] The retrieval and display module uses natural language processing technology to perform semantic analysis and understanding on the query statements input by users, convert the query statements into query paths in the knowledge graph, and then perform path search and matching in the knowledge graph, return semantically consistent retrieval results, and display the retrieval results to users.

[0139] It should be noted that the detailed description of a system for constructing and quickly retrieving a case library of operational tunnel defects provided in an embodiment of the present invention can be referred to the relevant description of a method for constructing and quickly retrieving a case library of operational tunnel defects provided in an embodiment of the present invention, which will not be repeated here.

[0140] In addition, an embodiment of the present invention further provides an electronic device, comprising: a processor and a memory; the memory is used to store one or more program instructions; the processor is used to run one or more program instructions to execute the steps of a method for constructing and quickly retrieving a case library of operating tunnel defects as described in any of the above items.

[0141] It should be noted that, for a detailed description of an electronic device provided in an embodiment of the present invention, reference can be made to the relevant description of a method for constructing and quickly retrieving a case library of operational tunnel defects provided in an embodiment of the present invention, which will not be repeated here.

[0142] In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of a method for constructing and quickly retrieving a case library of operating tunnel defects as described in any of the above items are implemented.

[0143] It should be noted that, for a detailed description of a computer-readable storage medium provided in an embodiment of the present invention, reference can be made to the relevant description of a method for constructing and quickly retrieving a case library of operational tunnel defects provided in an embodiment of the present invention, which will not be repeated here.

[0144] Those skilled in the art will appreciate that all or part of the functions of the various methods in the above-mentioned embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to implement the above-mentioned functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above-mentioned functions can be implemented. In addition, when all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash disk or mobile hard disk, and can be downloaded or copied and saved in the memory of the local device, or the system of the local device is updated, and when the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0145] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A method for constructing and quickly retrieving a case library of operating tunnel diseases, characterized in that: The method comprises: Collect the disease case data of operating tunnels and pre-process the disease case data; According to the types and characteristics of tunnel diseases, the disease case data is classified and labeled, and features are extracted; Based on the processed disease case data, a tunnel disease knowledge graph is constructed by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating; Using natural language processing technology, we perform semantic analysis and understanding on the query statements entered by users, convert the query statements into query paths in the knowledge graph, and then perform path search and matching in the knowledge graph, return semantically consistent retrieval results, and display the retrieval results to users.

2. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: Collect data on operational tunnel disease cases, including: The disease case data includes disease type, disease location, disease occurrence time, disease severity, and tunnel environmental parameters.

3. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: Preprocess the disease case data, including: Clean the collected data to remove duplicate data, erroneous data and noise data.

4. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: According to the types and characteristics of tunnel diseases, the disease case data is classified and labeled, and feature extraction is performed, including: Data classification and labeling: Establish a disease classification system to classify diseases; for each type of disease, further subdivide the specific type and degree; when labeling data, use standardized labeling methods and terminology to ensure data consistency and comprehensibility; Feature extraction: Use machine learning algorithms to extract key information that can reflect the characteristics of tunnel diseases from raw data.

5. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: Based on the processed disease case data, a tunnel disease knowledge graph is constructed by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating, including: Ontology modeling: Based on the domain knowledge and data characteristics of tunnel diseases, an ontology model of the tunnel disease knowledge graph is constructed. The ontology model defines the concepts, attributes, relationships and classification system of tunnel diseases. Knowledge extraction: extract knowledge from processed data and fill it into the knowledge graph; knowledge extraction includes entity recognition, relationship extraction and attribute extraction; entity recognition refers to identifying entities related to tunnel diseases from text data; relationship extraction refers to determining the relationship between entities; attribute extraction refers to extracting attribute information of entities; knowledge extraction adopts a method that combines natural language processing technology and image recognition technology; Knowledge fusion: Integrate knowledge extracted from different sources, eliminate duplicate and contradictory information, and build a unified knowledge graph. Knowledge fusion uses knowledge graph fusion algorithm and ontology mapping technology to achieve semantic alignment and integration between different knowledge sources. Knowledge storage and update: Store the constructed knowledge graph in the database and create an index for retrieval; use graph database technology to store the knowledge graph; establish an update mechanism for the knowledge graph to update the data in the knowledge graph regularly or in real time.

6. A method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 5, characterized in that: The knowledge extraction specifically includes: The K-means algorithm combined with the 3σ principle is used to extract knowledge from the processed data, and manual review and verification are combined to ensure that the extracted knowledge is accurate; If the quantitative indicators in the tunnel characteristics are directly extracted, the obtained data is not knowledge and cannot be used for feature calculation in the later stage. The K-means algorithm is used to group the data according to the degree of aggregation in the data set, and the data with similar indicator ranges are classified to facilitate knowledge extraction; The principles and steps of data aggregation through K-merns are as follows: First, randomly select K samples as the initial cluster centers, then calculate the distance from each sample to each cluster center, and assign the sample to the cluster with the closest distance; then recalculate the center of each cluster, that is, the mean of all samples in the cluster; repeat the corresponding process until the cluster center no longer changes or the preset number of iterations is reached; If there are abnormal data that are too large or too small in the aggregated data, in order to remove the abnormal values ​​in the aggregation results, the 3σ principle is used to eliminate the abnormal values ​​with large deviations in the grouped data and complete the setting of the upper and lower limits of each group.

7. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 5, characterized in that: The knowledge storage and updating specifically include: Data storage includes: Create nodes: Use Cypher query language to create nodes; Create relationship: establish the relationship between nodes; Create index: Create an index on a specific attribute; The integrated knowledge is stored in the Neo4j graph database and can be easily retrieved and queried.

8. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: Perform path search and matching in the knowledge graph and return semantically consistent search results, including: The core algorithms of semantic search include: TF-IDF: Calculate the frequency of keywords in disease knowledge and multiply it by the frequency in the disease knowledge set; TF-IDF t，d =TF t，d ×IDF t That is, the TF-IDF value of term t in document d is equal to the term frequency IDF t and Inverse Document Frequency TF t，d The product of By calculating the TF-IDF value, it is possible to determine which keywords are important keywords for the disease, thereby improving the precision and accuracy of the search in semantic search.

9. The method for constructing and quickly retrieving a case library of operating tunnel diseases as claimed in claim 1, characterized in that: Display the search results to the user, including: Design a friendly user interface to present the search results to users in an intuitive and clear manner; use a variety of forms including charts and images to display information including the location, characteristics, and treatment methods of the disease, making it easy for users to view and understand.

10. A system for constructing and quickly retrieving a case library of operational tunnel diseases, characterized in that: The system comprises: Data collection module, used to collect and pre-process the disease case data of operating tunnels; The data processing module is used to classify and label the tunnel disease case data according to the type and characteristics of the tunnel disease, and to extract features; The knowledge graph construction module is used to construct a tunnel disease knowledge graph based on the processed disease case data by adopting technologies including ontology modeling, knowledge extraction, knowledge fusion, and knowledge storage and updating; The retrieval and display module uses natural language processing technology to perform semantic analysis and understanding on the query statements input by users, convert the query statements into query paths in the knowledge graph, and then perform path search and matching in the knowledge graph, return semantically consistent retrieval results, and display the retrieval results to users.

Images