Intelligent decision-making method and device for prevention and control of insect-borne infectious diseases

Abstract

The invention provides an insect-borne infectious disease prevention and control intelligent decision-making method and device, and the method comprises the steps: obtaining insect-borne infectious disease data from a plurality of data sources, and constructing a knowledge graph based on the insect-borne infectious disease data; user input and a pre-integrated insect-borne infectious disease prevention and control decision-making system are obtained, and the insect-borne infectious disease prevention and control decision-making system inquires information from the knowledge graph according to the user input and generates a decision-making result related to insect-borne infectious disease prevention and control according to the inquired information; and outputting the decision result in a streaming output mode. Due to the flexibility of the knowledge graph and the intellectualization of the insect-borne infectious disease prevention and control decision system, the method can cope with the complexity and variability of the insect-borne infectious disease transmission mode. A user is allowed to adjust prevention and control measures in time according to the latest decision result, the requirement for prevention and control of the insect-borne infectious diseases is met, and therefore propagation of the insect-borne infectious diseases is more flexibly coped with.

Description

Technical Field

[0001] The present invention relates to the field of public health technology, and in particular to an intelligent decision-making method and device for the prevention and control of insect-borne infectious diseases. Background Art

[0002] Vector-borne infectious diseases (such as dengue fever, malaria, and Zika virus disease) pose a serious threat to human health around the world, especially in tropical and subtropical regions. Such diseases are transmitted by mosquitoes and other vectors, with a wide range of infections and rapid transmission speed, and are highly prevalent and destructive. Traditional means of prevention and control of vector-borne infectious diseases mainly rely on the experience of experts and historical data, and usually adopt chemical prevention and control (such as spraying insecticides), environmental governance, and data-driven prevention and control. Chemical prevention and control is to kill mosquitoes by spraying insecticides, but it is easy for mosquitoes to develop drug resistance. Environmental governance is to remove mosquito breeding grounds such as stagnant water, but it requires a lot of manpower and material resources, and the effect is difficult to last. Data-driven prevention and control is to use big data and artificial intelligence technology to predict and prevent infectious diseases, and provide more accurate prevention and control measures. However, traditional data-driven prevention and control methods lack systematic and intelligent analysis tools, and cannot effectively respond to the complex and changeable transmission patterns and prevention and control needs of vector-borne infectious diseases. Summary of the invention

[0003] The present invention provides an intelligent decision-making method and device for the prevention and control of vector-borne infectious diseases, which is used to solve the defect that the existing technology cannot effectively respond to the complex and changeable transmission mode and prevention and control needs of vector-borne infectious diseases, and can respond to the complexity and variability of the transmission mode of vector-borne infectious diseases, allowing users to adjust prevention and control measures in time according to the latest decision results, meet the needs of vector-borne infectious disease prevention and control, and thus respond to the spread of vector-borne infectious diseases more flexibly. The technical solutions proposed by the present invention are as follows: In a first aspect, the present invention provides an intelligent decision-making method for the prevention and control of insect-borne infectious diseases, comprising: Acquire insect-borne infectious disease data from multiple data sources, and construct a knowledge graph based on the insect-borne infectious disease data; Obtaining user input and a pre-integrated insect-borne infectious disease prevention and control decision system, wherein the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to insect-borne infectious disease prevention and control according to the queried information; The decision result is output in a streaming output manner.

[0004] Optionally, constructing a knowledge graph based on the insect-borne infectious disease data includes: Generate a corresponding node for each sub-data in the vector-borne infectious disease data, and define a relationship type according to the relationship between the sub-data to obtain the knowledge graph; wherein the relationship types include transmission, geographical distribution and prevention and control measures, transmission represents the transmission relationship between vectors and diseases, geographical distribution represents the distribution of vectors and diseases among regions, and prevention and control measures represent the association between prevention and control strategies and vectors or diseases.

[0005] Optionally, the insect-borne infectious disease prevention and control decision system includes a retrieval tool, an embedding model, a vector storage system and a large language model; the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to the insect-borne infectious disease prevention and control according to the queried information, including: The search tool receives user input and converts the user input into an input vector using the loaded embedding model; The retrieval tool retrieves a target vector that is most similar to the input vector in the vector storage system, and retrieves mosquito information in the knowledge graph according to the target vector; wherein the target vector corresponds to an existing node in the knowledge graph; The mosquito information is input into the large language model to generate decision results related to the prevention and control of insect-borne infectious diseases.

[0006] Optionally, the insect-borne infectious disease prevention and control decision-making system further includes a query tool; and the method further includes: The query tool is used to search for relevant documents in the database according to the target vector, and key information is extracted according to the title, abstract or keywords of the document.

[0007] Optionally, the method further comprises: An interface is established on the back-end server, wherein the interface is used to receive user input; The back-end server transmits the user input to the proxy executor so that the proxy executor uses the insect-borne infectious disease prevention and control decision-making system to process the user input.

[0008] Optionally, the method further comprises: defining a streaming inference function of the large language model, wherein the streaming inference function is used to receive mosquito information and gradually generate decision results; A questioning route is defined in the Flask framework, where the questioning route is used to receive the mosquito information and call the streaming inference function; The step of inputting the mosquito information into the large language model to generate a decision result related to the prevention and control of insect-borne infectious diseases includes: After receiving the mosquito information, the question routing passes it to the streaming reasoning function for step-by-step reasoning to generate a decision result related to the prevention and control of insect-borne infectious diseases.

[0009] In a second aspect, the present invention also provides an intelligent decision-making device for the prevention and control of insect-borne infectious diseases, comprising the following modules: A graph construction module, used to obtain insect-borne infectious disease data from multiple data sources and construct a knowledge graph based on the insect-borne infectious disease data; a result generation module, for obtaining user input and a pre-integrated insect-borne infectious disease prevention and control decision system, wherein the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to insect-borne infectious disease prevention and control according to the queried information; The stream output module is used to output the decision result in a stream output manner.

[0010] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the intelligent decision-making method for the prevention and control of insect-borne infectious diseases as described in the first aspect above is implemented.

[0011] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent decision-making method for the prevention and control of insect-borne infectious diseases as described in the first aspect above.

[0012] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the intelligent decision-making method for the prevention and control of insect-borne infectious diseases as described in the first aspect above.

[0013] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows: The intelligent decision-making method and device for the prevention and control of vector-borne infectious diseases provided by the present invention obtain comprehensive vector-borne infectious disease data from multiple data sources, which covers multiple dimensions such as vector species, distribution, transmission characteristics, disease types, treatment methods, and prevention and control strategies. Then, the knowledge graph is constructed using this knowledge, which can clearly show the correlation and hierarchy between the data. The knowledge graph not only contains the data itself, but also reveals the intrinsic connection between the data, providing a solid foundation for subsequent intelligent analysis. A pre-integrated vector-borne infectious disease prevention and control decision-making system is introduced, which can quickly retrieve relevant information from the knowledge graph according to the specific needs of the user. Due to the flexibility of the knowledge graph and the intelligence of the vector-borne infectious disease prevention and control decision-making system, the method can cope with the complexity and variability of the transmission mode of vector-borne infectious diseases. Whether it is the emergence of new vector species, changes in transmission paths, or adjustments to prevention and control strategies, the method can maintain its effectiveness by updating the knowledge graph and adjusting the algorithm of the vector-borne infectious disease prevention and control decision-making system. Unlike traditional batch output results, this method adopts a streaming output method, that is, as the user's query deepens and the data is continuously updated, the decision results will be presented to the user in real time. This output method not only improves the timeliness of information, but also allows users to adjust prevention and control measures in a timely manner based on the latest decision-making results to meet the needs of insect-borne infectious disease prevention and control, thereby responding to the spread of insect-borne infectious diseases more flexibly.

[0014] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0015] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 This is one of the flow charts of the intelligent decision-making method for the prevention and control of insect-borne infectious diseases provided by the present invention.

[0018] Figure 2 This is the second flow chart of the intelligent decision-making method for the prevention and control of insect-borne infectious diseases provided by the present invention.

[0019] Figure 3It is a structural schematic diagram of the intelligent decision-making device for the prevention and control of insect-borne infectious diseases provided by the present invention.

[0020] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0022] Combine the following Figure 1-Figure 3 The intelligent decision-making method and device for insect-borne infectious disease prevention and control of the present invention are described. First, the terms involved in the present invention are explained: Vector-borne Disease: An infectious disease transmitted by vectors such as mosquitoes, such as dengue fever, malaria, and Zika virus disease.

[0023] Large Language Model (LLM): A natural language processing model trained using deep learning techniques that can understand and generate natural language text.

[0024] Knowledge Graph: A technology for structured representation of knowledge, which uses nodes and edges to represent concepts and their relationships.

[0025] Internationalization (i18n): The system supports multiple languages ​​and regions.

[0026] Reference Figure 1 As shown, the intelligent decision-making method for the prevention and control of insect-borne infectious diseases includes the following: Step S110: Acquire insect-borne infectious disease data from multiple data sources, and construct a knowledge graph based on the insect-borne infectious disease data.

[0027] First, data on vector-borne infectious diseases will be obtained from multiple data sources (such as the Centers for Disease Control and Prevention, professional books, API interfaces, web crawlers, public health departments, research institutions, medical institutions, etc.) to ensure the comprehensiveness and timeliness of the data. Specifically, epidemiological data and prevention and control strategy information related to vector-borne infectious diseases can be obtained from the Centers for Disease Control and Prevention. Relevant information such as the biological characteristics and transmission routes of vectors can be extracted from professional books. Real-time data can be obtained using crawler technology or API interfaces. The above-mentioned data on vector-borne infectious diseases include the types, scientific names, characteristics of vectors, disease types and their treatment methods, distribution areas and regional characteristics, prevention and control strategies and precautions, as well as the transmission relationships between vectors and disease types, the distribution of vectors and disease types among regions, and the associations between prevention and control strategies and vectors or disease types.

[0028] Before constructing the knowledge graph, it is necessary to perform data cleaning and structuring on the collected data on vector-borne infectious diseases. Since the data on vector-borne infectious diseases come from multiple channels, such as public health departments, research institutions, medical institutions, literature databases, web crawlers, etc. The data may exist in various forms such as text, tables, images, videos, etc. Through API interfaces, data crawlers, manual input, etc., data from different sources are collected into a unified database. Check whether there are duplicate, invalid or irrelevant information in the dataset and delete it. Duplicate records are identified and deleted by comparing the unique identifiers of the data (such as ID, title, etc.). Invalid data such as null values, garbled characters, format errors, etc. need to be cleaned or corrected. Irrelevant information such as advertisements, copyright information, etc., content unrelated to vector-borne infectious diseases needs to be removed. Unify data formats, units, naming rules, etc. to ensure the consistency and comparability of the data. For example: convert dates uniformly to the standard date format (such as YYYY-MM-DD). Unify the units of numerical values (such as convert temperature to Celsius), and remove unnecessary commas, dots, etc. Standardize the naming of place names, vector names, disease names, etc. Verify the cleaned data to ensure its accuracy and integrity. Check whether there are logical contradictions or inconsistencies between the data.

[0029] Split text data into individual words or phrases for subsequent analysis and processing. Use Chinese word segmentation tools (such as jieba) to segment Chinese text. For English text, you can use spaces or punctuation marks for simple word segmentation. Identify named entities such as place names, insect vector names, and disease names from the text. Use pre-trained named entity recognition models for recognition. Standardize the identified named entities (such as unifying place names into Chinese full names). Extract relationship information such as the transmission relationship between insect vectors and diseases, and the distribution of insect vectors and diseases in different regions from the text. Use rule-based methods or deep learning models for relationship extraction. Verify and correct the extracted relationships to ensure their accuracy. Generate concise and clear text descriptions or summaries based on the extracted structured data. Use text generation models (such as GPT, BERT, etc.) for text generation. Polish and correct the generated text to ensure its fluency and accuracy.

[0030] Based on the characteristics of vector-borne infectious disease data, define related entities (such as vectors, diseases, regions, etc.) and their attributes (such as scientific names, characteristics, treatment methods, etc.). Define the relationships between entities (such as transmission relationships, distribution relationships, etc.), and determine the direction and type of relationships. Based on the design of entities, attributes, and relationships, build appropriate data structures (such as database tables, graph structures, etc.) to store and manage data.

[0031] Import the cleaned structured data into the knowledge graph construction tool or platform. Create corresponding entities and relationships in the knowledge graph according to the design of the data model. Optimize the created knowledge graph, such as removing redundant entities and relationships, merging similar entities, etc., to improve the quality and readability of the graph. Use the knowledge graph query language (such as SPARQL) or graphical interface for query and display, so that users can intuitively understand the relevant information of insect-borne infectious diseases and the relationship between them.

[0032] The above-mentioned knowledge graph is a data representation method with a graph structure. It uses nodes (representing entities, such as vectors, diseases, etc.) and edges (representing the relationship between entities, such as transmission relationships, geographical distribution relationships, prevention and control measures, etc.) to organize information. When constructing a knowledge graph, a node is created for each data item, and edges are created based on the relationship between data items, thus forming a complex and orderly information network. A knowledge graph containing multi-dimensional information such as vectors, diseases, regions, and prevention and control strategies has been constructed, realizing a systematic representation of knowledge on prevention and control of vector-borne infectious diseases.

[0033] Step S120: obtaining user input and a pre-integrated vector-borne infectious disease prevention and control decision system, wherein the vector-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to the vector-borne infectious disease prevention and control according to the queried information.

[0034] Users can enter their queries or needs through a user interface (such as a web page, mobile application, etc.). These queries may involve specific vectors, infectious diseases, or the need for prevention and control measures. The pre-integrated vector-borne infectious disease prevention and control decision-making system will query the knowledge graph for relevant information based on the user's input. This vector-borne infectious disease prevention and control decision-making system can use advanced natural language processing technology, machine learning algorithms, and database query technology to ensure that data related to user input can be found accurately and quickly.

[0035] Step S130: output the decision result in a streaming output manner.

[0036] The insect-borne infectious disease prevention and control decision system will generate decision results related to insect-borne infectious disease prevention and control based on the queried information. These results may include disease types, treatment methods, prevention and control strategies, etc. The decision results will be presented to users in a streaming output format. This means that the results will be generated and displayed gradually over time, rather than all at once. This approach ensures that users can see the latest and most relevant information in real time, while also reducing the load on the system.

[0037] The intelligent decision-making method for the prevention and control of vector-borne infectious diseases provided by the present invention first obtains comprehensive vector-borne infectious disease data from multiple data sources, which covers multiple dimensions such as the types, distribution, and transmission characteristics of vectors, as well as the epidemic trends of infectious diseases and the affected areas. Then, the knowledge is used to construct a knowledge graph, which is a structured data representation method that can clearly show the correlation and hierarchy between data. The knowledge graph not only contains the data itself, but also reveals the intrinsic connection between the data, providing a solid foundation for subsequent intelligent analysis. A pre-integrated vector-borne infectious disease prevention and control decision-making system is introduced, which can quickly retrieve relevant information from the knowledge graph according to the specific needs of the user. The vector-borne infectious disease prevention and control decision-making system adopts advanced algorithms and models, which can analyze complex data relationships, explore potential risk points and prevention and control strategies, and thus generate scientific and reasonable decision results. Unlike traditional batch output results, this method adopts a streaming output method, that is, as the user's query deepens and the data is continuously updated, the decision results will be presented to the user in real time. This output method not only improves the timeliness of information, but also allows users to adjust prevention and control measures in time according to the latest decision results, so as to more flexibly respond to the spread of vector-borne infectious diseases. Due to the flexibility of the knowledge graph and the intelligence of the decision-making system for the prevention and control of vector-borne infectious diseases, this method can cope with the complexity and variability of the transmission patterns of vector-borne infectious diseases. Whether it is the emergence of new vector species, changes in transmission paths, or adjustments to prevention and control strategies, this method can maintain its effectiveness by updating the knowledge graph and adjusting the algorithm of the decision-making system for the prevention and control of vector-borne infectious diseases. The above-mentioned intelligent decision-making method for the prevention and control of vector-borne infectious diseases effectively overcomes the shortcomings of traditional data-driven prevention and control methods through the characteristics of systematic data integration, intelligent vector-borne infectious disease prevention and control decision-making system, streaming output results, and adaptation to complex and changing transmission patterns, providing more scientific, efficient, and flexible tools and means for the prevention and control of vector-borne infectious diseases.

[0038] Moreover, by constructing a knowledge graph and using an advanced insect-borne infectious disease prevention and control decision-making system, the present invention can quickly extract useful information from a large amount of data and generate decision results related to insect-borne infectious disease prevention and control. This greatly improves the efficiency of decision-making and enables a faster response to the epidemic. Since the knowledge graph can clearly represent the relationship between entities, the insect-borne infectious disease prevention and control decision-making system can more accurately understand the needs of user input and find the most relevant information in the knowledge graph. This helps to generate more accurate and reliable decision results. The streaming output method enables users to see the latest decision results in real time without waiting for all results to be generated. This improves the user experience and enables them to better understand the epidemic dynamics and prevention and control measures.

[0039] The existing technology lacks the integration of multi-source heterogeneous data and cannot build a comprehensive knowledge graph, which limits the correlation analysis and utilization of information. The present invention ensures the comprehensiveness and high quality of data through data collection from multiple data sources and automated knowledge processing, providing a solid foundation for intelligent analysis. Multi-channel data collection ensures the richness and integrity of information, covering the latest research and real-time dynamics. The automated knowledge processing process reduces manual intervention and reduces the error rate of data processing. The automated process greatly shortens the cycle from data collection to application, supporting rapid updates and iterations.

[0040] In an optional embodiment, the step S110 described above of constructing a knowledge graph based on the insect-borne infectious disease data includes: Generate a corresponding node for each sub-data in the vector-borne infectious disease data, and define a relationship type according to the relationship between the sub-data to obtain the knowledge graph; wherein the relationship types include transmission, geographical distribution and prevention and control measures, transmission represents the transmission relationship between vectors and diseases, geographical distribution represents the distribution of vectors and diseases among regions, and prevention and control measures represent the association between prevention and control strategies and vectors or diseases.

[0041] When constructing a knowledge graph based on vector-borne infectious disease data, first, the vector-borne infectious disease data collected from multiple data sources (such as public health databases, scientific research institution reports, medical institution records, etc.) are cleaned and sorted to remove duplicate, erroneous or irrelevant information, ensure the accuracy and consistency of the data, and obtain preprocessed data. Traverse the preprocessed data and generate a corresponding node for each sub-data (such as vectors, diseases, regions, prevention and control strategies, etc.). Each node contains specific attributes and identifiers for unique identification and retrieval in subsequent steps. According to the relationship between sub-data, three main types of relationships are defined: transmission, geographical distribution, and prevention and control measures. Transmission represents the transmission relationship between vectors and diseases. For example, if a certain mosquito is a vector of dengue fever, the transmission relationship between the mosquito and dengue fever is established in the knowledge graph. Geographical distribution represents the distribution of vectors and diseases among regions. For example, in which regions are a certain vector distributed and which regions are high-incidence areas of a certain infectious disease, this information is represented in the knowledge graph in the form of geographical distribution relationships. Prevention and control measures represent the association between prevention and control strategies and vectors or diseases. For example, a certain vaccine is an effective means of preventing a certain infectious disease, or a certain environmental modification measure can reduce the breeding of insect vectors. This information is represented in the knowledge graph in the form of prevention and control measures relationships. Using these relationship types, connections are established between nodes to form a complex and orderly knowledge graph. The constructed knowledge graph is optimized to ensure the accuracy of the nodes and the integrity of the relationships. The accuracy and reliability of the knowledge graph are ensured by comparison and verification with actual data.

[0042] In order to construct a knowledge graph, we first need to define the types of nodes and relationships. The main node types include: vectors, diseases, regions, and prevention and control strategies. Among them, vectors represent the medium that spreads the disease, such as mosquitoes. Diseases represent diseases spread by vectors, such as dengue fever and malaria. Regions represent the distribution areas of vectors and diseases. Prevention and control strategies represent prevention and control measures for vectors and diseases. The main relationship types include: transmission, geographical distribution, and prevention and control measures. Among them, transmission represents the transmission relationship between vectors and diseases, geographical distribution represents the distribution of vectors and diseases among regions, and prevention and control measures represent the relationship between prevention and control strategies and vectors or diseases.

[0043] The present invention uses Cypher statements to construct a knowledge graph: Example 1: Create an insect vector node: CREATE (:insect{name: 'Aedes albopictus', scientific name: 'Aedes albopictus', characteristics: 'black and white with white markings', distribution: 'subtropical and temperate regions'}); Example 2: Create a disease node: CREATE (:disease{name: 'dengue fever', symptoms: 'high fever, headache, muscle and joint pain, rash', treatment: 'symptomatic supportive treatment'}); Example 3: Create a regional node: CREATE (:region{name: 'Yunnan Province', country: 'China', features: 'tropical monsoon climate, suitable for mosquito growth'}); Example 4: Create a prevention and control strategy node: CREATE (:control strategy{name: 'Spraying insecticides', method: 'Use chemicals to kill adult mosquitoes and larvae', precautions: 'Avoid mosquitoes from developing insecticide resistance'}); Example 5: Establishing the relationship between insect vectors and disease transmission: MATCH (a: insect vector {name: 'Aedes albopictus'}), (b: disease {name: 'dengue fever'}) CREATE (a)-[:spread{mode: 'bite'}]->(b); Example 6: Establishing the geographical distribution relationship of insect vectors: MATCH (a: insect vector {name: 'Aedes albopictus'}), (b: region {name: 'Yunnan Province'}) CREATE (a)-[:geographical distribution]->(b); Example 7: Establishing the link between control strategies and insect vectors: MATCH (a: prevention and control strategy {name: 'Spraying insecticides'}), (b: insect vector {name: 'Aedes albopictus'}) CREATE (a)-[:control object]->(b); Through the above Cypher statements, a knowledge graph containing multi-dimensional information such as vectors, diseases, regions and prevention and control strategies was constructed, realizing a systematic representation of knowledge on the prevention and control of vector-borne infectious diseases.

[0044] In the construction of knowledge graphs, other graph databases or knowledge management tools can be used, such as GraphQL, JanusGraph, etc. According to the data scale and performance requirements, select a suitable graph database and adjust the data model and query statements. Different graph databases can realize structured storage and query of knowledge, ensuring the normal knowledge management function of the system.

[0045] The knowledge graph constructed by the present invention represents complex information in a structured manner in the form of nodes and relationships, making information retrieval more efficient and accurate. Users can quickly find data and information related to vector-borne infectious diseases to provide support for decision-making. The knowledge graph reveals the intrinsic connections and relationships between data, allowing users to better understand the relationship between the transmission pattern, geographical distribution and prevention and control measures of vector-borne infectious diseases. This helps users discover potential risk points and prevention and control strategies, and improve the pertinence and effectiveness of prevention and control measures. The knowledge graph provides a basis for intelligent analysis. Through the above-mentioned vector-borne infectious disease prevention and control decision-making system, the data in the knowledge graph can be deeply mined and analyzed to provide more scientific and reasonable decision-making support for the prevention and control of vector-borne infectious diseases.

[0046] In an optional embodiment, the insect-borne infectious disease prevention and control decision-making system includes a retrieval tool, an embedding model, a vector storage system and a large language model. The retrieval tool is responsible for receiving user input and converting it into a vector form for retrieval. The embedding model is used to convert text data (such as user input) into a vector representation for efficient retrieval in the vector storage system. The vector storage system is used to store vectorized text data and supports fast retrieval and matching. The large language model is used to receive the retrieved information and generate decision results related to the prevention and control of insect-borne infectious diseases.

[0047] First, load and initialize the model, including: using the Flask framework to build the backend service and provide an API interface. Load the Ollama language model and specify the base URL and model. The base URL refers to the network address where the model service interface is located or the root address of the service. Model refers to the model name or identifier, which refers to the name of the specific model to be loaded and run, such as the "Qianwen" language model. The code is as follows: Python from flask import Flask from flask_cors import CORS from langchain_community.llms import Ollama app = Flask(__name__) CORS(app) # Initialize the "Qianwen" large language model llm = Ollama(base_url='http: / / server address:port number', model='qwen:7b') Load the HuggingFaceEmbeddings embedding model (such as bge-m3) and use the FAISS vector storage system to load local vector data. The code is as follows: Python from langchain_huggingface import HuggingFaceEmbeddings embeddings = HuggingFaceEmbeddings(model_name='bge-m3') Use the FAISS vector storage system to store vectorized text data for quick retrieval. The code is as follows: Python from langchain_community.vectorstores import FAISS vectorstore = FAISS.load_local("vector storage system path", embeddings) retriever = vectorstore.as_retriever() Then, we create and integrate tools, including: creating the create_retriever_tool retrieval tool to process the Chinese name, English name and disease-related information of mosquitoes. The code is as follows: Load the PubMed query tool to retrieve relevant literature from the PubMed database and provide reference materials. The code is as follows: Python from langchain_community.tools.pubmed.tool import PubmedQueryRun tools = [vectortool, PubmedQueryRun()] Create a REACT agent create_react_agent and pass the loaded tools and large language model as parameters to the agent. The code is as follows: Python from langchain.agents import create_react_agent agent = create_react_agent(llm, tools) Create an executor and save conversation records: Use AgentExecutor to execute agent operations and support saving conversation records through ConversationBufferMemory. The code is as follows: Python from langchain.memory import ConversationBufferMemory from langchain.agents import AgentExecutor memory = ConversationBufferMemory(memory_key="chat_history") executor = AgentExecutor(agent=agent, tools=tools, memory=memory) The insect-borne infectious disease prevention and control decision system described in step S120 above queries information from the knowledge graph according to the user input, and generates a decision result related to the insect-borne infectious disease prevention and control according to the queried information, including: S1201: The search tool receives user input and converts the user input into an input vector using the loaded embedding model.

[0048] S1202. The retrieval tool retrieves a target vector that is most similar to the input vector in the vector storage system, and retrieves mosquito information in the knowledge graph based on the target vector; wherein the target vector corresponds to an existing node in the knowledge graph.

[0049] S1203: Input the mosquito information into the large language model to generate decision results related to the prevention and control of insect-borne infectious diseases.

[0050] Users submit query information through the front-end interface (such as web pages, mobile applications) or API interfaces. This information is usually related to topics such as a certain mosquito and its related prevention and control recommendations, transmission characteristics, and geographical distribution. After receiving the user input, the system first performs preprocessing, including removing irrelevant characters, word segmentation, and removing stop words, to improve the accuracy and efficiency of retrieval. The preprocessed query information is fed into the loaded embedding model (such as HuggingFace's Transformers model). The embedding model converts text information into high-dimensional vector representations that can capture the semantic features of the text. The converted input vectors are fed into a vector storage system (such as FAISS). The vector storage system quickly retrieves the target vectors that are most similar to the input vector. These target vectors correspond to nodes or entities that already exist in the knowledge graph. Based on the retrieved target vectors, the system locates the corresponding nodes in the knowledge graph. Starting from these nodes, the system further retrieves detailed information related to the mosquito, including its transmission characteristics, geographical distribution, disease types, treatment methods, prevention and control measures, etc. The system integrates the information retrieved from the knowledge graph into a structured dataset. This dataset contains detailed information about mosquitoes, as well as other information that may be relevant to prevention and control strategies, treatment methods, etc. The integrated information is fed into a large language model (such as Ollama). The large language model uses its powerful natural language processing and knowledge reasoning capabilities to generate decision results related to the prevention and control of insect-borne infectious diseases. These results may include specific prevention and control strategies, recommended treatments, preventive measures, etc. The system returns the generated decision results to the user. This can be returned directly to the caller through the API interface, or it can be presented to the user in a friendly manner through the front-end interface. The user can view the generated decision results and make further queries or modifications as needed. The system can also collect user feedback to optimize model training and improve the accuracy of the system.

[0051] Based on user feedback and system usage, the system can be optimized and updated regularly. This includes adjusting the parameters of the embedding model, updating the content of the knowledge graph, optimizing the retrieval algorithm of the vector storage system, etc. The system can continuously improve its accuracy and performance through continuous learning. This can be achieved by regularly acquiring information from new data sources, updating models and data sets, and introducing new algorithms and technologies.

[0052] The existing system cannot fully utilize artificial intelligence technologies such as large language models for in-depth analysis and prediction, resulting in insufficient scientificity and effectiveness of prevention and control measures. The present invention uses a vector storage system and an embedded model to achieve rapid matching of user input with nodes in the knowledge graph, thereby improving retrieval efficiency. The retrieved information is deeply analyzed and understood using a large language model to generate intelligent and personalized prevention and control recommendations. It integrates multi-dimensional information such as mosquito information, disease types, and prevention and control measures to provide users with comprehensive and accurate prevention and control recommendations. The system supports loading different embedded models, vector storage systems, and large language models, and can be flexibly configured and optimized according to actual needs. The system can regularly update the knowledge graph and vector storage system from new data sources to maintain the timeliness and accuracy of the information.

[0053] The present invention constructs a multi-dimensional knowledge graph including insect vectors, disease types, epidemiology, and prevention and control strategies, realizes the systematic and structured representation of information, and facilitates the correlation analysis and utilization of information. The knowledge graph structures scattered information for easy retrieval and analysis. Through the advantages of the graph database, complex correlation queries can be efficiently executed. Help users discover hidden relationships and patterns, and support the in-depth development of scientific research and prevention and control work. Moreover, the streaming reasoning method is adopted to achieve real-time response and result output of the model, improve user experience and timeliness of decision-making, and meet the real-time requirements in practical applications. Using the streaming reasoning method, the model can output results step by step while generating them, reducing user waiting time. Real-time feedback makes users feel that the system is more intelligent and friendly, and improves interaction satisfaction. In emergency situations, such as when an epidemic breaks out, it is crucial for decision makers to obtain analysis results in a timely manner.

[0054] The intelligent decision-making method for the prevention and control of vector-borne infectious diseases provided by the present invention realizes intelligent analysis and real-time decision support for vector-borne infectious diseases by integrating the knowledge graph with the large language model and adopting streaming output technology. By integrating the large language model and the knowledge graph, the system can deeply understand and analyze the information related to vector-borne infectious diseases and realize intelligent decision support. Compared with the existing technology, the depth and breadth of the analysis are significantly improved. The large language model has powerful natural language understanding and generation capabilities, and can perform in-depth semantic analysis of information related to vector-borne infectious diseases. Combined with the structured knowledge of the knowledge graph, the system can discover complex associations, such as potential transmission paths, new prevention and control strategies, etc. Users can interact with the system through natural language to obtain accurate answers and suggestions.

[0055] In an optional embodiment, the insect-borne infectious disease prevention and control decision-making system further includes a query tool; the method further includes: The query tool is used to search for relevant documents in the database according to the target vector, and key information is extracted according to the title, abstract or keywords of the document.

[0056] The system uses specialized literature search tools (such as PubMed, Google Scholar, Web of Science, etc.) or self-built literature databases to search for literature based on relevant information of the target vector (such as mosquito species, control keywords, etc.). The query tool returns a list of documents that match the query conditions. These documents may include academic papers, research reports, control guidelines, etc. The system performs a preliminary screening of the retrieved documents and extracts key information from the title, abstract or keywords of the documents. This information may include the latest control strategies, experimental results, case studies, etc.

[0057] After adding the above query tool, the working process of the insect-borne infectious disease prevention and control decision system is as follows: the user submits query information about a certain mosquito and its prevention and control recommendations through the front-end interface or API interface. The system preprocesses the query information and converts the preprocessed information into a vector representation using an embedding model to obtain an input vector. The system uses the target vector that is most similar to the input vector using the search tool, and retrieves relevant mosquito information in the knowledge graph based on the target vector, such as propagation characteristics, geographical distribution, etc. The system uses a special query tool (such as PubMed, Google Scholar, Web of Science, etc.) or a self-built literature database to perform literature retrieval based on relevant information of the target vector (such as mosquito species, prevention and control keywords, etc.). The query tool returns a list of documents that match the query conditions. These documents may include academic papers, research reports, prevention and control guidelines, etc. The system preliminarily screens the retrieved documents and extracts key information from the title, abstract or keywords of the documents. This information may include the latest prevention and control strategies, experimental results, case studies, etc. The system integrates the mosquito information retrieved from the knowledge graph with the key information extracted from the documents. The integrated information forms a comprehensive and detailed data set, providing data support for subsequent decision-making. Before inputting the data into the large language model, the system can also perform further preprocessing on the data, such as removing duplicate information and organizing the data format. The integrated data set is input into the large language model. The large language model uses its powerful natural language processing and knowledge reasoning capabilities to analyze, understand and reason the input data. Based on the input data set, the large language model generates decision results related to the prevention and control of insect-borne infectious diseases. These results can include specific prevention and control strategies, recommended treatments, preventive measures, the latest scientific research results, etc. The system returns the generated decision results to the user and supports display through the API interface or front-end interface. Users can make further queries or modifications based on the results and provide feedback to optimize the system.

[0058] The way to integrate the mosquito information retrieved from the knowledge graph with the key information extracted from the literature can be to combine the prevention and control strategies in the literature with the geographical distribution information in the knowledge graph, or to compare and analyze the latest scientific research results with existing prevention and control measures. In this way, the system can generate more accurate and practical decision-making results and provide better services to users.

[0059] By integrating mosquito information in the knowledge graph and key information in the literature, the system can provide more comprehensive and accurate decision support. The system can regularly retrieve the latest literature and research results and integrate them into the decision-making process, thereby supporting the continuous learning and updating of the system. After introducing the literature retrieval and key information extraction functions, the system can handle different types of query requests more flexibly and adapt to the ever-changing prevention and control needs.

[0060] In order to achieve real-time response of large language models, the present invention provides two methods for implementing streaming output, which are suitable for different application scenarios. One is to call agent_executor.invoke in the ask route to implement streaming output, and the other is to implement streaming output by defining a streaming inference function. The technical implementation process, advantages and applications of these two methods are described in detail below.

[0061] In an optional embodiment, the insect-borne infectious disease prevention and control decision system sets up an interface on the back-end server for receiving user input. The back-end server passes the user input to the proxy executor, and the proxy executor uses the insect-borne infectious disease prevention and control decision system (including embedded models, knowledge graphs, literature retrieval tools, and large language models, etc.) to process the user input and generate corresponding decision results.

[0062] The method further comprises: S210. Establish an interface on the backend server, wherein the interface is used to receive user input.

[0063] The process of defining the backend server interface is: Use the Flask framework to build the backend server and define an / ask route to handle user question and answer requests. Flask is a lightweight web application framework that is suitable for rapid development of small to medium-sized web applications.

[0064] The process of setting up the agent executor is as follows: The agent executor (agent_executor) is a middle-layer component that receives user input from the backend server and calls various tool chains in the insect-borne infectious disease prevention and control decision-making system (such as embedding models, vector storage systems, query tools, retrieval tools, and large language models, etc.) to process requests. The agent executor can be a custom Python class that contains the invoke method to execute the entire processing flow.

[0065] In the / ask route of the Flask backend server, use request.json or request.form (depending on the format of the user input) to receive the user input (query). Call the agent_executor.invoke(query) method to pass the user input to the agent executor.

[0066] S220. The back-end server transmits the user input to the proxy executor, so that the proxy executor uses the insect-borne infectious disease prevention and control decision-making system to process the user input.

[0067] The following steps are performed inside the agent executor: a. Use the embedding model to convert user input into a vector.

[0068] b. Retrieve the target vector that is most similar to the input vector in the knowledge graph and obtain relevant mosquito information.

[0069] c. Use search tools to extract key information from relevant literature. Integrate mosquito information with key information to obtain integrated information.

[0070] d. Input the integrated information into the large language model to generate decision results.

[0071] The proxy executor returns the generated decision result to the backend server. The backend server sends the decision result returned by the proxy executor directly to the client to achieve synchronous streaming output. This can be achieved through Flask's jsonify function or Response object to ensure that the response result is sent to the client in JSON format. The code is as follows: The present invention realizes the separation of the front-end and the back-end by setting up an interface on the back-end server. The front-end interface can focus on user interaction and display, while the back-end server is responsible for processing complex business logic and data interaction. As an intermediate layer component, the agent executor can easily expand and modify the processing flow. For example, a new tool chain can be added, the performance of the existing tool chain can be optimized, or it can be replaced with a more advanced model. By calling agent_executor.invoke in the / ask route and realizing synchronous streaming output, it can be ensured that the client can receive the processing results in real time, thereby improving the user experience. The client only needs to interact with the back-end server through HTTP requests, without having to care about the specific details of the back-end processing. This simplifies the logic of the client and reduces development costs.

[0072] In an optional embodiment, in the insect-borne infectious disease prevention and control decision-making system, in order to more efficiently process a large amount of data and provide real-time decision support, the present invention introduces the streaming reasoning function of the large language model. This function allows the system to gradually receive and process mosquito information, and at the same time generate decision results related to insect-borne infectious disease prevention and control in real time. By defining a streaming reasoning function and using the routing mechanism of the Flask framework, real-time data push and streaming output are achieved.

[0073] The method further comprises: S310. Define a stream inference function stream_inference of the large language model, where the stream inference function is used to receive mosquito information and gradually generate decision results.

[0074] Write a streaming inference function stream_inference that uses the large language model's stream method (or a similar API) to gradually generate the model's output. Inside the function, use Python's yield keyword to return the model's output in chunks. In this way, each time the model generates a portion of the output, the function pauses and returns that portion of the output, and then waits for the next call to continue generating the remaining output. The function receives mosquito information as input, which can include mosquito species, quantity, geographical distribution, propagation characteristics, etc.

[0075] S320. Define a question route ( / ask route) in the Flask framework, where the question route is used to receive the mosquito information and call a streaming inference function.

[0076] The step of inputting the mosquito information into the large language model to generate a decision result related to the prevention and control of insect-borne infectious diseases includes: After receiving the mosquito information, the question routing passes it to the streaming reasoning function for step-by-step reasoning to generate a decision result related to the prevention and control of insect-borne infectious diseases.

[0077] The / ask route handler receives the mosquito information submitted by the user through an HTTP POST request (usually sent in JSON format). The route handler passes the received mosquito information to the stream_inference function and obtains a generator object. This generator object will gradually generate decision results related to the prevention and control of insect-borne infectious diseases. Use Flask's Response object and use the generator object returned by the stream_inference function as the response body. Set the MIME type to application / json to ensure that the client can correctly parse the received data. The Response object processes the generator object through its internal mechanism (such as using the iter_encoded method of Werkzeug's BaseResponse class) to achieve real-time data push. Whenever the generator object produces a new decision result block, the block is immediately encoded in JSON format and sent to the client through an HTTP response. The client (such as a web front-end or a mobile application) sends mosquito information to the server through an HTTP request and listens to the streaming response returned by the server. Each decision result block received by the client is in JSON format and can be directly parsed into a JavaScript object or other corresponding data structure. The client can update the user interface in real time based on the received decision results, display them to the user, or take corresponding actions (such as sending alarms, launching prevention and control measures, etc.). In this way, the insect-borne infectious disease prevention and control decision-making system can generate decision results based on mosquito information in real time and send these results to the client in real time, thereby improving the system's response speed and decision-making efficiency.

[0078] The streaming reasoning function designed in the present invention allows the system to process mosquito information in real time and generate decision results related to the prevention and control of vector-borne infectious diseases. This improves the response speed and decision-making efficiency of the system. By returning the output content of the model in blocks, the system can use memory and computing resources more effectively. This helps to process large-scale data sets and complex model reasoning tasks. The function of pushing data in real time enables the client to see the processing results immediately without waiting for the entire processing process to complete. This improves user experience and satisfaction. The design of the streaming reasoning function enables the system to be easily expanded to support more types of mosquito information and more complex decision logic. This will help the system adapt to changing prevention and control needs in the future.

[0079] It should be noted that if the real-time requirement is not high, batch processing can be used instead of the above streaming output for data processing and model reasoning. Specifically, the system architecture can be adjusted to change streaming processing to scheduled batch processing to reduce the complexity and resource consumption of the system. Although the real-time performance is reduced, the core functions and analytical capabilities of the system are not affected.

[0080] Reference Figure 2 As shown, the above-mentioned knowledge graph includes insect vector graph, disease graph, epidemiological graph and prevention and control strategy graph. For the insect vector graph, the front end of the present invention adopts Vue3 framework to realize a responsive and dynamic user interface. The specific process is as follows: Initialize a new Vue3 project using tools such as Vue CLI or Vite. Design different Vue components such as navigation bar, content display area, sidebar, etc. according to functional requirements. Use Vue's two-way data binding feature to bind user input and interface elements together for real-time updates. Use Vue's responsive layout and media query techniques to ensure that the interface can be displayed well on different devices and screen sizes.

[0081] Responsive design enables the interface to maintain a good user experience on different devices. Vue3's component-based development improves code reusability and development efficiency. Two-way data binding ensures real-time synchronization between user input and interface elements.

[0082] For the epidemiological graph, the present invention uses Echarts to draw a tree diagram to show the hierarchical structure of the knowledge graph. The process is as follows: Introduce the Echarts library in the Vue project. Obtain the node and link data of the knowledge graph from the backend and convert them into the format required by Echarts. Configure the option object of Echarts according to requirements, including series type, layout mode, data, label display, etc. Pass the configured option object to the Echarts instance and call the setOption method to render the chart. The code is as follows: The present invention can intuitively display the hierarchy and relationship of the knowledge graph through a tree diagram. Echarts provides rich interactive functions, such as zooming, dragging, etc., which enhances the user experience. Complex data is displayed in a graphical way, which is convenient for users to understand and analyze.

[0083] For the disease atlas, the present invention also uses the Neo4j D3 visualization tool to realize knowledge retrieval and dynamic display. The specific process is: configure the connection information of the Neo4j database, and write query statements to realize the connection of the Neo4j database. Introduce the D3.js library in the Vue project. Send a query request to the Neo4j database through HTTP request and obtain the returned data. Use D3.js to process the returned data and draw the corresponding visualization graphics. The code is as follows: The present invention realizes the retrieval function of knowledge in Neo4j database. D3.js provides powerful data visualization capability and can dynamically display query results. The visualization graphics based on data can be automatically updated as the data is updated.

[0084] For the prevention and control strategy map, the present invention also integrates carousel technology to dynamically display key information in the knowledge graph. The specific process is as follows: Select a suitable carousel component library, such as Swiper, Slick, etc. Obtain key information in the knowledge graph from the back end and convert it into the format required by the carousel component. Configure the parameters of the carousel component according to requirements, such as carousel speed, animation effects, etc. Embed the configured carousel component into the Vue component and render it. The present invention can highlight the key information in the knowledge graph through carousel technology. The dynamically changing carousel effect can attract the user's attention. The carousel content can be automatically updated as the back-end data is updated.

[0085] The present invention also introduces the i18n internationalization library into the Vue project. Corresponding language files are prepared for languages ​​such as Chinese, English, Lao and Khmer. Functions such as language switching buttons or drop-down menus are implemented in the Vue component. The text content on the interface is dynamically replaced according to the currently selected language. The needs of users of different languages ​​are met, and the internationalization level of the system is improved. Users can choose the appropriate language version according to their language habits, which improves user satisfaction.

[0086] By adopting Vue3 framework, Echarts, Neo4j D3 visualization tools, carousel technology and internationalization support, a knowledge graph display system with rich functions and good user experience has been built. The system can intuitively display the hierarchy and relationship of the knowledge graph, realize the retrieval and dynamic display of knowledge, and support multiple language versions to meet the needs of different users.

[0087] The dynamic and interactive visualization method enhances the user's interactive experience. At the same time, it supports multi-language versions, expands the application scope of the system, and has good international promotion potential. Dynamic visualization makes complex data and relationships clear at a glance, lowering the threshold for understanding. Users can interact with the visualization interface to get a richer experience. International support allows the system to serve users with different languages ​​and cultural backgrounds, and has the potential for global promotion.

[0088] The intelligent decision-making device for the prevention and control of vector-borne infectious diseases provided by the present invention is described below. The intelligent decision-making device for the prevention and control of vector-borne infectious diseases described below and the intelligent decision-making method for the prevention and control of vector-borne infectious diseases described above can be referenced to each other.

[0089] The intelligent decision-making device for insect-borne infectious disease prevention and control provided by the present invention refers to Figure 3 As shown, including: A graph construction module 410, configured to obtain insect-borne infectious disease data from multiple data sources and construct a knowledge graph based on the insect-borne infectious disease data; A result generation module 420, for obtaining user input and a pre-integrated insect-borne infectious disease prevention and control decision system, wherein the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to insect-borne infectious disease prevention and control according to the queried information; The streaming output module 430 is used to output the decision result in a streaming output manner.

[0090] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the intelligent decision-making method for the prevention and control of insect-borne infectious diseases.

[0091] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0092] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the intelligent decision-making methods for the prevention and control of vector-borne infectious diseases provided by the above-mentioned methods.

[0093] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the intelligent decision-making method for the prevention and control of insect-borne infectious diseases provided by the above-mentioned methods.

[0094] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent decision-making method for the prevention and control of insect-borne infectious diseases, characterized in that: include: Acquire insect-borne infectious disease data from multiple data sources, and construct a knowledge graph based on the insect-borne infectious disease data; Obtaining user input and a pre-integrated insect-borne infectious disease prevention and control decision system, wherein the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to insect-borne infectious disease prevention and control according to the queried information; The decision result is output in a streaming output manner.

2. The intelligent decision-making method for the prevention and control of insect-borne infectious diseases according to claim 1, characterized in that: The constructing of a knowledge graph based on the insect-borne infectious disease data includes: Generate a corresponding node for each sub-data in the vector-borne infectious disease data, and define a relationship type according to the relationship between the sub-data to obtain the knowledge graph; wherein the relationship types include transmission, geographical distribution and prevention and control measures, transmission represents the transmission relationship between vectors and diseases, geographical distribution represents the distribution of vectors and diseases among regions, and prevention and control measures represent the association between prevention and control strategies and vectors or diseases.

3. The intelligent decision-making method for the prevention and control of insect-borne infectious diseases according to claim 1, characterized in that: The insect-borne infectious disease prevention and control decision system includes a retrieval tool, an embedding model, a vector storage system and a large language model; the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to the insect-borne infectious disease prevention and control according to the queried information, including: The search tool receives user input and converts the user input into an input vector using the loaded embedding model; The retrieval tool retrieves a target vector that is most similar to the input vector in the vector storage system, and retrieves mosquito information in the knowledge graph according to the target vector; wherein the target vector corresponds to an existing node in the knowledge graph; The mosquito information is input into the large language model to generate decision results related to the prevention and control of insect-borne infectious diseases.

4. The intelligent decision-making method for the prevention and control of insect-borne infectious diseases according to claim 3, characterized in that: The insect-borne infectious disease prevention and control decision-making system also includes a query tool; the method also includes: The query tool is used to search for relevant documents in the database according to the target vector, and key information is extracted according to the title, abstract or keywords of the document.

5. The intelligent decision-making method for the prevention and control of insect-borne infectious diseases according to claim 1, characterized in that: The method further comprises: An interface is established on the back-end server, wherein the interface is used to receive user input; The back-end server transmits the user input to the proxy executor so that the proxy executor uses the insect-borne infectious disease prevention and control decision-making system to process the user input.

6. The intelligent decision-making method for the prevention and control of insect-borne infectious diseases according to claim 3, characterized in that: The method further comprises: defining a streaming inference function of the large language model, wherein the streaming inference function is used to receive mosquito information and gradually generate decision results; A questioning route is defined in the Flask framework, where the questioning route is used to receive the mosquito information and call the streaming inference function; The step of inputting the mosquito information into the large language model to generate a decision result related to the prevention and control of insect-borne infectious diseases includes: After receiving the mosquito information, the question routing passes it to the streaming reasoning function for step-by-step reasoning to generate a decision result related to the prevention and control of insect-borne infectious diseases.

7. An intelligent decision-making device for the prevention and control of insect-borne infectious diseases, characterized in that: include: A graph construction module, used to obtain insect-borne infectious disease data from multiple data sources and construct a knowledge graph based on the insect-borne infectious disease data; a result generation module, for obtaining user input and a pre-integrated insect-borne infectious disease prevention and control decision system, wherein the insect-borne infectious disease prevention and control decision system queries information from the knowledge graph according to the user input, and generates a decision result related to insect-borne infectious disease prevention and control according to the queried information; The stream output module is used to output the decision result in a stream output manner.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the intelligent decision-making method for the prevention and control of vector-borne infectious diseases as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the intelligent decision-making method for the prevention and control of insect-borne infectious diseases as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the intelligent decision-making method for the prevention and control of insect-borne infectious diseases as described in any one of claims 1 to 6 is implemented.

Images