Geological disaster knowledge graph construction method and system and storage medium

Abstract

The invention relates to the technical field of geological disaster knowledge graph construction, in particular to a geological disaster knowledge graph construction method and system and a storage medium. Comprising the following steps: firstly, correcting multi-modal information by using an information correction method; according to the method, information is matched and corrected by establishing an expert knowledge base; the data information is subjected to statistical verification, and information with problems is corrected; accurate correction of original information is realized by combining expert knowledge correction and machine knowledge correction; secondly, proposing a geological disaster knowledge graph generation model; according to the model, a geological disaster knowledge graph is constructed by analyzing the dependency relationship between correction information and disaster types; and finally, providing a geological disaster knowledge graph optimization method. According to the method, correction auxiliary description information is used for carrying out association calculation to obtain auxiliary elements, so that the knowledge graph is optimized. By integrating the above methods, accurate construction of the geological disaster knowledge map is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of constructing a geological disaster knowledge graph, and specifically to a method, system and storage medium for constructing a geological disaster knowledge graph. Background Art

[0002] Geological disasters refer to geological actions or phenomena that are caused by natural or human factors and cause damage to the environment, including landslides, mudslides, earthquakes, collapses and other types. These disasters have a very serious impact on humans and the environment. Therefore, a deep understanding of the occurrence mechanism, influencing factors and their interrelationships of geological disasters is crucial for disaster prevention, risk management and emergency response.

[0003] With the continuous development of computer technology, geologists have gradually applied knowledge graph technology to the field of geological hazards. Knowledge graph is a technology that organizes and represents knowledge in a graphical way, establishing the relevance of knowledge through nodes (entities) and edges (relationships). The advantage of knowledge graph is that it can integrate a large amount of heterogeneous data into a unified platform, thus forming a knowledge network from a global perspective.

[0004] In the field of geological disasters, the construction of knowledge graphs can effectively integrate various information related to geological disasters, including: types of disasters, such as the definition, characteristics and occurrence conditions of various geological disasters such as landslides, mudslides and earthquakes; cause analysis, how natural factors (such as rainfall, earthquakes) and human factors (such as land use changes, construction activities) lead to geological disasters; influencing factors, the impact of different geological environments, climatic conditions and human activities on the occurrence of disasters; for the above-mentioned contents, professionals use a systematic approach to collect, organize and analyze data based on different information sources to construct a comprehensive geological disaster knowledge graph; however, the reliability of information cannot be guaranteed in this process; the existing technology lacks an effective information correction mechanism, which will lead to errors in the construction of the geological disaster knowledge graph and an accurate geological disaster knowledge graph cannot be obtained.

[0005] To this end, the present invention proposes a method, system and storage medium for constructing a geological disaster knowledge graph. Summary of the invention

[0006] The purpose of the present invention is to provide a method, system and storage medium for constructing a geological disaster knowledge graph for constructing a geological disaster knowledge graph, which can be used for functions such as geological disaster information query, prediction and disaster assessment; the specific functions include: first, obtaining multimodal information related to geological disasters through different channels, and correcting the multimodal information; for information correction, the present invention proposes a method for integrating expert knowledge correction and machine knowledge correction; wherein, in the expert knowledge correction process, an expert knowledge base is established to generate expert knowledge matching documents for different types of geological disasters, which are used to generate candidate expert correction information sets for the collected information; machine knowledge correction obtains different types of geological disaster description information by classifying and counting the collected multimodal information, and calculates the probability of information co-occurrence; cross-validates the information with high co-occurrence rate and performs corresponding correction to obtain a selected machine correction information set; synthesizes the candidate expert correction information set and the selected machine correction information set. The correction information set jointly generates a comprehensive correction information set to correct the original information data; secondly, the present invention proposes a geological disaster knowledge graph generation model; the model takes the corrected disaster description information and the initial geological disaster map as input data; the dependency relationship between information features is obtained through the relationship dependency mining layer, and a multi-condition constraint random forest model is used as the dependency mining model in the relationship dependency mining layer; the geological disaster knowledge graph generation layer constructs a geological disaster knowledge graph according to the structured geological disaster dependency matrix and the geological disaster map; finally, the present invention proposes a geological disaster knowledge graph optimization method; the method converts the geological disaster knowledge graph into a geological disaster multiple linked list, and explains the internal connection of the geological disaster knowledge graph in the form of a linked list; the addition of content in the geological disaster multiple linked list and the update of data are realized by using the auxiliary information elements in the corrected disaster auxiliary description information; and the optimized geological disaster knowledge graph is obtained by reversing the rule.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention proposes a method for constructing a geological disaster knowledge graph, comprising:

[0009] Collect multimodal information related to geological disasters and perform deduplication processing to obtain deduplicated multimodal information; wherein the multimodal information includes: text information, video information and audio information;

[0010] Further, the deduplicated multimodal information is classified according to the disaster type to obtain disaster description information and disaster auxiliary description information;

[0011] The classification process of the deduplicated multimodal information includes: obtaining description information corresponding to different types of geological disasters;

[0012] Further, keyword extraction is performed on the description information to obtain disaster type keywords;

[0013] Further, the disaster type keywords are vectorized to obtain disaster type matching word vectors;

[0014] Further, the deduplicated multimodal information is divided into a text data set, a video data set and an audio data set according to information type;

[0015] Further, all text descriptions are extracted from the text data set to obtain text description information; all video descriptions are extracted from the video data set to obtain video description information; all audio descriptions are extracted from the audio data set to obtain audio description information;

[0016] Further, feature extraction is performed on the text description information, the video description information and the audio data respectively to obtain text description features, video description features and audio description features;

[0017] Further, the similarities between the text description feature, the video feature, and the audio description feature and the disaster type matching word vector are calculated respectively to obtain text similarity, video similarity, and audio similarity;

[0018] Furthermore, when the text similarity, the video similarity and the audio similarity exceed a similarity matching threshold, the deduplicated multimodal information is used as the disaster description information corresponding to the disaster type; when the text similarity, the video similarity and the audio similarity do not exceed a similarity matching threshold, the deduplicated multimodal information is used as the disaster auxiliary description information.

[0019] Further, the disaster description information and the disaster auxiliary description information are corrected to obtain corrected disaster description information and corrected disaster auxiliary description information;

[0020] The correction of the disaster description information and the disaster auxiliary description information includes: an expert knowledge correction mechanism and a machine knowledge correction mechanism;

[0021] The expert knowledge correction mechanism corrects information through the knowledge of geologists, including: obtaining relevant knowledge of N geologists on geological hazards and building an expert knowledge base;

[0022] Further, generating expert knowledge matching text from the information in the expert knowledge base;

[0023] Further, a multi-layer knowledge matching index is established for the expert knowledge matching text;

[0024] Further, the disaster description information and the disaster auxiliary description information are matched with the expert knowledge matching text using the multi-layer knowledge matching index;

[0025] Further, information consistency calculation is performed based on the matching results to obtain an information consistency value;

[0026] Further, when the information consistency value does not exceed the consistency threshold, an information correction algorithm is used to obtain a candidate expert corrected information set;

[0027] The machine knowledge correction mechanism performs information correction through a machine statistical algorithm, including: using a named entity recognition algorithm to identify entity fields of the disaster description information and the disaster auxiliary description information;

[0028] Further, data statistics are performed on the entity fields, and the co-occurrence probabilities of the entity fields are calculated to obtain an entity field co-occurrence matrix;

[0029] Further, the disaster description information and the disaster auxiliary description information are screened according to the entity field co-occurrence matrix to obtain a geological disaster screening information set;

[0030] Further, cross-validating the geological disaster screening information set;

[0031] Further, when the cross-validation value of any information in the geological disaster screening information set is lower than the cross-validation threshold, the information that does not meet the conditions is corrected according to the same type of information to obtain a candidate machine corrected information set;

[0032] Furthermore, the candidate expert correction information set is merged with the candidate machine correction information set to obtain a comprehensive correction information set; when the candidate expert correction information set and the candidate machine correction information set have the same correction items; if the correction information of the correction items are the same, they are merged; if the correction information of the correction items are different, they are merged after regularization; and the comprehensive correction information set is used to correct the disaster description information and the disaster auxiliary description information.

[0033] Further, an initial geological disaster map is constructed; wherein the initial geological disaster map is represented by: {Node(GHT,GHAT),Edg(NULL)}; wherein Node() represents the initial node; GHT represents the disaster type; GHAT represents the geological disaster area type; Edg() represents the initial edge, and the initial value is NULL, that is, a null value;

[0034] Furthermore, the corrected disaster description information and the initial geological disaster map are input into a geological disaster knowledge graph generation model; the geological disaster knowledge graph generation model includes:

[0035] The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map;

[0036] A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features;

[0037] The relationship dependency mining layer is used to mine the dependency relationship of the structured geological hazard description features, including: establishing a structured geological hazard description feature matrix according to the structured geological hazard description features; wherein the structured geological hazard description feature matrix is ​​expressed as: Among them, GHTC i The hazard type characteristics represented by the i-th structural geological hazard description characteristics; GHATC i The geological hazard regional type characteristics represented by the i-th structured geological hazard description characteristics; It is represented as the jth related information feature vector of the i-th structured geological hazard description feature; M is represented as the total number of the structured geological hazard description features; N is represented as the total number of related information feature vectors in the structured geological hazard description features;

[0038] Input the structured geological hazard description feature matrix into the multi-conditional constrained random forest model to obtain the structured geological hazard dependency matrix;

[0039] Among them, the multi-conditional constraint random forest model includes: an input layer, used to receive the structured geological disaster description feature matrix; a tree generation layer, used to randomly generate multiple trees; wherein each tree independently selects feature data corresponding to different data disaster types; a condition constraint setting layer, used to set restrictions for each tree to constrain the branches of the tree; wherein the constraint rules of the condition constraint setting layer are set to: the occurrence reliability constraint between the disaster type and the geological disaster area type; the correlation constraint between the disaster type and the factors affecting the occurrence of the disaster; the impact constraint between the geological disaster area type and the disaster impact; a voting layer, used to summarize the analysis results of each tree; a dependency calculation layer, used to calculate the dependency based on the analysis results of the voting layer; and the output layer, used to output the structured geological disaster dependency matrix.

[0040] The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph;

[0041] A visualization output layer, used for visualizing the geological disaster knowledge graph;

[0042] Further, the geological disaster knowledge graph is optimized using the disaster correction auxiliary description information to obtain an optimized geological disaster knowledge graph;

[0043] The optimization process of the geological disaster knowledge graph includes:

[0044] The geological disaster knowledge graph is converted into a geological disaster multiple linked list; the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength;

[0045] Further, the correlation between the corrected disaster auxiliary description information and the data in the geological disaster multiple linked list is calculated;

[0046] Further, extracting auxiliary information elements of the corrected disaster auxiliary description information according to the correlation calculation result;

[0047] Further, the auxiliary information elements are integrated with the linked list nodes and the linked list multiple edges of the geological disaster multiple linked list, and the connection strength is updated and the connection direction is modified;

[0048] Furthermore, the updated geological disaster multiple linked list is reversely converted into the optimized geological disaster knowledge graph.

[0049] In a second aspect, the present invention proposes a system for constructing a geological disaster knowledge graph, the system comprising:

[0050] An information collection unit, used to collect information related to geological hazards from different channels;

[0051] An information classification unit, used to classify geological disaster information;

[0052] An information correction unit, used to correct the collected relevant information; wherein the information correction unit includes: an expert knowledge correction module and a machine knowledge correction module;

[0053] The expert knowledge correction module comprises:

[0054] Obtain relevant knowledge of geological hazards from N geologists and build an expert knowledge base;

[0055] Further, generating expert knowledge matching text from the information in the expert knowledge base;

[0056] Further, a multi-layer knowledge matching index is established for the expert knowledge matching text;

[0057] Further, matching the disaster description information and the disaster auxiliary description information with the expert knowledge matching text according to the multi-layer knowledge matching index;

[0058] Further, information consistency calculation is performed according to the matching result to obtain an information consistency value; when the information consistency value does not exceed the consistency threshold, an information correction algorithm is used to obtain a candidate expert corrected information set;

[0059] The machine knowledge correction module includes:

[0060] Using a named entity recognition algorithm to identify entity fields of the disaster description information and the disaster auxiliary description information;

[0061] Further, data statistics are performed on the entity fields, and the co-occurrence probabilities of the entity fields are calculated to obtain an entity field co-occurrence matrix;

[0062] Further, the disaster description information and the disaster auxiliary description information are screened according to the entity field co-occurrence matrix to obtain a geological disaster screening information set; the geological disaster screening information set is cross-validated; when the cross-validation value of any information in the geological disaster screening information set is lower than the cross-validation threshold, the information that does not meet the conditions is corrected according to similar information to obtain a candidate machine correction information set;

[0063] Furthermore, the candidate expert correction information set is merged with the candidate machine correction information set to obtain a comprehensive correction information set; when the candidate expert correction information set and the candidate machine correction information set have the same correction items; if the correction information of the correction items are the same, they are merged; if the correction information of the correction items are different, they are merged after regularization; and the comprehensive correction information set is used to correct the disaster description information and the disaster auxiliary description information.

[0064] A knowledge graph construction unit is used to construct a geological disaster knowledge graph; wherein the knowledge graph construction unit includes:

[0065] Construct an initial geological hazard map;

[0066] Inputting the corrected disaster description information and the initial geological disaster map into a geological disaster knowledge graph generation model; wherein the geological disaster knowledge graph generation model includes:

[0067] The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map;

[0068] A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features;

[0069] The relationship dependency mining layer is used to mine the dependency relationship of the structured geological disaster description features, including: establishing a structured geological disaster description feature matrix according to the structured geological disaster description features; inputting the structured geological disaster description feature matrix into a multi-conditional constrained random forest model to obtain a structured geological disaster dependency matrix;

[0070] The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph;

[0071] The visualization output layer is used to visualize the geological disaster knowledge graph.

[0072] A knowledge graph optimization unit is used to optimize the geological disaster knowledge graph; wherein the knowledge graph optimization unit includes:

[0073] The geological disaster knowledge graph is converted into a geological disaster multiple linked list; the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength;

[0074] Further, the correlation between the corrected disaster auxiliary description information and the data in the geological disaster multiple linked list is calculated;

[0075] Further, extracting auxiliary information elements of the corrected disaster auxiliary description information according to the correlation calculation result;

[0076] Further, the auxiliary information elements are integrated with the linked list nodes and the linked list multiple edges of the geological disaster multiple linked list, and the connection strength is updated and the connection direction is modified;

[0077] Furthermore, the updated geological disaster multiple linked list is reversely converted into an optimized geological disaster knowledge graph.

[0078] A visualization unit is used to visualize the generated geological disaster knowledge graph;

[0079] Maintenance unit, used to regularly update the geological disaster knowledge graph;

[0080] In the third aspect, the present invention proposes a storage medium for constructing a geological hazard knowledge graph, on which a program for constructing a geological hazard knowledge graph is stored. When the program for constructing a geological hazard knowledge graph is executed by a processor, the construction of any one of the geological hazard knowledge graphs in the first aspect or the second aspect is realized.

[0081] Compared with the prior art, the present invention has the following beneficial effects:

[0082] 1. The present invention proposes a method for correcting geological hazard information; the method comprehensively utilizes expert knowledge correction and machine knowledge correction; wherein, in the expert knowledge correction process, an expert knowledge base is established to generate expert knowledge matching texts for different types of geological hazards, which are used to generate candidate expert correction information sets for the collected information; machine knowledge correction obtains different types of geological hazard description information by classifying and counting the collected multimodal information, and calculates the information co-occurrence probability; cross-validates the information with high co-occurrence rate and performs corresponding correction to obtain a selected machine correction information set; the candidate expert correction information set and the selected machine correction information set are combined to jointly generate a comprehensive correction information set to correct the original information data; the accurate construction of the geological hazard knowledge graph is realized, thereby improving the availability of the geological hazard knowledge graph.

[0083] 2. The present invention proposes a geological disaster knowledge graph generation model; the model takes the corrected disaster description information and the initial geological disaster map as input data; obtains the dependency relationship between information features through the relationship dependency mining layer, and adopts a multi-condition constraint random forest model as the dependency mining model in the relationship dependency mining layer; the geological disaster knowledge graph generation layer constructs a geological disaster knowledge graph according to the structured geological disaster dependency matrix and the geological disaster map; this is conducive to the accurate construction of the geological disaster knowledge graph.

[0084] 3. The present invention proposes a method for optimizing a geological disaster knowledge graph; the method converts the geological disaster knowledge graph into a multiple linked list of geological disasters, and illustrates the internal connection of the geological disaster knowledge graph in the form of a linked list; by correcting the auxiliary information elements in the disaster auxiliary description information, the content in the multiple linked list of geological disasters is increased and the data is updated; and the optimized geological disaster knowledge graph is obtained by reversing the rules; the method optimizes the accuracy of the information of the geological disaster knowledge graph and enriches the information content of the geological disaster knowledge graph. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 A flowchart of a method for constructing a geological disaster knowledge graph provided by an embodiment of the present invention;

[0086] Figure 2 A structural diagram of a geological disaster knowledge graph construction system provided by an embodiment of the present invention;

[0087] Figure 3 A structural diagram of an information correction unit provided by an embodiment of the present invention;

[0088] Figure 4 A structural diagram of a multi-layer knowledge matching index provided by an embodiment of the present invention;

[0089] Figure 5 A schematic diagram of an initial geological disaster map provided by an embodiment of the present invention;

[0090] Figure 6 A structural diagram of a geological disaster knowledge graph generation model provided by an embodiment of the present invention;

[0091] Figure 7 A schematic diagram of a geological disaster knowledge graph provided by an embodiment of the present invention;

[0092] Figure 8 A schematic diagram of a multiple linked list of geological hazards provided by an embodiment of the present invention;

[0093] Fig. 9 A schematic diagram of an optimized geological disaster knowledge graph provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0094] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0095] Geological disasters are natural disasters that are caused by geological actions or changes in the geological environment and cause damage to humans and the natural environment. They include many types, such as earthquakes, volcanic eruptions, landslides, mudslides, collapses, and ground subsidence. In the process of studying these disasters, geologists are committed to revealing their causes, processes, and potential impacts.

[0096] In the study of geological disasters, due to the complex background factors and the diversity of geological disasters, geologists cannot rely on a single method or model to fully understand and analyze all types of geological disasters, making the research in the field of geological disasters very challenging. In recent years, with the continuous development of computer technology, geologists have applied computer technology to the field of geological disasters. Through research, it is found that combining knowledge graph technology with geological disasters can deeply explore the occurrence mechanism, influencing factors and their mutual relationship of geological disasters; knowledge graph is a technology that organizes and represents knowledge in a graphical way, and establishes the association of knowledge through nodes (entities) and edges (relationships). The advantage of knowledge graph is that it can integrate a large amount of heterogeneous data into a unified platform, thereby forming a knowledge network from a global perspective.

[0097] In the prior art, researchers achieve multiple tasks such as prediction, evaluation and management of geological disasters by establishing a geological disaster knowledge graph; whether these can achieve good results depends mainly on the construction of the geological disaster knowledge graph; the construction of the geological disaster knowledge graph collects a large amount of information, including: relevant literature, research reports and related data, etc., to form a huge knowledge base, and then analyzes the relationship between knowledge in the knowledge base to construct a knowledge graph; however, in this process, since the reliability of the information cannot be verified, the constructed geological disaster knowledge graph has certain risks; in order to solve the problem of data errors in the process of constructing the geological disaster knowledge graph, the present invention proposes a construction method, system and storage medium for a geological disaster knowledge graph to improve the accuracy of the geological disaster knowledge graph; the following will explain in detail from two embodiments.

[0098] Embodiment 1:

[0099] In the embodiments of the present application, the construction method and system of the geological disaster knowledge graph proposed by the present invention are used to realize the construction of the geological disaster knowledge graph; in order to illustrate the effectiveness of the present invention, in the first embodiment, earthquakes, landslides and mud-rock flows, three common types of geological disasters, are taken as examples to establish a geological disaster knowledge graph targeting earthquakes, landslides and mud-rock flows. The method and system proposed by the present invention are explained in the process of establishing the geological disaster knowledge graph; Figure 1 and Figure 2 The flowchart of the method of the present invention and the structure diagram of the system are respectively given; Figure 1 and Figure 2 As shown, the method steps of the present invention include: S10. Collecting multimodal information related to geological disasters; S20. Deduplication processing of the multimodal information; S30. Classifying the deduplication multimodal information; S40. Correcting the information; S50. Constructing a geological disaster knowledge graph; S60. Optimizing the geological disaster knowledge graph; Figure 2The construction system of the geological disaster knowledge graph given in the paper includes: information collection unit, information classification unit, information correction unit, knowledge graph construction unit, knowledge graph optimization unit, visualization unit and maintenance unit; the specific geological disaster knowledge graph construction process includes:

[0100] The information collection unit of the system is used to collect multimodal data on earthquakes, landslides and debris flows, corresponding to Figure 1 Step S10; wherein the information collection unit mainly obtains text information, video information and audio information related to earthquakes, landslides and debris flows through the Internet; the text information is mainly obtained from literature, forums, social media and official electronic reports; the video information is mainly obtained from news reports, social media videos and official geological report analysis videos; the audio information is mainly obtained from podcasts, news broadcasts and geologist interviews.

[0101] Further, the multimodal data is deduplicated through step S20 to obtain deduplicated multimodal data;

[0102] Further, the deduplicated multimodal data is classified into disaster description information and disaster auxiliary description information by using the information classification unit of the system, corresponding to step S30; wherein the disaster description information refers to the description information related to the specific disaster type in the collected multimodal information; the disaster auxiliary description information refers to the relevant information without directly describing the disaster type; for example, the disaster description information is about the epicenter location, magnitude and impact of a certain earthquake; the disaster auxiliary description information introduces the relevant geological structure, soil type and climate conditions;

[0103] The classification process of the deduplicated multimodal information includes:

[0104] Obtain descriptive information on geological hazards (earthquakes, landslides and debris flows);

[0105] Further, keyword extraction is performed on the description information to obtain disaster type keywords;

[0106] Furthermore, the disaster type keywords are vectorized to obtain disaster type matching word vectors; see Table 1, which shows the conversion process of earthquake, landslide and mud-rock flow description information to disaster type matching word vectors;

[0107] Table 1 Conversion of disaster type description vectors

[0108]

[0109] Table 1 shows the transformation process from "description information" to "disaster type matching word vector". In this process, Word2Vec technology is used, a neural network-based model that converts words into vectors of fixed dimensions to capture the semantic relationship between words. Word vectors are generated through the Skip-gram or CBOW (Continuous Bag of Words) model.

[0110] Further, the deduplicated multimodal information is divided into a text data set, a video data set, and an audio data set according to information type;

[0111] Further, all text descriptions are extracted from the text data set using NLP (natural language processing) technology to obtain text description information;

[0112] Extract all video descriptions from the video data set to obtain video description information; wherein the process of obtaining the video description information includes: obtaining video metadata of the videos in the video data set: each video usually contains some metadata (such as title, description, duration and release time, etc.); using a video processing library (such as OpenCV, FFmpeg and Moviepy, etc.) to analyze the video metadata and extract text information in the frame; organizing the extracted description information to create a structured data format (such as a list, a dictionary or a data frame);

[0113] Extract all audio descriptions from the audio data set to obtain audio description information; including: extracting audio metadata: each audio file usually contains some metadata (such as title, duration and sampling rate, etc.), and some audio processing libraries (such as pydub, mutagen, etc.) can be used to extract these metadata; using audio processing libraries: basic information of audio files can be read through these libraries; using speech recognition technology to transcribe audio into text to obtain audio description information.

[0114] Further, feature extraction is performed on the text description information, the video description information and the audio data respectively to obtain text description features, video description features and audio description features;

[0115] Further, the similarities between the text description feature, the video description feature, and the audio description feature and the disaster type matching word vector are calculated respectively to obtain text similarity, video similarity, and audio similarity;

[0116] The calculation formula for the text similarity, the video similarity and the audio similarity is expressed as:

[0117]

[0118] Among them, CosSim() represents the similarity calculation function, that is, cosine similarity; Represented as the text description feature; Represented as the video description feature; represented as the audio description feature; It represents the disaster type matching word vector; |||| represents the norm operator;

[0119] Furthermore, when the text similarity, the video similarity and the audio similarity exceed the similarity matching threshold; the deduplicated multimodal information is used as the disaster description information corresponding to the disaster type; when the text similarity, the video similarity and the audio similarity do not exceed the similarity matching threshold, the deduplicated multimodal information is used as the disaster auxiliary description information; wherein, in order to obtain more disaster description information in the embodiment of the present application, the similarity matching threshold is set to 0.6.

[0120] In the embodiment of the present application, the multimodal information after deduplication is classified into disaster description information and disaster auxiliary description information; in the classification process, the feature similarity of the word vector is matched with the characteristics of text information, video information and audio information and the disaster type, and the classification effect is achieved through the similarity threshold. It can not only improve the accuracy and availability of the geological disaster knowledge map, but also provide a solid data foundation for future disaster management and emergency response.

[0121] Further, the disaster description information and the disaster auxiliary description information are corrected by the information correction unit of the system to obtain the corrected disaster description information and the corrected disaster auxiliary description information, corresponding to step S40; wherein, Figure 3 As shown, the information correction unit includes: an expert knowledge correction module and a machine knowledge correction module;

[0122] The expert knowledge correction module includes: obtaining relevant knowledge of N geological experts on earthquakes, landslides and debris flows, and constructing an expert knowledge base; in the embodiment of the present application, relevant knowledge of 10 geologists on earthquakes, landslides and debris flows is obtained; the structure of the expert knowledge base includes: expert number attribute description, disaster type attribute description, disaster area type attribute description, occurrence mechanism attribute description, influencing factor attribute description and factor action attribute description; refer to Table 2, which illustrates the knowledge provided by the geological expert numbered P001 in the expert knowledge base;

[0123] Table 2 Part of the expert knowledge in the expert knowledge base Table 2 above gives the content description of each attribute under different disaster types. The content in the expert knowledge base is not only included here. A more detailed knowledge base can be established based on the knowledge provided by geological experts, and the technical solutions contained therein do not deviate from the present invention.

[0124] Further, generating expert knowledge matching text from the information in the expert knowledge base;

[0125] Furthermore, a multi-layer knowledge matching index is established for the expert knowledge matching text; wherein the multi-layer knowledge matching index includes: establishing attribute key values ​​according to disaster type, disaster area type, occurrence mechanism, influencing factors and factor effects; each key value is associated with the corresponding expert knowledge matching text; wherein the expert knowledge matching text stores content index values; specific information in the expert knowledge base can be queried according to the content index values; refer to Figure 4 , gives a visual description of the multi-layer knowledge matching index; from Figure 4 It can be seen that the multi-layer knowledge matching index is divided into four layers. The first layer is the attribute key value layer corresponding to the disaster type, where the attribute key value correspondence is: "attribute key value 1 <-> earthquake"; "attribute key value 2 <-> landslide"; "attribute key value 3 <-> mudslide"; the second layer is the attribute key value layer corresponding to the disaster area type, and the attribute key value correspondence is, for example, "attribute key value A1 <-> mountainous area" and "attribute key value A2 <-> coastal area"; the third layer is the relevant attributes of the disaster occurrence, including: occurrence mechanism, influencing factors and factor effects, and the corresponding attribute key values; the fourth layer is the link layer, which is linked to the expert knowledge base through the database structure to obtain the corresponding expert index; Figure 4 The first layer and the second layer are bidirectionally linked. The corresponding disaster area type can be queried by the disaster type, and the corresponding disaster type can also be queried by the disaster area type;

[0126] Further, matching the disaster description information and the disaster auxiliary description information with the expert knowledge matching text according to the multi-layer knowledge matching index;

[0127] Further, information consistency calculation is performed based on the matching results to obtain an information consistency value;

[0128] Further, information consistency calculation is performed according to the matching result to obtain an information consistency value; when the information consistency value does not exceed the consistency threshold, an information correction algorithm is used to obtain a candidate expert corrected information set;

[0129] The above information matching and consistency calculation process is described in the following example; for example, a description of a piece of information in the disaster description information is obtained as follows: "A strong earthquake occurred in mountain area A, with the epicenter located at longitude 114.3 and latitude 30.5, and a magnitude of 6.8. The earthquake was so strong that residents of several surrounding villages felt obvious shaking. The geological structure of the area is complex, and the earthquake caused violent movement of the earth's crust and triggered a chain reaction. Due to the occurrence of the earthquake, landslides occurred in many mountainous areas in the area, especially when the weather was clear and the soil was saturated with water, which made the risk of landslides high. The earthquake also caused some buildings to collapse, resulting in road closures and serious traffic obstruction. Influencing factors: (1) Geological structure: The geological structure of the area is loose and easily affected by earthquakes; (2) Distance from the epicenter: The epicenter is close to nearby villages, which aggravates the impact; (3) Focal depth: The epicenter is shallow, resulting in significant ground shaking and increasing the occurrence of secondary disasters. Description of secondary disasters: Landslide: The landslide accident caused by the earthquake caused dozens of roads to be blocked and some villages lost contact with the outside world. Building damage: Many buildings were severely damaged due to the shaking, and some houses collapsed, trapping people. "Based on the above information, a keyword list can be obtained;

[0130] Table 3 Example keyword list

[0131]

[0132] According to the disaster type ("earthquake, landslide") in Table 3, attribute key value 1 and attribute key value 2 are obtained according to the first layer of the multi-layer knowledge matching index;

[0133] Furthermore, the consistency calculation formula is used to calculate the regional type in Table 3. The consistency of the second-layer disaster regional type is indexed according to the attribute key value 1 and the attribute key value 2. When the consistency exceeds the consistency threshold, it is matched to the third layer. If it does not exceed, the corresponding keyword of the regional type in Table 3 is recorded as modification information.

[0134] According to Table 3, the keyword corresponding to the region type is "region type", which is consistent with the second-level index;

[0135] Further, match to the third layer; obtain the corresponding knowledge of the expert knowledge base through the attribute key values ​​of the occurrence mechanism, influencing factors and factor effects of the third layer; calculate the consistency of the acquired knowledge with the keywords corresponding to the epicenter information, impact description, influencing factors, secondary disasters and meteorological conditions in Table 3;

[0136] In the consistency calculation process, the consistency calculation formula between the keyword "clear weather" corresponding to the meteorological conditions in Table 3 and the expert knowledge is: Where C represents the consistency value (usually ranging from 0 to 1, 1 represents complete consistency, and 0 represents complete inconsistency); P represents "clear weather"; Q represents the disaster type; W(P,Q) represents the consistency weight between "clear weather" and the disaster type; P′ represents all possible weather conditions under the disaster type;

[0137] Refer to Table 4, where the consistency weights are given in the case of “geology and landslides”;

[0138] Table 4 Distribution of meteorological consistency weights for geological and landslide conditions

[0139]

[0140] Calculate the consistency value:

[0141] Further, the calculated consistency value does not exceed the consistency threshold; therefore, the keyword corresponding to the meteorological condition in Table 3 is error information; wherein, in the embodiment of the present application, the consistency threshold for the meteorological condition is set to 0.7; in summary, the calculation of the consistency value for other items of information is the same as the above principle, and the consistency thresholds for matching different information are all set to high thresholds to ensure the accuracy of information verification;

[0142] An information correction algorithm is used to correct the erroneous information; wherein the information correction algorithm mainly includes the following steps:

[0143] Step 1: Query the knowledge base: retrieve information similar to the error from the expert knowledge base and compare;

[0144] Step 2: Matching degree calculation: Calculate the matching degree between the marked information and the relevant information in the knowledge base and quantify it;

[0145] Step 3: Find matching high-consistency information from the knowledge base and replace the original error information;

[0146] Step 4: Update the original information according to the correction strategy to form a candidate expert correction information set;

[0147] The machine knowledge correction module includes:

[0148] A named entity recognition algorithm is used to identify entity fields of the disaster description information and the disaster auxiliary description information; wherein the named entity recognition algorithm uses a dictionary matching method and the process is as follows:

[0149] Matching is performed through a predefined entity dictionary, which is suitable for applications in earthquake, landslide and mud-rock flow scenarios; wherein the predefined content of the entity dictionary is shown in Table 5;

[0150] Table 5 Predefined contents of entity dictionary

[0151]

[0152] Furthermore, Jieba is used to perform word segmentation on the disaster description information and the disaster auxiliary description information; Jieba is a very popular Chinese word segmentation tool, mainly used for word segmentation of Chinese texts. It combines dictionary-based rules and statistical-based models to efficiently and accurately complete word segmentation tasks.

[0153] Furthermore, by traversing the word segmentation results, the predefined entity dictionary is used for matching to identify the entity fields in the text;

[0154] Furthermore, data statistics are performed on the entity fields, and the co-occurrence probabilities of the entity fields are calculated to obtain an entity field co-occurrence matrix; wherein the acquisition process of the entity field co-occurrence matrix includes: creating a co-occurrence matrix, the rows and columns of which are all the identified entities;

[0155] Furthermore, each group of identified entities is traversed to count the number of co-occurrences between them;

[0156] Furthermore, each value in the co-occurrence matrix is ​​divided by the total number of entities to obtain the entity field co-occurrence matrix;

[0157] Further, the disaster description information and the disaster auxiliary description information are screened according to the entity field co-occurrence matrix to obtain a geological disaster screening information set; wherein, when the co-occurrence rate in the entity field co-occurrence matrix is ​​greater than a preset threshold, the entity fields are associated; for example, if the co-occurrence rate between "earthquake" and "building collapse" is higher than the threshold, it can be considered that the two entities are significantly associated; in the embodiment of the present application, the preset threshold is set to 0.5;

[0158] Further, the geological disaster screening information set is cross-validated; wherein, in the embodiment of the present application, a K-fold cross-validation method is used to divide the geological disaster screening information set into K parts;

[0159] Further, when the cross-validation value of any information in the geological disaster screening information set is lower than the cross-validation threshold, the information that does not meet the conditions is corrected according to the same type of information to obtain a candidate machine corrected information set;

[0160] For example, according to the following entities: Entity A: Earthquake; Entity B: Building Collapse; Entity C: Landslide; Entity D: Flood.

[0161] Calculate the entity field co-occurrence matrix of the above entities; as shown in Table 6:

[0162] Table 6 Example entity field co-occurrence matrix description

[0163] entity earthquake Building collapse landslide flood earthquake 1 0.6 0.4 0.2 Building collapse 0.6 1 0.3 0.1 landslide 0.4 0.3 1 0.5 flood 0.2 0.1 0.5 1

[0164] Based on the preset threshold of 0.5, entities with significant correlation will be screened out according to the co-occurrence rate; including: earthquake and building collapse: the co-occurrence rate is 0.6 (greater than 0.5); earthquake and landslide: the co-occurrence rate is 0.4 (less than 0.5, not retained); building collapse and landslide: the co-occurrence rate is 0.3 (less than 0.5, not retained); landslide and flood: the co-occurrence rate is 0.5 (equal to 0.5, retained); other combinations are all less than the threshold.

[0165] Based on the above screening, a geological disaster screening information set is obtained, including: earthquakes and building collapse; landslides and floods.

[0166] Perform K-fold cross validation; set K=2 and divide the screening information set into 2 parts; Part 1: earthquake and building collapse; Part 2: landslide and flood.

[0167] The two parts are trained and verified separately, and different training data are used to verify the effect of the model;

[0168] Evaluate the cross-validation value: The cross-validation value of part 1 is 0.4 (below the threshold, the cross-validation threshold is set to 0.5); The cross-validation value of part 2 is 0.7 (above the threshold)

[0169] Since the cross-validation value of part 1 is lower than the threshold of 0.5, the information in this part needs to be corrected, and it can be adjusted based on similar information (such as other information related to earthquakes and building collapse);

[0170] Further, the candidate expert correction information set is merged with the candidate machine correction information set to obtain a comprehensive correction information set;

[0171] Furthermore, when the candidate expert correction information set and the candidate machine correction information set have the same correction items; if the correction information of the correction items are the same, they are merged; if the correction information of the correction items are different, they are merged after regularization;

[0172] Among them, in the embodiment of the present application, the regularized formula is set to: Among them, C r Represented as the regularized information; C c Represents the candidate expert correction information; C m Indicates machine correction information; w c and w m Represents different regularization weights, which can be set based on historical accuracy, reliability, or other indicators;

[0173] Furthermore, the disaster description information and the disaster auxiliary description information are corrected using the comprehensive correction information set.

[0174] The information correction process is as follows:

[0175] The candidate expert correction information set includes: Item 1: "The landslide may cause road interruption and traffic diversion is required immediately."; Item 2: "Residents in the affected area should be prepared to evacuate.";

[0176] The candidate machine correction information set includes,item 1: "The landslide will cause road closures, and it is recommended to implement traffic control immediately.";item 2: "In the affected area, residents need to evacuate in time to ensure safety.";

[0177] Merging process: Merge the candidate expert correction information set with the candidate machine correction information set to obtain a comprehensive correction information set.

[0178] (1) Check the same correction items

[0179] For item 1: Expert information: "Landslides may cause road interruptions and traffic diversion is required immediately."; Machine information: "Landslides may cause road closures and it is recommended to implement traffic control immediately."; The information is different and needs to be regularized.

[0180] The regularization result is: C r = "Landslide may lead to slope failure. It is recommended to implement traffic control and diversion immediately.";

[0181] (2) For the merging of item 2, since the two pieces of information are similar, they can be merged directly:

[0182] The combined result is: "Residents in the affected areas should be prepared to evacuate and need to evacuate in time to ensure safety.".

[0183] The final comprehensive correction information set is:

[0184] Item 1: "The landslide may cause road interruption. It is recommended to implement traffic control and traffic diversion immediately.";

[0185] Item 2: "Residents in the affected areas should be prepared to evacuate and should evacuate in a timely manner to ensure safety.";

[0186] Use the comprehensive correction information set to make corrections; Next, apply the comprehensive correction information set to the correction of the disaster description information and the auxiliary description information. Hypothetical disaster description information: "The recent landslide event caused many roads to be closed and residents need to evacuate.";

[0187] Correction process: Update the original description using information from the comprehensive correction information set to: "Recent landslides have caused the closure of multiple roads. Immediate traffic control and diversion are recommended. Residents in the affected areas should be prepared to evacuate and need to evacuate in a timely manner to ensure safety."

[0188] The final corrected disaster description information is: "The recent landslide event has caused the closure of multiple roads. It is recommended to immediately implement traffic control and traffic diversion. Residents in the affected areas should be prepared to evacuate and need to evacuate in time to ensure safety."

[0189] In the embodiment of the present application, a geological disaster information correction method is used to correct the information; the method performs comprehensive information correction through expert knowledge correction and machine knowledge correction; wherein, the expert knowledge correction establishes an expert knowledge base based on multiple geological experts; and matches the acquired information with the expert knowledge through indexing, and generates a candidate expert correction information set according to the matching result; the machine knowledge correction performs data statistics on the entity information of the collected information, screens the information according to the co-occurrence relationship of the entity fields, and cross-validates the screened information to obtain a machine correction information set; the original information is corrected according to the candidate expert correction information set and the machine correction information set; this can not only improve the accuracy and reliability of the information, but also enhance the richness and adaptability of the knowledge graph, and provide a sufficient data basis for the construction of the geological disaster knowledge graph.

[0190] Furthermore, the knowledge graph construction unit of the system is used to construct a knowledge graph of geological disasters related to landslides, debris flows and earthquakes, corresponding to step S50; the specific process includes:

[0191] Construct an initial geological hazard map;

[0192] The initial geological disaster map is represented by: {Node(GHT,GHAT),Edg(NULL)}; wherein Node() represents the initial node; GHT represents the disaster type; GHAT represents the geological disaster area type; Edg() represents the initial edge, and the initial value is NULL, that is, a null value; see Figure 5 , giving an initial geological disaster map of the embodiment of the present application;

[0193] The corrected disaster description information and the initial geological disaster map are input into the geological disaster knowledge graph generation model; wherein, refer to Figure 6 , the geological disaster knowledge graph generation model includes:

[0194] The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map;

[0195] A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features;

[0196] The relationship dependency mining layer is used to mine the dependency relationship of the structured geological disaster description features, including: establishing a structured geological disaster description feature matrix according to the structured geological disaster description features; inputting the structured geological disaster description feature matrix into a multi-conditional constrained random forest model to obtain a structured geological disaster dependency matrix;

[0197] The multi-conditional constraint random forest model includes: an input layer for receiving the structured geological disaster description feature matrix; a tree generation layer for randomly generating multiple trees; wherein each tree independently selects feature data corresponding to different data disaster types; a condition constraint setting layer for setting restriction conditions for each tree to constrain the branches of the tree; wherein the constraint rules of the condition constraint setting layer are set as follows: the occurrence reliability constraint between the disaster type and the geological disaster area type; the correlation constraint between the disaster type and the factors affecting the occurrence of the disaster; the impact constraint between the geological disaster area type and the disaster impact;

[0198] The rule constraints for the conditional constraint setting layer are as follows:

[0199] (1) Reliability constraint between disaster type and geological disaster area type: P(DR)≥θ1; where P(DR) represents the probability of occurrence of disaster type D given area type R; θ1 represents the set reliability threshold, which represents the minimum probability requirement for the occurrence of the disaster;

[0200] (2) Correlation constraint between disaster type and influencing factors of disaster occurrence: I(D, F) ≥ θ2; where I(D, F) represents the correlation coefficient between disaster type D and influencing factor F; θ2 represents the set correlation threshold, which represents the minimum requirement for correlation to be achieved;

[0201] (3) Constraints between geological hazard area types and hazard impacts: C(R,I D )≥θ3; where C(R,I D ) represents the impact of regional type D on disaster I D θ3 represents the impact threshold, which represents the minimum requirement for the impact level.

[0202] In an embodiment of the present application, the correlation between geological disaster attributes is obtained through the relationship dependency mining layer of the geological disaster knowledge graph generation model; wherein, a multi-condition constraint random forest model is established in the relationship dependency mining layer to perform relationship mining; the multi-condition constraint random forest model is based on the random forest model, and a condition constraint setting layer is set; the actual attribute relationship in the geological disaster is set as a rule to filter the connection between unrelated attributes; the multi-condition constraint random forest model can significantly enhance the construction quality and practicality of the geological disaster knowledge graph.

[0203] The voting layer is used to summarize the analysis results of each tree; the dependency calculation layer is used to calculate the dependency according to the analysis results of the voting layer; and the output layer is used to output the structured geological hazard dependency matrix.

[0204] The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph;

[0205] The visualization output layer is used to visualize the geological disaster knowledge graph; see Figure 7 ;

[0206] Further, the knowledge graph optimization unit of the system is used to optimize the geological disaster knowledge graph to obtain an optimized geological disaster knowledge graph, corresponding to the above step S60; the geological disaster knowledge graph optimization process includes:

[0207] The geological disaster knowledge graph is converted into a geological disaster multiple linked list; wherein the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength; refer to Figure 8 , Figure 8 Content is for Figure 7 The multiple linked lists of geological disasters converted from the geological disaster knowledge graph;

[0208] Furthermore, the correlation between the corrected disaster auxiliary description information and the data in the geological disaster multiple linked list is calculated; the specific data correlation calculation process includes:

[0209] Processing the corrected disaster auxiliary description information, including removing irrelevant words, unifying terminology, etc.;

[0210] Performing data standardization on the processed disaster correction auxiliary description information to obtain standard disaster correction auxiliary description information;

[0211] Vectorize the standard corrected disaster auxiliary description information and extract relevant features;

[0212] Construct a feature matrix using the acquired features;

[0213] The chi-square test is used to conduct correlation analysis between the characteristics of the characteristic matrix and the data of the geological disaster multiple linked list; the steps are:

[0214] Step 1: According to the relationship between the features of the feature matrix and the nodes of the multi-link list, select appropriate features for analysis;

[0215] Step 2: Construct a contingency table based on the frequency between features and nodes;

[0216] See Table 7 for an example of a contingency table of embodiments of the present application;

[0217] Table 7 Contingency table examples

[0218] landslide mudslide earthquake total Sand 10 5 2 17 Clay 8 12 4 24 Loam 6 3 1 10 total 24 20 7 51

[0219] Table 7 shows the relationship between different soil types and different geohazards; the rows represent different soil types (sand, clay, and loam). The columns represent different geohazard types (landslides, debris flows, and earthquakes). The numbers in Table 7 represent the number of specific geohazards observed under a specific soil type. The "Total" columns and rows at the far right and bottom represent the totals for each soil type and each hazard, respectively.

[0220] Step 3: Perform a chi-square test to evaluate the independence between the features and the multiple linked lists, and calculate the association value to determine the significance of the association.

[0221] Further, extracting auxiliary information elements of the corrected disaster auxiliary description information according to the correlation calculation result; including:

[0222] (1) Identify characteristics that are significantly associated with the type of geological hazards, such as soil type, rainfall, and topographic features;

[0223] (2) Assign weights to different features based on the strength of the association and identify the most influential features;

[0224] (3) Extract information elements and classify information extraction: Soil type: Extract the physical and chemical properties of various types of soil, such as permeability, viscosity, etc. Geological structure: Collect information on relevant geological structures, such as the existence and distribution of faults and folds.

[0225] Numerical information extraction: Meteorological data: Extract historical meteorological data related to geological disasters, such as precipitation, temperature, wind speed, etc. in a specific period.

[0226] Further, the auxiliary information elements are integrated with the linked list nodes and the linked list multiple edges of the geological disaster multiple linked list, and the connection strength is updated and the connection direction is modified; including:

[0227] (1) Identify target nodes: Based on the extracted auxiliary information elements, determine the linked list nodes that need to be integrated. For example, nodes that may need to be updated include "landslide" and "soil type".

[0228] (2) Update multiple edge information:

[0229] Enhance edge information: Add auxiliary information elements to existing multiple edges, such as incorporating rainfall, historical disaster data, etc. into the information attributes of related edges.

[0230] New Edges: If there is a new relationship between an auxiliary information feature and an existing node, you can create multiple edges. For example, add an edge from the "Rainfall" node to the "Landslide" node.

[0231] (3) Update connection strength:

[0232] Calculate new connection strength: Use statistical methods (such as linear regression, logistic regression, etc.) to calculate the new connection strength between auxiliary information elements and geological hazards. For example, analyze the relationship between rainfall and landslide incidence.

[0233] (4) Modify the connection direction: Determine the direction of the edge based on the updated connection strength.

[0234] (5) Update nodes: Update the integrated auxiliary information elements to the corresponding nodes to ensure the integrity of the information.

[0235] Furthermore, the updated geological disaster multi-link list is reversely converted into an optimized geological disaster knowledge graph. Fig. 9 , Fig. 9 This is the optimized geological disaster knowledge graph.

[0236] In the embodiment of the present application, a geological disaster knowledge graph optimization method is proposed to optimize the knowledge graph generated by the knowledge graph construction unit; the method converts the geological disaster knowledge graph into a multiple linked list to clearly reflect the relationship between the nodes; and obtains auxiliary information elements by correcting the correlation between the disaster auxiliary description information and the multiple linked list nodes, thereby realizing the update of edges and nodes; the updated multiple linked list is reversely converted to optimize the geological disaster knowledge graph; the method not only further improves the accuracy of the geological disaster knowledge graph, but also enriches the content of the geological disaster knowledge graph.

[0237] Furthermore, the geological disaster knowledge graph is displayed by the visualization unit of the system;

[0238] Furthermore, the maintenance unit is used to update the geological disaster knowledge graph by acquiring new information.

[0239] In the embodiment of the present application, the method and system proposed in the present invention are combined to realize the construction of the geological disaster knowledge map of earthquakes, landslides and mud-rock flows; it mainly starts from the following points: first, multimodal information about geological disasters is obtained from multiple channels; in order to ensure the reliability of information, the present invention proposes an information correction method; the method includes: expert knowledge correction and machine knowledge correction; wherein, the expert knowledge correction establishes an expert knowledge base of geologists and corrects the original information through the information matching method; the machine knowledge correction performs statistical analysis on the collected information to find erroneous information and correct it; the information corrected by expert knowledge and machine knowledge is comprehensively corrected to ensure the reliability of information. The correctness of the knowledge graph is verified; secondly, a geological disaster knowledge graph generation model is proposed, which generates the geological disaster knowledge graph according to the corrected disaster description information and the initial geological disaster map; in this model, the dependency between information and disasters is mined through the relationship dependency mining layer to establish the corresponding knowledge graph; this is conducive to the construction of the geological disaster knowledge graph; finally, a geological disaster knowledge graph optimization method is proposed, which converts the geological disaster knowledge graph into a multi-linked list to clearly reflect the relationship between each node; and obtains auxiliary information elements by correcting the correlation between the auxiliary disaster description information and the multi-linked list nodes, so as to update the edges and nodes; according to the updated multi-linked list, it is converted into an optimized geological disaster knowledge graph. The above methods are combined to achieve the accurate construction of the geological disaster knowledge graph.

[0240] Embodiment 2:

[0241] In the first implementation, the method and system of the present invention realize the construction of the geological disaster knowledge map of earthquake, landslide and debris flow; in the application embodiment, the geological disaster type will be added to illustrate that it has good flexibility and can realize the expansion of the geological disaster map; the specific process includes:

[0242] Collect multimodal information related to geological disasters and perform deduplication processing to obtain deduplicated multimodal information;

[0243] Further, the deduplicated multimodal information is classified according to the disaster type to obtain disaster description information and disaster auxiliary description information;

[0244] Further, the disaster description information and the disaster auxiliary description information are corrected to obtain corrected disaster description information and corrected disaster auxiliary description information;

[0245] Further, an initial geological hazard map is constructed;

[0246] Furthermore, the corrected disaster description information and the initial geological disaster map are input into a geological disaster knowledge graph generation model; the geological disaster knowledge graph generation model includes:

[0247] The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map;

[0248] A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features;

[0249] The relationship dependency mining layer is used to mine the dependency relationship of the structured geological hazard description features, including: establishing a structured geological hazard description feature matrix according to the structured geological hazard description features; wherein the structured geological hazard description feature matrix is ​​expressed as: Among them, GHTC i The hazard type characteristics represented by the i-th structural geological hazard description characteristics; GHATC i The geological hazard regional type characteristics represented by the i-th structured geological hazard description characteristics; It is represented as the jth related information feature vector of the i-th structured geological hazard description feature; M is represented as the total number of the structured geological hazard description features; N is represented as the total number of related information feature vectors in the structured geological hazard description features;

[0250] Input the structured geological hazard description feature matrix into the multi-conditional constrained random forest model to obtain the structured geological hazard dependency matrix;

[0251] Among them, the multi-conditional constraint random forest model includes: an input layer, used to receive the structured geological disaster description feature matrix; a tree generation layer, used to randomly generate multiple trees; wherein each tree independently selects feature data corresponding to different data disaster types; a condition constraint setting layer, used to set restrictions for each tree to constrain the branches of the tree; wherein the constraint rules of the condition constraint setting layer are set to: the occurrence reliability constraint between the disaster type and the geological disaster area type; the correlation constraint between the disaster type and the factors affecting the occurrence of the disaster; the impact constraint between the geological disaster area type and the disaster impact; a voting layer, used to summarize the analysis results of each tree; a dependency calculation layer, used to calculate the dependency based on the analysis results of the voting layer; and the output layer, used to output the structured geological disaster dependency matrix.

[0252] The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph;

[0253] A visualization output layer, used for visualizing the geological disaster knowledge graph;

[0254] Further, the geological disaster knowledge graph is optimized using the disaster correction auxiliary description information to obtain an optimized geological disaster knowledge graph;

[0255] The optimization process of the geological disaster knowledge graph includes:

[0256] The geological disaster knowledge graph is converted into a geological disaster multiple linked list; the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength;

[0257] Further, the correlation between the corrected disaster auxiliary description information and the data in the geological disaster multiple linked list is calculated;

[0258] Further, extracting auxiliary information elements of the corrected disaster auxiliary description information according to the correlation calculation result;

[0259] Further, the auxiliary information elements are integrated with the linked list nodes and the linked list multiple edges of the geological disaster multiple linked list, and the connection strength is updated and the connection direction is modified;

[0260] Furthermore, the updated geological disaster multiple linked list is reversely converted into the optimized geological disaster knowledge graph.

[0261] Embodiment three:

[0262] A storage medium for constructing a geological hazard knowledge graph, wherein a program for constructing a geological hazard knowledge graph is stored on the storage medium. When the program for constructing a geological hazard knowledge graph is executed by a processor, all the contents of Embodiment 1 or Embodiment 2 are implemented.

[0263] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a geological disaster knowledge graph, characterized in that: include: Collect multimodal information of relevant geological disasters and perform deduplication processing to obtain deduplicated multimodal information; Classifying the deduplicated multimodal information according to disaster types to obtain disaster description information and disaster auxiliary description information; Correcting the disaster description information and the disaster auxiliary description information to obtain corrected disaster description information and corrected disaster auxiliary description information; Construct an initial geological hazard map; Inputting the corrected disaster description information and the initial geological disaster map into a geological disaster knowledge graph generation model; The geological disaster knowledge graph generation model includes: The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map; A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features; The relationship dependency mining layer is used to mine the dependency relationship of the structured geological disaster description features, including: establishing a structured geological disaster description feature matrix according to the structured geological disaster description features; inputting the structured geological disaster description feature matrix into a multi-conditional constrained random forest model to obtain a structured geological disaster dependency matrix; The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph; A visualization output layer, used for visualizing the geological disaster knowledge graph; The geological disaster knowledge graph is optimized using the disaster correction auxiliary description information to obtain an optimized geological disaster knowledge graph.

2. The method for constructing a geological disaster knowledge graph according to claim 1, characterized in that: The classification process of the deduplicated multimodal information includes: obtaining descriptive information corresponding to different types of geological disasters; extracting keywords from the descriptive information to obtain disaster type keywords; vectorizing the disaster type keywords to obtain disaster type matching word vectors; dividing the deduplicated multimodal information into text data sets, video data sets and audio data sets according to information types; extracting all text descriptions from the text data set to obtain text description information; extracting all video descriptions from the video data set to obtain video description information; extracting all audio descriptions from the audio data set to obtain audio description information; respectively extracting the text description information, the The video description information and the audio data are subjected to feature extraction to obtain text description features, video description features and audio description features; the similarities between the text description features, the video description features and the audio description features and the disaster type matching word vector are respectively calculated to obtain text similarity, video similarity and audio similarity; when the text similarity, the video similarity and the audio similarity exceed a similarity matching threshold; the deduplicated multimodal information is used as the disaster description information corresponding to the disaster type; when the text similarity, the video similarity and the audio similarity do not exceed a similarity matching threshold, the deduplicated multimodal information is used as the disaster auxiliary description information.

3. The method for constructing a geological disaster knowledge graph according to claim 1, characterized in that: Correcting the disaster description information and the disaster auxiliary description information includes: an expert knowledge correction mechanism and a machine knowledge correction mechanism; The expert knowledge correction mechanism corrects information through the knowledge of geologists, including: obtaining relevant knowledge of geological disasters of N geologists to construct an expert knowledge base; generating expert knowledge matching text from the information in the expert knowledge base; establishing a multi-layer knowledge matching index for the expert knowledge matching text; matching the disaster description information and the disaster auxiliary description information with the expert knowledge matching text by using the multi-layer knowledge matching index; calculating the information consistency according to the matching result to obtain the information consistency value; when the information consistency value does not exceed the consistency threshold, using the information correction algorithm to obtain a candidate expert correction information set; The machine knowledge correction mechanism corrects information through a machine statistical algorithm, including: using a named entity recognition algorithm to identify the entity fields of the disaster description information and the disaster auxiliary description information; performing data statistics on the entity fields, and calculating the co-occurrence probability of the entity fields to obtain an entity field co-occurrence matrix; screening the disaster description information and the disaster auxiliary description information according to the entity field co-occurrence matrix to obtain a geological disaster screening information set; cross-validating the geological disaster screening information set; when the cross-validation value of any information in the geological disaster screening information set is lower than the cross-validation threshold, correcting the information that does not meet the conditions according to similar information to obtain a candidate machine-corrected information set; The candidate expert correction information set is merged with the candidate machine correction information set to obtain a comprehensive correction information set; when the candidate expert correction information set and the candidate machine correction information set have the same correction items; if the correction information of the correction items are the same, they are merged; if the correction information of the correction items are different, they are merged after regularization; and the disaster description information and the disaster auxiliary description information are corrected using the comprehensive correction information set.

4. The method for constructing a geological disaster knowledge graph according to claim 3, characterized in that: The multi-conditional constrained random forest model includes: An input layer, for receiving the structured geological hazard description feature matrix; The tree generation layer is used to randomly generate multiple trees, where each tree independently selects feature data corresponding to different data disaster types; The conditional constraint setting layer is used to set restriction conditions for each tree to constrain the branches of the tree; wherein the constraint rules of the conditional constraint setting layer are set as follows: the occurrence reliability constraint between the disaster type and the geological disaster area type; the correlation constraint between the disaster type and the factors affecting the occurrence of the disaster; the impact constraint between the geological disaster area type and the disaster impact; The voting layer is used to summarize the analysis results of each tree; A dependency calculation layer, used to calculate the dependency according to the analysis results of the voting layer; The output layer is used to output the structured geological hazard dependency matrix.

5. The method for constructing a geological disaster knowledge graph according to claim 1, characterized in that: The optimization process of the geological disaster knowledge graph includes: The geological disaster knowledge graph is converted into a geological disaster multiple linked list; the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength; Calculate the correlation between the disaster correction auxiliary description information and the data in the geological disaster multiple linked list; Extracting auxiliary information elements of the disaster auxiliary explanation information for correction according to the correlation calculation result; Integrate the auxiliary information elements with the linked list nodes and the linked list multiple edges of the geological disaster multiple linked list, update the connection strength, and modify the connection direction; The updated geological disaster multi-link list is reversely converted into the optimized geological disaster knowledge graph.

6. A system for constructing a geological disaster knowledge graph, characterized in that: The system includes: an information collection unit for collecting information related to geological disasters from different channels; an information classification unit for classifying geological disaster information; an information correction unit for correcting the collected relevant information; a knowledge graph construction unit for constructing a geological disaster knowledge graph; a knowledge graph optimization unit for optimizing the geological disaster knowledge graph; a visualization unit for visually displaying the generated geological disaster knowledge graph; and a maintenance unit for regularly updating the geological disaster knowledge graph.

7. A geological disaster knowledge graph construction system according to claim 6, characterized in that: The information correction unit includes: an expert knowledge correction module and a machine knowledge correction module; Wherein, the expert knowledge correction module includes: Obtain relevant knowledge of geological hazards from N geologists and build an expert knowledge base; Generate expert knowledge matching text from the information in the expert knowledge base; Establishing a multi-layer knowledge matching index for the expert knowledge matching text; Matching the disaster description information and the disaster auxiliary description information with the expert knowledge matching text according to the multi-layer knowledge matching index; Calculate the information consistency according to the matching results to obtain an information consistency value; when the information consistency value does not exceed the consistency threshold, use an information correction algorithm to obtain a candidate expert corrected information set; The machine knowledge correction module includes: Using a named entity recognition algorithm to identify entity fields of the disaster description information and the disaster auxiliary description information; Performing data statistics on the entity fields, and calculating the co-occurrence probability of the entity fields, to obtain an entity field co-occurrence matrix; The disaster description information and the disaster auxiliary description information are screened according to the entity field co-occurrence matrix to obtain a geological disaster screening information set; the geological disaster screening information set is cross-validated; when the cross-validation value of any information in the geological disaster screening information set is lower than the cross-validation threshold, the information that does not meet the conditions is corrected according to the same type of information to obtain a candidate machine correction information set; The candidate expert correction information set is merged with the candidate machine correction information set to obtain a comprehensive correction information set; when the candidate expert correction information set and the candidate machine correction information set have the same correction items; if the correction information of the correction items are the same, they are merged; if the correction information of the correction items are different, they are merged after regularization; and the disaster description information and the disaster auxiliary description information are corrected using the comprehensive correction information set.

8. A geological disaster knowledge graph construction system according to claim 6, characterized in that: The knowledge graph construction unit includes: Construct an initial geological hazard map; Inputting the corrected disaster description information and the initial geological disaster map into a geological disaster knowledge graph generation model; wherein the geological disaster knowledge graph generation model includes: The receiving layer includes: a disaster information receiving module for receiving the corrected disaster description information; a map receiving module for receiving the initial geological disaster map; A processing layer, used for processing the corrected disaster description information to obtain structured geological disaster description features; The relationship dependency mining layer is used to mine the dependency relationship of the structured geological disaster description features, including: establishing a structured geological disaster description feature matrix according to the structured geological disaster description features; inputting the structured geological disaster description feature matrix into a multi-conditional constrained random forest model to obtain a structured geological disaster dependency matrix; The geological disaster knowledge graph generation layer is used to generate the geological disaster knowledge graph; The visualization output layer is used to visualize the geological disaster knowledge graph.

9. A geological disaster knowledge graph construction system according to claim 6, characterized in that: The knowledge graph optimization unit includes: The geological disaster knowledge graph is converted into a geological disaster multiple linked list; the geological disaster multiple linked list includes: linked list nodes, linked list multiple edges and linked list connection directions; wherein the linked list nodes are nodes in the geological disaster knowledge graph; the linked list multiple edges are connection relationships between nodes in the geological disaster knowledge graph; the linked list multiple edges record various information elements related to geological disasters; the linked list connection direction is the connection strength between nodes in the geological disaster knowledge graph, and the connection direction of the linked list multiple edges is set according to the connection strength; Calculate the correlation between the auxiliary explanation information for correcting disasters and the data in the geological disaster multiple linked list; extract the auxiliary information elements of the auxiliary explanation information for correcting disasters according to the correlation calculation result; integrate the auxiliary information elements with the linked list nodes and the multiple edges of the linked list of the geological disaster multiple linked list, and update the connection strength and modify the connection direction; reversely transform the updated geological disaster multiple linked list into an optimized geological disaster knowledge graph.

10. A storage medium for constructing a geological disaster knowledge graph, characterized in that: The storage medium stores a program for constructing a geological hazard knowledge graph, and when the program for constructing a geological hazard knowledge graph is executed by a processor, a method for constructing a geological hazard knowledge graph as described in any one of claims 1 to 5 is implemented.

Images