Data query method and device, computer equipment and storage medium

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to a data query method and device, computer equipment and a storage medium, and the method comprises the steps: judging whether a query request sent by a user is received or not; wherein the query request carries query content; if yes, acquiring identity information of the user, and calling a target assistant instance corresponding to the identity information; based on the target assistant instance, retrieving resource data corresponding to the query content from an independent resource pool; searching knowledge data corresponding to the query content from a personal knowledge base; performing comprehensive reasoning processing on the resource data and the knowledge data based on a reasoning strategy to generate a corresponding reasoning result; generating target answer data based on the reasoning result; and returning the target answer data to the user. In addition, the invention also relates to a block chain technology, and the target answer data can be stored in a block chain. The method can be applied to data query scenes in the financial science and technology field and the digital medical field, and the accuracy and personalization of query processing are effectively improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology and can be applied to fields such as fintech and digital healthcare, particularly to data query methods, devices, computer equipment, and storage media. Background Technology

[0002] In current business system application scenarios, a common problem exists: systems struggle to provide in-depth, professional services for specific business domains. Existing systems often employ rigid, generic models, failing to effectively learn and adapt to the individual characteristics of different business roles, and even the analytical thinking, terminology, and analytical paths of different users within the same role. This leads to systems easily providing incorrect or confusing answers, resulting in low accuracy of generated response data and severely impacting the user experience.

[0003] In the financial sector, taking credit risk assessment as an example, traditional credit risk assessment systems often rely solely on limited financial indicators (such as income and liabilities) and basic credit history. However, customers in different industries and with different business models exhibit significantly different risk characteristics. For instance, for an emerging technology company, traditional methods may underestimate its development potential and repayment ability by neglecting key factors such as its R&D investment ratio, number of technology patents, and market prospects, leading to incorrect risk assessment results, affecting financial institutions' lending decisions, and even causing the loss of high-quality customers. In the medical field, taking drug efficacy prediction as an example, traditional drug efficacy prediction systems mainly rely on the general pharmacological effects of drugs and clinical trial data from some common populations for prediction. However, different patients have vastly different physical conditions, genetic characteristics, and underlying diseases, all of which significantly impact drug efficacy. For example, for patients with cancers suffering from specific gene mutations, traditional systems may only rely on drug efficacy data for common cancer types, ignoring the specific impact of the gene mutation on drug metabolism and mechanisms of action, thus providing inaccurate drug efficacy prediction results. This can lead doctors to develop inappropriate treatment plans, affecting the patient's treatment outcome and recovery process.

[0004] Therefore, there is an urgent need for a system that can be deeply adapted to specific business areas and accurately meet the personalized needs of different users in order to improve the professionalism and accuracy of services. Summary of the Invention

[0005] The purpose of this application is to provide a data query method, apparatus, computer device, and storage medium to solve the technical problem that existing service systems often adopt rigid, general patterns, which cannot effectively learn and adapt to different business roles, resulting in the system easily giving incorrect or confusing answers and generating low accuracy of the response data.

[0006] Firstly, a data query method is provided, including: Determine whether a query request from a user has been received; wherein the query request carries query content; If so, obtain the user's identity information and call the target assistant instance corresponding to the identity information; Based on the target assistant instance, resource data corresponding to the query content is retrieved from a preset independent resource pool; Retrieve knowledge data corresponding to the query content from a preset personal knowledge base; Based on a preset reasoning strategy, the resource data and the knowledge data are comprehensively reasoned and processed to generate corresponding reasoning results; Based on the reasoning results, corresponding target answer data is generated; The target answer data is returned to the user.

[0007] Secondly, a data query device is provided, comprising: The judgment module is used to determine whether a query request from a user has been received; wherein the query request carries query content; The processing module is used to, if so, obtain the user's identity information and call the target assistant instance corresponding to the identity information; The retrieval module is used to retrieve resource data corresponding to the query content from a preset independent resource pool based on the target assistant instance; The search module is used to retrieve knowledge data corresponding to the query content from a preset personal knowledge base; The reasoning module is used to perform comprehensive reasoning processing on the resource data and the knowledge data based on a preset reasoning strategy, and generate corresponding reasoning results. The generation module is used to generate corresponding target answer data based on the reasoning results; The return module is used to return the target answer data to the user.

[0008] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described data query method.

[0009] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described data query method.

[0010] In the above-described data query method, apparatus, computer equipment, and storage medium, the solution first determines whether a query request from a user has been received; wherein the query request carries query content; if so, the user's identity information is obtained, and a target assistant instance corresponding to the identity information is invoked; then, based on the target assistant instance, resource data corresponding to the query content is retrieved from a preset independent resource pool; subsequently, knowledge data corresponding to the query content is retrieved from a preset personal knowledge base; subsequently, the resource data and the knowledge data are comprehensively reasoned based on a preset reasoning strategy to generate a corresponding reasoning result; further, corresponding target answer data is generated based on the reasoning result; finally, the target answer data is returned to the user. Based on the above automated processing flow, after receiving a query request from a user, this application intelligently invokes a target assistant instance corresponding to the user's identity information, and then, based on the use of the target assistant instance, retrieves resource data and knowledge data corresponding to the query content from a preset independent resource pool and a personal knowledge base, and performs comprehensive reasoning on the resource data and knowledge data based on the use of a reasoning strategy to generate a reasoning result, and then generates target answer data based on the reasoning result, and returns the target answer data to the user. This enables the use of target assistant instances to accurately understand user needs and generate target answer data that meets user expectations, ensuring the accuracy and personalization of query processing and improving the accuracy of the generated target answer data. Attached Figure Description

[0011] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is an exemplary system architecture diagram to which this application can be applied; Figure 2 This is a flowchart of an embodiment of the data query method according to this application; Figure 3 This is a schematic diagram of the structure of one embodiment of the data query device according to this application; Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0014] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0016] like Figure 1 As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Terminal device 101 may be a laptop 1011, tablet 1012, or mobile phone 1013. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0017] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0018] Terminal device 101 can be various electronic devices with a display screen and support web browsing. In addition to laptops 1011, tablets 1012, or mobile phones 1013, terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, and a desktop computer, etc.

[0019] Server 103 can be a server that provides various services, such as a backend server that provides support for the interface displayed on terminal device 101.

[0020] It should be noted that the data query method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the data query device is generally set in the server / terminal device.

[0021] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0022] Continue to refer to Figure 2 The flowchart illustrates an embodiment of the data query method according to this application. Depending on different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted. The data query method provided in this application embodiment can be applied to any scenario requiring data querying, and therefore can be applied to products in these scenarios, such as data query products in the financial and insurance fields. The data query method includes the following steps: Step S201: Determine whether a query request from the user has been received; wherein the query request carries query content.

[0023] In this embodiment, the data query method runs on an electronic device (e.g., Figure 1The server / terminal device shown can receive user query requests via wired or wireless connection. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wideband) connections, and other currently known or future wireless connection methods. The executing entity of this application is a data query system based on an assistant instance, which can be simply referred to as the system. Users can describe their query needs in natural language through the input boxes provided on the system interface, and submit a query request containing the query content to the system by clicking the query button or pressing the Enter key. Upon receiving the query request, the system will invoke the target instance assistant matched with the user to initiate the subsequent retrieval and reasoning process, and return the matching answer data to the user, thereby completing the response processing for the query request.

[0024] This application can be applied to data query scenarios in the fintech and digital healthcare fields. For example, in a product comparison query scenario within the financial insurance sector, a user (such as customer Xiao Wang, who intends to purchase insurance) could enter the query: "I want to buy critical illness insurance for myself when I turn 30. Among the mainstream critical illness insurance products on the market, which one offers the best value for money, and what are its coverage and claim conditions?" Alternatively, in an insurance claim progress query scenario, a user (such as customer Xiao Li, who has submitted a claim) could enter the query: "I submitted a car insurance claim last week. What stage is the claim at now, and how long will it take to complete?"

[0025] In the digital healthcare field, a user (such as Xiao Sun, a patient diagnosed with diabetes) might enter the following query: "I was just diagnosed with type 2 diabetes. What are the common treatment options available? What are the advantages, disadvantages, and suitable populations for each treatment option?" Alternatively, in a health indicator interpretation scenario, a user (such as Xiao Zhou, an office worker concerned about his health) might enter the following query: "My recent physical examination report shows that my blood pressure is a bit high, with a systolic pressure of 145 mmHg and a diastolic pressure of 95 mmHg. What do these values ​​mean, and what should I pay attention to in order to improve them?"

[0026] In addition, the system has assistant creation and resource configuration functions. The specific implementation process includes: 1) Determining business needs and assistant types: Administrators first communicate in depth with various business departments to understand their specific needs in data analysis. For example, the finance department needs to process financial data, perform cost accounting, and analyze financial statements; the store operations department focuses on indicators such as sales data, inventory management, and customer traffic. Based on these needs, the type of assistant to be created is determined, such as a financial analysis assistant or a store operations assistant. 2) Selecting assistant templates: The assistant template library in the domain assistant factory provides pre-built assistant configuration templates for different business functions. Administrators select appropriate templates from the template library according to the determined assistant type. These templates contain the initial domain knowledge skeleton; for example, the finance template may contain basic financial indicators and report structures, while the store operations template may contain sales indicators and inventory management-related content. 3) Instantiating assistants: Administrators use the assistant instantiation engine (referred to as assistant instance) to quickly create and deploy a domain-specific assistant instance based on the selected template. During the instantiation process, administrators can make certain adjustments and customizations to the template according to the specific needs of the business departments, such as adding specific business rules or data fields. 4) Define Resource Scope: Allocate and manage an independent set of resources for each assistant instance. Indicator Library Configuration: Configure relevant business indicators for each assistant based on its business domain. For example, the indicator library for the financial analysis assistant includes financial indicators such as "gross profit margin," "accounts receivable turnover," "net profit margin," and "debt-to-equity ratio"; the indicator library for the store operations assistant includes operational indicators such as "sales per square meter," "average transaction value," "out-of-stock rate," and "sales ranking." These indicators are calibrated and defined to ensure accurate use during query processing. Data Model Encapsulation: Encapsulate data tables and relationships related to the domain. The financial analysis assistant may encapsulate general ledger data tables, cost center data tables, etc., and clarify the relationships between them, such as the correspondence between cost centers and general ledger accounts; the store operations assistant may encapsulate POS transaction data tables, store master data tables, etc., and the relationships between them, such as the relationship between stores and sales transactions. Data Source Connection Configuration: Configure the physical data sources or logical data views accessible to the assistant instance. The data source for the financial analysis assistant connects to the ERP system to obtain accurate financial data; the data source for the store operations assistant connects to the business middle platform to obtain real-time sales and operational data. 5) Resource isolation settings: Through namespaces or permission logic, ensure that resources of different assistant instances are not visible to each other and do not interfere with each other. For example, set up different namespaces for the financial analysis assistant and the store operation assistant, so that they can only access resources under their own namespaces, thus eliminating cross-contamination of data and standards from the root.

[0027] In this way, by creating dedicated assistant instances for different business departments and defining independent resource scopes, it is ensured that each assistant can focus on data and knowledge in a specific domain. The resource isolation mechanism guarantees data security and the purity of the business context, so that users in different departments are not affected by data from other departments when performing queries and processing, thereby obtaining accurate and professional analysis results.

[0028] Step S202: If yes, obtain the user's identity information and call the target assistant instance corresponding to the identity information.

[0029] In this embodiment, during the system design phase, association rules between users and assistant instances are preset based on the enterprise's organizational structure and business processes. For example, by department, each department corresponds to one or more assistants with specific functions; or by job level, users in different positions use assistants with different permissions and functions. When a user logs into the system by entering identity information (such as name, department, job title, contact information, etc.), the system automatically associates and invokes the user with the corresponding assistant instance according to the preset rules. For example, after a sales department employee successfully logs in, the system automatically associates them with a dedicated sales assistant, which has functions for sales operations such as sales data analysis and customer follow-up reminders.

[0030] The system includes features that allow users to personalize the matched assistant, specifically including: 1) Personal terminology dictionary maintenance: Add Terminology: Users can add frequently used business terms and their explanations or mappings to their personal terminology dictionary based on their work habits and business understanding. For example, a user can personalize the definition of "cash flow" as "the total daily inflows and outflows of the company's bank account." In subsequent queries, when "cash flow" is mentioned, the assistant can understand and process it according to Zhang's definition. Modify Terminology: If a user is not satisfied with the previously added term definitions or if the business situation changes, the term definitions can be modified. For example, if a user finds that the previous definition of "cash flow" is not accurate enough, it can be modified to "the total daily inflows and outflows of the company's bank account related to its main business." 2) Use of the Analysis Process Recorder: Recording Query and Analysis Paths: The system can record users' frequently used query and analysis paths or custom calculation formulas. For example, when conducting cost analysis, users often first query the department's expense details and then calculate the percentage of each expense; the system will record these steps. Save Analysis Templates: Users can save the recorded query and analysis paths as "My Analysis Templates." For example, Zhang saved the above cost analysis steps as a template named "Department Cost Percentage Analysis" for easy use later. 3) Knowledge Inheritance and Overriding Mechanism Settings: User-level knowledge takes effect on top of the domain-wide knowledge, allowing for personal customization, with higher priority given to personalized customization. For example, in the financial indicator library, "cost" has a general definition, but if a user has created a personal definition for "cost" in their personal knowledge base, the assistant will prioritize using the accountant's personal definition when the user initiates a query related to "cost".

[0031] Thus, the personalized configuration process allows each user to have customized settings that align with their work habits and business understanding when using the assistant. A personal terminology dictionary and analytical thought process recorder enable the assistant to better understand user needs and provide analytical results that are more relevant to the user's actual work. The knowledge inheritance and overriding mechanism balances the consistency of public knowledge with the personalization of individual knowledge, ensuring both the assistant's professionalism and accuracy within its domain and meeting users' personalized needs, thereby improving user experience and work efficiency.

[0032] Step S203: Based on the target assistant instance, retrieve resource data corresponding to the query content from a preset independent resource pool.

[0033] In this embodiment, the specific implementation process of retrieving resource data corresponding to the query content from a preset independent resource pool based on the target assistant instance will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0034] Step S204: Retrieve knowledge data corresponding to the query content from a preset personal knowledge base.

[0035] In this embodiment, the specific implementation process of searching for knowledge data corresponding to the query content from a preset personal knowledge base will be described in more detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0036] Step S205: Based on a preset reasoning strategy, perform comprehensive reasoning processing on the resource data and the knowledge data to generate corresponding reasoning results.

[0037] In this embodiment, the specific implementation process of performing comprehensive reasoning processing on the resource data and the knowledge data based on the preset reasoning strategy to generate the corresponding reasoning result will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0038] Step S206: Generate corresponding target answer data based on the reasoning result.

[0039] In this embodiment, the specific implementation process of generating the corresponding target answer data based on the reasoning result will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0040] Step S207: Return the target answer data to the user.

[0041] In this embodiment, the specific implementation process of returning the target answer data to the user will be described in more detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0042] This application first determines whether a query request from a user has been received; wherein the query request carries query content; if so, it obtains the user's identity information and calls the target assistant instance corresponding to the identity information; then, based on the target assistant instance, it retrieves resource data corresponding to the query content from a preset independent resource pool; subsequently, it searches for knowledge data corresponding to the query content from a preset personal knowledge base; subsequently, it performs comprehensive reasoning processing on the resource data and the knowledge data based on a preset reasoning strategy to generate a corresponding reasoning result; further, it generates corresponding target answer data based on the reasoning result; finally, it returns the target answer data to the user. Based on the above automated processing flow, after receiving a query request from a user, this application intelligently calls the target assistant instance corresponding to the user's identity information, and then, based on the use of the target assistant instance, retrieves resource data and knowledge data corresponding to the query content from a preset independent resource pool and a personal knowledge base, and performs comprehensive reasoning on the resource data and knowledge data based on the use of a reasoning strategy to generate a reasoning result, and then generates target answer data based on the reasoning result, and returns the target answer data to the user. This enables the use of target assistant instances to accurately understand user needs and generate target answer data that meets user expectations, ensuring the accuracy and personalization of query processing and improving the accuracy of the generated target answer data.

[0043] In some alternative implementations, step S203 includes the following steps: Based on the target assistant instance, the query content is semantically parsed to obtain the corresponding semantic parsing result.

[0044] In this embodiment, the target assistant instance (hereinafter referred to as the assistant) has a semantic parsing rule base, which contains semantic mappings of various business terms and common expressions. By matching the query content with the rule base, the assistant can understand the basic intent of the query and generate corresponding semantic parsing results. For example, when a user queries "What are the costs this month (such as insurance sales costs)?", the assistant will recognize that "this month" is a time range and "cost" is a core business concept.

[0045] Based on the semantic parsing results, the domain knowledge index in the independent resource pool is queried to obtain the corresponding target domain knowledge.

[0046] In this embodiment, the assistant queries the domain knowledge index in its independent resource pool based on the results of semantic parsing. The domain knowledge index is a structured knowledge catalog that categorizes and associates various concepts, indicators, and data models within the business domain. The assistant searches the domain knowledge index for entries related to core business concepts from the semantic parsing results, such as "cost," and finds associated indicator libraries, data models, and data source connection information. For example, it finds the cost center model, which defines the classification and accounting methods of costs; cost-related indicators in the financial statement indicator library, such as direct costs and indirect costs; and data source connection information pointing to the tables and fields storing cost data in the ERP system. Furthermore, the assistant further explores the relationships between identified related business domain knowledge. For example, it understands how the cost center model and cost indicators in the financial statement indicator library influence each other, and how these indicators correspond to data in the ERP system. Through this knowledge mining, the assistant can construct a complete knowledge network, providing more comprehensive support for subsequent retrieval.

[0047] Based on a preset data acquisition strategy, the target domain knowledge is processed to obtain the corresponding relevant data.

[0048] In this embodiment, the above-mentioned data acquisition and processing process for the target domain knowledge includes: 1) Indicator library query: The assistant queries the indicator library based on the determined relevant indicators (core business concepts in the semantic parsing results). The indicator library stores the definitions, calculation methods, and related explanations of various business indicators. For the "cost" indicator, the assistant obtains its definition from the financial indicator library, such as cost refers to the various expenses incurred by an enterprise in the production and operation process; it also obtains the calculation method, such as the cost calculation formula may be direct cost plus indirect cost. In addition, the indicator library may also contain information such as the scope of application and business significance of the indicator. 2) Data model parsing: The assistant parses the determined data model. The data model describes the storage structure and relationships of business data. Taking the cost center model as an example, the assistant will understand which tables are included in the model and what the relationships are between the tables, such as the cost center table and the cost details table being associated through the cost center number. By parsing the data model, the assistant can know how to obtain cost data from the data source and how to organize and analyze this data. 3) Data source connection and data acquisition: The assistant establishes a connection with the corresponding data source based on the data source connection information. The financial analysis assistant connects to the ERP system. Based on the previously analyzed data model and information from the indicator library, the assistant retrieves relevant cost data from the ERP system. For example, it queries the ERP system for direct and indirect cost data incurred this month, according to the time frame "this month" and the definition of cost indicators. During the data acquisition process, the assistant further handles data format conversion and data cleaning to ensure the accuracy and usability of the acquired data.

[0049] The relevant data is used as the resource data.

[0050] Based on the above processing flow, this application performs semantic parsing on the query content using a target assistant instance to obtain the corresponding semantic parsing results; then, based on the semantic parsing results, it queries and processes the domain knowledge index in the independent resource pool to obtain the corresponding target domain knowledge; subsequently, it processes the target domain knowledge for data acquisition based on a preset data acquisition strategy to obtain the corresponding related data; and finally, it uses the related data as resource data. Thus, by using a target assistant instance, this application can automatically and accurately perform a comprehensive search of the independent resource pool to obtain resource data corresponding to the query content, improving the efficiency of resource data retrieval and ensuring the accuracy and personalization of the resource data.

[0051] In some optional implementations of this embodiment, step S204 includes the following steps: Retrieve target personal terms related to the query content from the personal knowledge base.

[0052] In this embodiment, the aforementioned personal knowledge base refers to the user's personal knowledge base. The assistant searches for personal terminology definitions related to the query content in this personal knowledge base, i.e., target personal terms. The personal knowledge base contains a user terminology database, which stores user-defined business terms and their explanations or mapping relationships. The assistant matches key terms in the query content with entries in the user terminology database. For example, for the user query "How are the costs this month?", the assistant searches for a personal definition of "cost" in the user's personal terminology database. If the assistant finds a relevant personal terminology definition in the user terminology database, such as the user defining "cost" as "the sum of direct and indirect costs," the assistant will parse this definition. It will understand the scope and meaning of this definition and compare and integrate it with general definitions in the domain knowledge. For example, the assistant will know that in subsequent processing, when the concept of "cost" is involved, calculations and analyses should be performed according to the user's definition.

[0053] In addition, the assistant updates relevant related knowledge based on the user's personal terminology definition. For example, if the user's definition of "cost" differs from the general definition in the domain knowledge, the assistant will adjust the calculation of "cost"-related indicators and the use of data models to ensure that the user's personal definition can be correctly applied in subsequent reasoning and result generation.

[0054] Retrieve target analysis templates related to the query content from the personal knowledge base.

[0055] In this embodiment, the assistant similarly searches for relevant analysis templates in the user's personal knowledge base. The personal knowledge base contains a template library that stores various analysis ideas and operational steps saved by the user. The assistant searches the template library for potentially matching analysis templates based on the content and semantics of the query. For example, for a user query "What are the costs this month?", the assistant will search for templates related to departmental cost analysis. Then, the assistant will match and filter the found analysis templates. It compares the similarity of the query conditions, analysis indicators, and operational steps in the templates with the current query content. For example, the assistant will check whether the analysis template involves cost indicator analysis and whether it considers factors such as time range. If multiple analysis templates are similar to the current query, the assistant will select the most suitable target analysis template based on certain rules (such as template usage frequency, user preferences, etc.). If the assistant finds a suitable target analysis template, it will parameterize that template. For example, the target analysis template may contain some fixed parameters, such as department name and time range. The assistant will replace these parameters with actual values ​​based on the current query content. For user queries, the assistant will set the time range parameter in the template to "this month" so that subsequent data queries and analyses can be performed according to this parameter.

[0056] The target personal terms and the target analysis template are integrated to obtain the corresponding integrated data.

[0057] In this embodiment, the identified target personal terms and target analysis templates can be integrated, and the resulting integrated data can be used as corresponding knowledge data.

[0058] The integrated data is used as the knowledge data.

[0059] Based on the above processing flow, this application retrieves target personal terms related to the query content from the personal knowledge base; and retrieves target analysis templates related to the query content from the personal knowledge base; then, it integrates the target personal terms and target analysis templates to obtain corresponding integrated data; subsequently, the integrated data is used as knowledge data. Thus, by using a target assistant instance, this application can automatically and accurately perform a comprehensive search of the personal knowledge base to obtain knowledge data corresponding to the query content, improving the efficiency of knowledge data retrieval and ensuring the accuracy and personalization of the knowledge data.

[0060] In some alternative implementations, step S205 includes the following steps: The resource data and the knowledge data are fused together to obtain the corresponding fused data.

[0061] In this embodiment, domain knowledge (resource data) from an independent resource pool and personalized settings (knowledge data) from a personal knowledge base are merged using a target assistant instance to obtain fused data. For example, in the case of a financial analysis assistant, the assistant merges the general concept of "cost" from the financial indicator library with the range of "cost" defined by Accountant Zhang personally. It comprehensively considers the two definitions to determine how to use them in subsequent analysis. If the personal definition supplements or refines the general definition, the assistant will adjust according to the personal definition; if there is a conflict between the two, the assistant will resolve it according to certain rules (such as prioritizing the personal definition or making a compromise).

[0062] Furthermore, the fused data can be formatted according to the requirements of the analysis results. For example, for financial analysis reports, the Target Assistant instance will arrange cost data in a specific format, such as displaying various cost indicators (direct costs, indirect costs, total costs, etc.) in separate rows and annotating the corresponding values ​​next to them. At the same time, for data such as cost percentages, the assistant will convert them into percentage form and retain an appropriate number of decimal places to ensure the readability and accuracy of the data.

[0063] The fused data is subjected to causal relationship analysis to obtain the corresponding analysis results.

[0064] In this embodiment, a causal relationship analysis is performed on the fused data using a target assistant instance (hereinafter referred to as the assistant), and corresponding analysis results are generated. Specifically, the causal relationship analysis process includes: 1) Feature identification: The assistant first performs feature identification on the fused data. In a financial analysis scenario, it identifies various data features related to costs, such as raw material purchase prices, business volume indicators (such as production quantity, sales quantity), and indirect cost components (such as rent, utilities, employee salaries, etc.). 2) Association rule analysis: Association rule mining algorithms, such as the Apriori algorithm, are used to perform association analysis on the fused data. Taking raw material purchase prices and direct costs as an example, the assistant will statistically analyze the changes in direct costs under different raw material purchase price ranges. By setting support and confidence thresholds, frequently occurring data combinations with strong correlations are identified. For example, it is found that when the raw material purchase price is in a certain range (such as an increase of 5%-10% compared to the previous month), the direct costs increase frequently, and this increase is strongly correlated with the change in raw material purchase prices, thus initially determining that the increase in raw material prices is a possible cause of the increase in direct costs. Business Logic Verification: The uncovered relationships are verified using business domain knowledge. The assistant consults the business domain knowledge base to understand the impact mechanism of rising raw material prices on direct costs. For example, in manufacturing, raw materials are the main input for production, and their price increases inevitably lead to an increase in direct costs (such as raw material procurement costs). If the business logic matches the results of the data correlation analysis, this causal relationship is further confirmed. 2) Multi-Factor Comprehensive Analysis. Factor Decomposition: For cost changes, the assistant decomposes costs into direct and indirect costs, and then analyzes the factors affecting them separately. Taking indirect costs as an example, the assistant analyzes the impact of business expansion on indirect costs. By comparing indirect cost data at different business volume levels, it is found that as business volume increases, indirect cost items such as rent and utilities also increase. Weight Determination: Weight analysis is used to determine the degree of influence of each factor on cost changes. For example, when analyzing changes in direct costs, the contribution rates of rising raw material prices and increased business volume to changes in direct costs are calculated. Assuming the total increase in direct costs is 1 million yuan, of which 700,000 yuan is due to the increase in raw material prices and 300,000 yuan is due to the increase in business volume, then it can be determined that the increase in raw material prices is the main driver of the increase in direct costs.

[0065] The fused data is processed for trend prediction and evaluation to obtain the corresponding trend evaluation results.

[0066] In this embodiment, a target assistant instance is used to predict and evaluate trends based on historical and current fused data. For cost data, the assistant analyzes its long-term trend to determine if there is a sustained upward or downward trend. Simultaneously, the assistant assesses whether the current cost situation is within a reasonable range and aligns with the company's cost control objectives. For example, if the monthly cost growth rate is significantly higher than the historical average, the assistant generates a corresponding warning to remind the user to pay attention to cost control issues.

[0067] Specifically, the process of trend prediction and evaluation of fused data includes: 1) Trend prediction. Time series analysis: For cost data, the assistant uses time series analysis methods for trend prediction. Time series models can include moving average models, exponential smoothing models, and ARIMA models. Moving average model: Calculates the moving average of historical cost data to smooth short-term fluctuations in the data, thereby revealing long-term trends. For example, calculating the moving average of costs over the past 12 months, and observing the trend of the moving average, determines whether costs are rising, falling, or stable. Exponential smoothing model: Weighted averages are applied to historical cost data, giving greater weight to recent data. By adjusting the smoothing coefficient, the sensitivity of the model to historical data is controlled. Exponential smoothing models can better capture recent trends in data, thus more accurately predicting future cost trends. ARIMA model: The ARIMA (Autoregressive Integrated Moving Average) model is a more complex time series analysis method that combines autoregression, differencing, and moving average. The assistant determines the parameters (p, d, q) of the ARIMA model by performing stationarity tests, autocorrelation, and partial autocorrelation analyses on cost data, and then uses this model to predict future costs. Trend visualization: Historical cost data and prediction results are visualized, such as by drawing line charts. In the line chart, the horizontal axis represents time, and the vertical axis represents cost amount. By observing the trend of the line chart, the trend of cost changes can be intuitively judged. 2) Trend assessment. Comparative analysis: The predicted cost trend is compared with the company's cost control target. For example, the company's cost control target is to keep the cost growth rate within 5%, while the prediction result shows that the cost growth rate this month will reach 8%, which is significantly higher than the target value. Reasonableness judgment: Combining industry data and the company's historical data, the reasonableness of the current cost situation is assessed. The assistant will collect cost data from other companies in the same industry and analyze the industry average cost level. At the same time, the company's cost data over the past few years will be reviewed to understand the pattern of cost changes. If the current cost growth rate is much higher than the industry average and does not conform to the company's historical cost change trend, it is believed that there may be a problem with the current cost situation. Warning Generation: Based on the results of comparative analysis and rationality judgment, corresponding warning prompts are generated. If the cost growth rate is significantly higher than the historical average or the company's cost control target, the assistant will add a prominent warning message to the analysis results, such as "This month's cost growth rate is 3 percentage points higher than the historical average, exceeding the company's cost control target, and cost control issues need to be closely monitored."

[0068] The analysis results and the trend assessment results are integrated to obtain the corresponding integrated results.

[0069] In this embodiment, the obtained analysis results and trend assessment results can be integrated and processed, and the resulting integrated result can be used as the corresponding reasoning result.

[0070] The integration result is used as the reasoning result.

[0071] Based on the above processing flow, this application integrates resource data and knowledge data to obtain corresponding integrated data; then, it performs causal relationship analysis on the integrated data to obtain corresponding analysis results; and finally, it performs trend prediction and evaluation on the integrated data to obtain corresponding trend evaluation results. The analysis results and trend evaluation results are then integrated to obtain corresponding integrated results. These integrated results are subsequently used as inference results. Thus, this application, by integrating resource data and knowledge data to obtain integrated data, and then performing causal relationship analysis and trend prediction and evaluation on the integrated data, and then integrating the analysis results and trend evaluation results, uses the generated integrated results as corresponding inference results. This enables automatic, efficient, and accurate comprehensive inference processing of resource data and knowledge data, improving the efficiency of inference processing and ensuring the accuracy and comprehensiveness of the generated inference results.

[0072] In some alternative implementations, step S206 includes the following steps: Information is extracted from the reasoning results to obtain the corresponding key information.

[0073] In this embodiment, key information can be extracted from the causal relationship analysis and trend prediction assessment results included in the above reasoning results. In the causal relationship analysis section, the main causes of cost changes are extracted, such as rising raw material prices and business expansion; in the trend prediction assessment section, the cost growth rate, comparison with the target, and early warning information are extracted and integrated to obtain key information.

[0074] The key information is logically sorted and organized to obtain the corresponding initial answer data.

[0075] In this embodiment, the extracted key information is logically organized and arranged in a certain order to generate the corresponding initial response data. For example, the changes in costs are first described (e.g., total costs increased by 15% compared to the previous month), then the main reasons for the cost changes are explained (rising raw material prices led to an increase in direct costs, and business expansion led to an increase in indirect costs), and finally the assessment results and early warning information of the current cost situation are mentioned (the current cost situation is within a reasonable range, but the subsequent cost change trend needs to be monitored).

[0076] The initial response data is subjected to language optimization processing to obtain the corresponding first response data.

[0077] In this embodiment, the language optimization process includes: Conciseness: Using clear and concise language to express conclusions, avoiding overly complex or technical jargon. For example, simplifying "Total costs this month showed a 15% increase compared to last month" to "Total costs this month increased by 15% compared to last month." Highlighting Key Information: Emphasizing key information in the conclusions through bolding, color changing, etc. In financial analysis reports, key content such as cost growth rates, main reasons, and warning information is displayed in bold to attract user attention.

[0078] The first answer data is formatted and optimized to obtain the corresponding second answer data.

[0079] In this embodiment, the above format optimization process includes: Position optimization: Determining the presentation position of the conclusions based on the overall layout of the analysis report. Generally, the conclusions can be placed at the beginning of the report as a brief summary of the report's content, allowing users to understand the main analysis results at a glance; or they can be placed at the end of the report to summarize and elevate the entire analysis process. Layout optimization: Optimizing the layout of the conclusions to ensure they are clear and easy to read. For example, appropriately setting line spacing, font size, and color to distinguish the conclusions from other content in the report without making them appear abrupt.

[0080] The second answer data is used as the target answer data.

[0081] Based on the above processing flow, this application extracts information from the reasoning results to obtain corresponding key information; then, it logically organizes and processes this key information to obtain initial answer data; subsequently, it optimizes the language of the initial answer data to obtain the first answer data; and then optimizes the format of the first answer data to obtain the second answer data; finally, it uses the second answer data as the target answer data. In this way, this application improves the efficiency of answer data generation by extracting key information from the reasoning results and logically organizing and processing this key information to obtain initial answer data. Furthermore, it intelligently optimizes the language and format of the initial answer data, thereby effectively improving the accuracy and standardization of the generated target answer data.

[0082] In some optional implementations of this embodiment, step S207 includes the following steps: Obtain the preset target display method.

[0083] In this embodiment, the specific implementation process of obtaining the preset target display method will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0084] The target display method is optimized to obtain the optimized target display method.

[0085] In this embodiment, after determining the target display method, it will be further optimized, including: for table-style displays, adjusting the column width and row height to ensure clear data display; adding table borders and background colors to improve readability; and highlighting certain cells in the table based on data importance, such as using different colors or font styles. For displays combining images and text, the assistant will adjust the image size and position to ensure harmonious pairing between images and text; and add text descriptions explaining the characteristics and relevant information of the products in the images.

[0086] Call the preset target interface.

[0087] In this embodiment, the target interface is an interface constructed based on the requirements of the aforementioned target display method. Specifically, the interface design and layout of the target interface include: for tabular financial analysis reports, table areas, title areas, and operation button areas can be added to the interface. The table area is used to display data, the title area is used to display the report title and description, and the operation button area is used to provide sorting, filtering, downloading, and other operation functions. Furthermore, the layout of interface elements is adjusted and optimized to ensure that the interface is simple, beautiful, and easy to use. For example, the assistant will place the table area in the center of the interface to make it the focus of the user's attention; it will place the operation button area at the bottom or side of the interface for easy user operation; and it will reasonably set the spacing and margins between various areas to avoid the interface elements being too crowded or scattered.

[0088] Based on the optimized target display method, the target answer data is displayed and processed in the target interface.

[0089] In this embodiment, the generated target number of answers can be displayed on the target interface according to the planned interface layout. For example, for a financial analysis report, the table data will be filled into the table area, and the title and description will be displayed in the title area; for a product list, product images and text information will be arranged on the interface in a certain order.

[0090] Based on the above processing flow, this application obtains a preset target display method; then optimizes the target display method to obtain an optimized target display method; subsequently, it calls a preset target interface; and then, based on the optimized target display method, it displays the target answer data in the target interface. In this way, this application obtains an optimized target display method by optimizing the obtained target display method, and then displays the target answer data in the target interface based on the optimized target display method. This achieves the goal of presenting the target answer data to the user in a suitable way, meeting the user's query needs, enabling the user to obtain accurate and useful information in a timely manner, thereby making more scientific and reasonable decisions, and improving work efficiency and business level. At the same time, a suitable display method also enhances the user experience and strengthens the user's trust and reliance on the assistant.

[0091] In some optional implementations of this embodiment, obtaining the preset target display method includes the following steps: The target response data is evaluated for its dimensionality and complexity to obtain the corresponding evaluation results.

[0092] In this embodiment, the generated target answer data can be evaluated for data dimensions and complexity using a target assistant example (hereinafter referred to as the assistant) to generate corresponding evaluation results. For example, for a financial analysis report, the assistant will analyze the data dimensions contained in the report, such as the time dimension (this month, last month), cost item dimensions (direct costs, indirect costs, etc.), and other possible dimensions (such as departments, products, etc.). Simultaneously, the assistant will evaluate the complexity of the data, such as the relationships between data and the complexity of calculation formulas. If the report contains a large amount of data and complex calculation relationships, the assistant will tend to choose a more detailed and comprehensive display method.

[0093] Obtain the user's needs information.

[0094] In this embodiment, the assistant can analyze the user's needs and preferences regarding the way answers are displayed by combining the user's query history, usage habits, and feedback information. For example, if the user frequently uses tables to view data in previous queries and uses the sorting and filtering functions of tables frequently, the assistant will assume that the user prefers a table-based display. Furthermore, the assistant will also consider the user's business scenario and needs; for example, financial analysis reports may require more detailed text descriptions, while product lists may require more intuitive image displays.

[0095] Get the available display options.

[0096] In this embodiment, the above-mentioned available display methods may include at least table-based display, image display, text display, and a combination of image and text display, etc.

[0097] Based on the evaluation results and the requirement information, the corresponding designated display method is determined from the available display methods.

[0098] In this embodiment, based on the obtained evaluation results and demand information, the assistant selects the most suitable display method from the available options as the specified display method. Specifically, for financial analysis reports with a large amount of data and many data dimensions, the assistant will choose to display them in a tabular format, because tables can clearly present various data and facilitate users to perform sorting, filtering, and other operations. For product lists, the assistant will choose to display them using a combination of images and text, which can more intuitively show the appearance and basic information of the products.

[0099] The specified display method is used as the target display method.

[0100] Based on the above processing flow, this application evaluates the target answer data regarding data dimensions and complexity to obtain corresponding evaluation results; acquires user demand information; then obtains available display methods; subsequently, based on the evaluation results and demand information, determines the corresponding specified display method from the available display methods; and finally, uses the specified display method as the target display method. Thus, by combining the evaluation results and demand information, this application can automatically and intelligently select the most suitable display method from the available display methods as the required target display method, effectively improving the accuracy and intelligence of target display method selection.

[0101] In some alternative implementations, the user information obtained is subject to user consent and complies with relevant laws and policies.

[0102] Furthermore, any software tools or components not belonging to our company that appear in the embodiments of this application are merely illustrative examples and do not represent actual use.

[0103] Furthermore, this application has the following advantages: 1. It achieves depth and professionalism in analysis: Each assistant specializes in a specific field, with a highly focused knowledge base and resources, enabling them to provide deeper business insights rather than generalities.

[0104] 2. Ensures data security and clarity: The resource isolation mechanism ensures data access security and the purity of the business context. Users will only have access to data and metrics relevant to their work, allowing for a more focused experience.

[0105] 3. Improved user stickiness and efficiency: The user-level knowledge base enables the assistant to "remember" the user's habits, making it easier to use the more it is used, and truly becoming a personalized intelligent assistant for employees, rather than a cold tool.

[0106] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0107] It should be emphasized that, to further ensure the privacy and security of the aforementioned target response data, the target response data can also be stored in a blockchain node.

[0108] The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0109] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results. Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0110] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0111] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0112] Further reference Figure 3 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of a data query device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0113] like Figure 3 As shown, the data query device 300 described in this embodiment includes: a judgment module 301, a processing module 302, a retrieval module 303, a search module 304, a reasoning module 305, a generation module 306, and a return module 307. Wherein: The judgment module 301 is used to determine whether a query request sent by the user has been received; wherein the query request carries query content; The processing module 302 is used to, if yes, obtain the user's identity information and call the target assistant instance corresponding to the identity information; The retrieval module 303 is used to retrieve resource data corresponding to the query content from a preset independent resource pool based on the target assistant instance; The search module 304 is used to search for knowledge data corresponding to the query content from a preset personal knowledge base; The reasoning module 305 is used to perform comprehensive reasoning processing on the resource data and the knowledge data based on a preset reasoning strategy, and generate corresponding reasoning results. The generation module 306 is used to generate corresponding target answer data based on the reasoning result; The return module 307 is used to return the target answer data to the user.

[0114] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here.

[0115] In some optional implementations of this embodiment, the retrieval module 303 includes: The parsing submodule is used to perform semantic parsing on the query content based on the target assistant instance to obtain the corresponding semantic parsing result; The query submodule is used to query the domain knowledge index in the independent resource pool based on the semantic parsing results to obtain the corresponding target domain knowledge; The first acquisition submodule is used to perform data acquisition and processing on the target domain knowledge based on a preset data acquisition strategy to obtain corresponding relevant data. The first determining submodule is used to treat the relevant data as the resource data.

[0116] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here.

[0117] In some optional implementations of this embodiment, the lookup module 304 includes: The first search submodule is used to search for target personal terms related to the query content from the personal knowledge base; The second search submodule is used to search for target analysis templates related to the query content from the personal knowledge base; The first integration submodule is used to integrate the target personal terms and the target analysis template to obtain the corresponding integrated data; The second determining submodule is used to treat the integrated data as the knowledge data.

[0118] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here.

[0119] In some optional implementations of this embodiment, the inference module 305 includes: The fusion submodule is used to fuse the resource data and the knowledge data to obtain corresponding fused data; The analysis submodule is used to perform causal relationship analysis on the fused data to obtain corresponding analysis results; The evaluation submodule is used to perform trend prediction and evaluation processing on the fused data to obtain the corresponding trend evaluation results. The second integration submodule is used to integrate the analysis results and the trend assessment results to obtain the corresponding integration results; The third determining submodule is used to use the integration result as the reasoning result.

[0120] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here.

[0121] In some optional implementations of this embodiment, the generation module 306 includes: The extraction submodule is used to extract information from the reasoning results to obtain the corresponding key information; The processing submodule is used to logically sort and organize the key information to obtain the corresponding initial answer data. The first optimization submodule is used to perform language optimization processing on the initial answer data to obtain the corresponding first answer data; The second optimization submodule is used to perform format optimization processing on the first answer data to obtain the corresponding second answer data; The fourth determining submodule is used to use the second answer data as the target answer data.

[0122] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here. In some optional implementations of this embodiment, the return module 307 includes: The second acquisition submodule is used to acquire the preset target display method; The third optimization submodule is used to optimize the target display method to obtain the optimized target display method; Calling a submodule is used to invoke a preset target interface; The display submodule is used to display the target answer data in the target interface based on the optimized target display method.

[0123] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here.

[0124] In some optional implementations of this embodiment, the second acquisition submodule includes: An evaluation unit is used to evaluate the target answer data in terms of data dimension and complexity, and obtain the corresponding evaluation results. The first acquisition unit is used to acquire the user's demand information; The second acquisition unit is used to acquire available display methods; The first determining unit is used to determine the corresponding designated display method from the available display methods based on the evaluation results and the requirement information; The second determining unit is used to select the specified display method as the target display method.

[0125] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the data query method in the aforementioned embodiments, and will not be repeated here. To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0126] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0127] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0128] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for data querying methods. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0129] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the data query method.

[0130] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0131] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the data query method described above.

[0132] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0133] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A data query method, characterized in that, Includes the following steps: Determine whether a query request from a user has been received; wherein the query request carries query content; If so, obtain the user's identity information and call the target assistant instance corresponding to the identity information; Based on the target assistant instance, resource data corresponding to the query content is retrieved from a preset independent resource pool; Retrieve knowledge data corresponding to the query content from a preset personal knowledge base; Based on a preset reasoning strategy, the resource data and the knowledge data are comprehensively reasoned and processed to generate corresponding reasoning results; Based on the reasoning results, corresponding target answer data is generated; The target answer data is returned to the user.

2. The data query method according to claim 1, characterized in that, The step of retrieving resource data corresponding to the query content from a preset independent resource pool based on the target assistant instance specifically includes: Based on the target assistant instance, the query content is semantically parsed to obtain the corresponding semantic parsing results; Based on the semantic parsing results, the domain knowledge index in the independent resource pool is queried to obtain the corresponding target domain knowledge. Based on a preset data acquisition strategy, the target domain knowledge is acquired and processed to obtain corresponding relevant data. The relevant data is used as the resource data.

3. The data query method according to claim 1, characterized in that, The step of retrieving knowledge data corresponding to the query content from a preset personal knowledge base specifically includes: Retrieve target personal terms related to the query content from the personal knowledge base; Retrieve target analysis templates related to the query content from the personal knowledge base; The target personal terms and the target analysis template are integrated to obtain the corresponding integrated data; The integrated data is used as the knowledge data.

4. The data query method according to claim 1, characterized in that, The step of performing comprehensive reasoning processing on the resource data and the knowledge data based on a preset reasoning strategy to generate corresponding reasoning results specifically includes: The resource data and the knowledge data are fused together to obtain the corresponding fused data; The fused data is subjected to causal relationship analysis to obtain the corresponding analysis results; The fused data is subjected to trend prediction and evaluation processing to obtain the corresponding trend evaluation results; The analysis results and the trend assessment results are integrated to obtain the corresponding integrated results; The integration result is used as the reasoning result.

5. The data query method according to claim 1, characterized in that, The step of generating corresponding target answer data based on the reasoning result specifically includes: Information is extracted from the reasoning results to obtain the corresponding key information; The key information is logically sorted and organized to obtain the corresponding initial answer data. The initial response data is subjected to language optimization processing to obtain the corresponding first response data; The first answer data is formatted and optimized to obtain the corresponding second answer data. The second answer data is used as the target answer data.

6. The data query method according to claim 1, characterized in that, The step of returning the target answer data to the user specifically includes: Obtain the preset target display method; The target display method is optimized to obtain an optimized target display method; Call the preset target interface; Based on the optimized target display method, the target answer data is displayed and processed in the target interface.

7. The data query method according to claim 6, characterized in that, The step of obtaining the preset target display method specifically includes: The target answer data is evaluated for its dimensionality and complexity to obtain the corresponding evaluation results. Obtain the user's needs information; Obtain available display options; Based on the evaluation results and the requirement information, a corresponding designated display method is determined from the available display methods; The specified display method is used as the target display method.

8. A data query device, characterized in that, include: The judgment module is used to determine whether a query request from a user has been received; wherein the query request carries query content; The processing module is used to, if so, obtain the user's identity information and call the target assistant instance corresponding to the identity information; The retrieval module is used to retrieve resource data corresponding to the query content from a preset independent resource pool based on the target assistant instance; The search module is used to retrieve knowledge data corresponding to the query content from a preset personal knowledge base; The reasoning module is used to perform comprehensive reasoning processing on the resource data and the knowledge data based on a preset reasoning strategy, and generate corresponding reasoning results. The generation module is used to generate corresponding target answer data based on the reasoning results; The return module is used to return the target answer data to the user.

9. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the data query method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the data query method as described in any one of claims 1 to 7.

Images