Investigation report generation method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of market research automation and artificial intelligence, and discloses a survey report generation method and device, an electronic device and a storage medium. The survey report generation method comprises the following steps: obtaining a survey instruction; obtaining multi-source information based on the survey instruction; performing data processing and analysis on the multi-source information to obtain a data analysis result; performing natural speech understanding processing on the data analysis result to generate core viewpoint data; and generating a target survey report based on the core viewpoint data. Through the above method, the whole process of information collection, analysis and extraction and report writing can be automatically completed based on an input research theme instruction, and the research efficiency and intelligence are effectively improved.

Description

Technical Field

[0001] This application relates to the fields of market research automation and artificial intelligence technology, and in particular to a research report generation method, apparatus, electronic device, and storage medium. Background Technology

[0002] Market research reports are crucial for most companies in making investment decisions, project positioning, and marketing plans. Researchers need to regularly or as needed collect information from multiple dimensions, including macroeconomic data, policies and regulations, competitor projects, and customer profiles, and then summarize and analyze this information to ultimately produce a research report.

[0003] Currently, when writing customer and market research reports, companies often need to collect, screen, and organize a large amount of information from multiple scattered channels such as the Internet and industry databases. After manually analyzing the data, they organize the analysis conclusions into reports. This results in problems such as incomplete information collection, huge time consumption, unstable data analysis, and low report writing efficiency. Summary of the Invention

[0004] The embodiments of this application aim to at least partially address one of the technical problems in the related art. To this end, embodiments of this application propose a research report generation method, apparatus, electronic device, and storage medium.

[0005] The embodiments of this application provide a method for generating a research report, which includes: obtaining a research instruction; obtaining multi-source information based on the research instruction; performing data processing and analysis on the multi-source information to obtain data analysis results; performing natural language understanding processing on the data analysis results to generate core viewpoint data; and generating a target research report based on the core viewpoint data.

[0006] In some implementations, a target research report is generated based on core viewpoint data, including: processing data analysis results and core viewpoint data based on a preset report template to generate a target research report.

[0007] In some implementations, generating a target research report based on core viewpoint data includes: generating an initial research report based on the core viewpoint data; responding to receiving an interactive instruction for the initial research report, performing data processing and analysis on at least one of the initial research report, data analysis results, and core viewpoint data to obtain an interactive response result, wherein the interactive response result includes at least one of report supplementation results, report modification results, and report deepening processing results; and generating a target research report based on the interactive response result and the initial research report.

[0008] In some implementations, acquiring multi-source information based on survey instructions includes: acquiring multi-source information in parallel from at least one external data source and / or at least one internal data source based on survey instructions.

[0009] In some implementations, based on the survey instructions, multi-source information is obtained, including: identifying at least one of the survey topic, survey area, and survey time information in the survey instructions to obtain an intent recognition result; based on the intent recognition result, the survey instructions are decomposed into multiple sub-tasks to be processed; and the multiple sub-tasks to be processed are executed to obtain multi-source information.

[0010] In some implementations, the method further includes updating the target survey report at preset intervals and marking the updated content for output.

[0011] In some implementations, data processing and analysis of multi-source information to obtain data analysis results includes: preprocessing the multi-source information to obtain preprocessing results, wherein the preprocessing includes at least one of information cleaning, information integration, and information statistical analysis; performing chart processing on the multi-source information and / or preprocessing results to obtain chart processing results, wherein the chart processing results include at least one of chart data and tabular data; and obtaining data analysis results based on the preprocessing results and chart processing results.

[0012] The embodiments of this application provide a research report generation device, which includes: a first acquisition module for acquiring research instructions; a second acquisition module for acquiring multi-source information based on the research instructions; a first processing module for performing data processing and analysis on the multi-source information to obtain data analysis results; a second processing module for performing natural language understanding processing on the data analysis results to generate core viewpoint data; and a generation module for generating a target research report based on the core viewpoint data.

[0013] The embodiments of this application provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method of any of the above embodiments.

[0014] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any of the above embodiments. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a survey report generation method provided for an embodiment of this application; Figure 2 A schematic diagram illustrating the generation of a survey report for a multi-agent system driven by an AI large model, provided for the implementation of this application. Figure 3 A schematic diagram of a survey report generation apparatus provided for an embodiment of this application; Figure 4 A block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0016] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0017] Market research reports are crucial for most companies in making investment decisions, project positioning, and marketing plans. Researchers need to regularly or based on specific needs (such as researching a plot of land to be auctioned) collect information from multiple dimensions, including macroeconomics, policies and regulations, the land market, competitor projects, and customer profiles, and then summarize and analyze this information to ultimately produce a research report.

[0018] Currently, when writing customer and market research reports, companies often need to collect, screen, and organize a large amount of information from multiple scattered channels such as the Internet and industry databases. After manually analyzing the data, they organize the analysis conclusions into reports. This results in problems such as incomplete information collection, huge time consumption, unstable data analysis, and low report writing efficiency.

[0019] Specifically, for example, common technical problems encountered by research or investment departments of real estate development companies when writing client and market research reports: 1. Information collection and organization are extremely time-consuming: Writing an in-depth report requires collecting, filtering, and organizing a large amount of information from multiple scattered channels such as the Internet, industry databases, government websites, and internal business systems. This process takes up most of the researchers' time.

[0020] 2. High barriers to entry for data analysis and insights: Transforming collected raw data into valuable conclusions and insights requires researchers to have strong data analysis skills and industry experience. The quality of output is inconsistent and highly dependent on individual ability.

[0021] 3. Low report writing efficiency: Organizing the analysis conclusions into a well-structured, logically clear, and visually appealing report is a time-consuming writing task, especially in scenarios where reports need to be issued regularly and frequently.

[0022] One technology can generate research reports using web crawling and data visualization (BI) tools. Companies can use web crawlers to scrape data from specific websites, import it into BI tools, and researchers can manually drag and drop dimensions and metrics to generate charts. Finally, screenshots of the charts are taken and pasted into Word or PowerPoint, along with text descriptions. This technical solution has the following limitations: 1. Incomplete and unintelligent information gathering: Web crawlers typically require programming for fixed websites and page structures, making them inflexible in dynamically and intelligently discovering and collecting information across the entire web based on a single research topic. They also cannot process reports in unstructured documents such as PDFs.

[0023] 2. Analysis and insights still rely on human intervention: BI tools are merely data "display" tools. They cannot automatically analyze the relationships between data, let alone extract "insights" or "conclusions." The most crucial step from "data" to "insights" relies entirely on human intervention.

[0024] 3. Fragmented report generation process: The entire process consists of three separate stages: "data capture - data analysis - report writing". Data and charts cannot be dynamically updated in the final report. Every time the report is updated, the entire cumbersome process needs to be repeated.

[0025] Therefore, this application proposes a research report generation method that can automatically complete the entire process of information collection, analysis and refinement, and report writing based on input research topic instructions, effectively improving research efficiency and intelligence.

[0026] In the technical solution of this application embodiment, a survey instruction is first obtained, then multi-source information is obtained based on the survey instruction, and then the multi-source information is processed and analyzed to obtain data analysis results. Natural speech understanding processing is performed on the data analysis results to generate core viewpoint data. Finally, a target survey report is generated based on the core viewpoint data. Thus, the entire process of information collection, analysis and refinement and report writing can be automatically completed based on the input research topic instruction, which effectively improves the efficiency and intelligence of the survey.

[0027] Figure 1 This is a flowchart illustrating a research report generation method provided for an embodiment of this application.

[0028] like Figure 1 As shown, the survey report generation method 100 provided in this application includes, for example, steps S110-S150.

[0029] Step S110: Obtain survey instructions.

[0030] For example, a survey instruction can include information such as the survey topic, the survey area, and the survey time. For instance, a survey instruction can be obtained by receiving a survey instruction (survey instruction) entered by the user in a conversational or form-based interface.

[0031] Step S120: Based on the survey instructions, obtain information from multiple sources.

[0032] For example, the survey instructions input by the user can be decomposed to obtain information from different data sources in parallel. The multi-source information can be, for example, raw data related to the survey instructions obtained from multiple data sources.

[0033] Step S130: Perform data processing and analysis on the multi-source information to obtain data analysis results.

[0034] For example, the acquired multi-source information, i.e., the raw data, can be processed by data cleaning, data integration, and statistical analysis to obtain data analysis results, which may include, for example, visualization charts or chart data.

[0035] Step S140: Perform natural language processing on the data analysis results to generate core viewpoint data.

[0036] For example, the data analysis results can be processed by natural speech understanding based on natural language generation (NLG) capabilities, and the analyzed data can be processed into fluent text descriptions with core viewpoints to extract viewpoints, i.e., core viewpoint data.

[0037] Step S150: Generate a target research report based on the core viewpoint data.

[0038] For example, text descriptions (core viewpoint data) generated from natural speech understanding can be processed based on further user feedback and needs to obtain the final target research report.

[0039] In the technical solution of this application embodiment, a survey instruction is first obtained, then multi-source information is obtained based on the survey instruction, and then the multi-source information is processed and analyzed to obtain data analysis results. Natural speech understanding processing is performed on the data analysis results to generate core viewpoint data. Finally, a target survey report is generated based on the core viewpoint data. Thus, the entire process of information collection, analysis and refinement and report writing can be automatically completed based on the input research topic instruction, which effectively improves the efficiency and intelligence of the survey.

[0040] In one example, the survey report generation method of this application is executed based on a multi-agent system driven by a large AI model. This system can automatically decompose the survey instructions input by the user, acquire data, analyze and generate relevant target survey reports, realizing end-to-end automated survey report generation. The following is combined with... Figure 2 Detailed description.

[0041] Figure 2 A schematic diagram illustrating the generation of a research report for a multi-agent system driven by an AI large model, provided as an embodiment of this application.

[0042] like Figure 2 As shown, the AI-driven multi-agent system includes task decomposition agents, information gathering agents, data analysis agents, opinion extraction agents, and report writing agents. The research report generation method based on the AI-driven multi-agent system includes steps S201-S210.

[0043] S201, users enter research instructions in a conversational or form-based interface, including information such as topic, region, and time range (research instructions).

[0044] S202, AI large model receives instructions.

[0045] S203, the system splits instructions into multiple subtasks through a task decomposition agent.

[0046] S204, Information gathering agents acquire data in parallel from multiple channels, such as external data sources from the Internet (news / reports, etc.) and internal data sources from enterprise databases (sales / customer research, etc.) (multi-source information).

[0047] S205, the data analysis agent cleans, integrates, analyzes, and generates charts and other analysis results from multi-source data.

[0048] S206, Opinion Extraction: The agent summarizes core insights (core viewpoints and data) from the analysis results.

[0049] S207, The report writing agent organizes charts and viewpoints into a document according to the template.

[0050] S208 generates a draft of a structured report containing text, charts, and tables.

[0051] S209, the front end displays the report to the user in the form of rich text or an editable interface.

[0052] In S210, users can interactively modify and ask follow-up questions, such as requesting "a detailed explanation of the second part of the report." The system provides feedback to confirm the final draft, thereby obtaining the target research report.

[0053] For example, based on the survey instruction, multi-source information is obtained. For instance, firstly, at least one of the survey topic, survey area, and survey time information in the survey instruction is identified to obtain the intent recognition result; then, based on the intent recognition result, the survey instruction is decomposed into multiple sub-tasks to be processed; and then the multiple sub-tasks to be processed are executed to obtain multi-source information.

[0054] Specifically, users can first input a research instruction (survey instruction) in a conversational or form-based interface, specifying the core research topic (survey theme), geographical scope (survey area), time span (survey time information), etc. The AI ​​big model decomposes the task based on the user's input instruction, breaking it down into a series of executable sub-tasks, such as "collecting land auction data for the region over the past year", "analyzing the unit types and prices of three major competing properties", and "grasping the latest housing purchase policies" (sub-tasks to be processed). These sub-tasks are then executed to obtain multi-source information.

[0055] In the technical solution of this application embodiment, at least one of the research topic, research area, and research time information in the research instruction is first identified to obtain the intent recognition result. Then, based on the intent recognition result, the research instruction is decomposed into multiple sub-tasks to be processed. The multiple sub-tasks to be processed are then executed to obtain multi-source information. Compared with the traditional role of AI as a passive command execution tool, it can autonomously decompose tasks, collect data, analyze and write reports based on the new paradigm of AI "autonomous research", thus achieving a high degree of intelligence.

[0056] For example, based on a survey instruction, multi-source information is obtained, such as obtaining multi-source information from at least one external data source and / or at least one internal data source in parallel based on the survey instruction.

[0057] Specifically, multiple activated "information gathering agents" can retrieve information from different data sources in parallel. Some agents are responsible for accessing internal databases (such as relevant internal enterprise databases, i.e., internal data sources), while others obtain publicly available data, industry reports, etc. (external data sources) from the Internet through search engines, API interfaces, etc., thereby obtaining multi-source data information (multi-source information).

[0058] In the technical solution of this application embodiment, based on the survey instruction, multi-source information is obtained in parallel from at least one external data source or at least one internal data source, thereby enabling dynamic and intelligent data collection across the entire network, providing strong support for data analysis and report generation.

[0059] For example, data processing and analysis are performed on multi-source information to obtain data analysis results. For instance, firstly, the multi-source information is preprocessed to obtain preprocessing results, wherein the preprocessing includes at least one of information cleaning, information integration, and information statistical analysis; then, the multi-source information and / or preprocessing results are subjected to chart processing to obtain chart processing results, wherein the chart processing results include at least one of chart data and tabular data; finally, based on the preprocessing results and chart processing results, data analysis results are obtained.

[0060] Specifically, a "data analysis agent" can receive all collected raw data (multi-source information), perform preprocessing such as cleaning, integration, and statistical analysis (i.e., information cleaning, information integration, and information statistical analysis) to obtain preprocessing results, and call chart generation tools to automatically create relevant data pivot charts, trend charts, regional heat maps, and other visualization elements based on multi-source information or preprocessing results, generating analyzed data and charts (chart processing results, including at least one of chart data and tabular data). Based on the preprocessing results and chart processing results, the data analysis results are obtained.

[0061] In the technical solution of this application embodiment, multi-source information is first preprocessed to obtain preprocessing results, then the multi-source information or preprocessing results are subjected to graph processing to obtain graph processing results, and finally, based on the preprocessing results and graph processing results, data analysis results are obtained. Through the collaborative work of multiple artificial intelligence agents, information collection and analysis are carried out. Through effective task decomposition and collaboration, complex and long-chain research tasks that were difficult for a single model to handle in the past have been completed, achieving a high degree of intelligence.

[0062] Next, based on the data analysis results and the core viewpoint data obtained from natural speech understanding processing, the final research report (targeted research report) is generated, which will be elaborated below.

[0063] For example, a target research report can be generated based on core viewpoint data. For instance, a target research report can be generated by processing data analysis results and core viewpoint data based on a preset report template.

[0064] For example, a target research report is generated based on core viewpoint data. For instance, firstly, an initial research report is generated based on the core viewpoint data; then, in response to receiving an interactive instruction for the initial research report, data processing and analysis are performed on at least one of the initial research report, data analysis results, and core viewpoint data to obtain an interactive response result, wherein the interactive response result includes at least one of the report supplementation result, report modification result, and report deepening processing result; finally, a target research report is generated based on the interactive response result and the initial research report.

[0065] Specifically, for example, the analyzed data and charts (one of the data analysis results) can be filled in according to a preset report template (such as the structure of "market macro environment - land market - commercial housing market - customer group analysis - conclusions and recommendations"). Natural language generation (NLG) capabilities can then be used to refine the analyzed data into fluent text descriptions with core viewpoints, thus obtaining core viewpoint data related to the research report.

[0066] Specifically, the text descriptions obtained above can be used to generate a draft report (initial research report) in a rich text editor and presented to the user. The user can not only modify the text in the draft like a regular document, but also interact with the AI ​​by asking follow-up questions. For example, the user can select a chart and ask the AI ​​to "explain the reason behind this chart", or ask the AI ​​to "add more cases to the third part" (interaction instructions). The AI ​​model will provide corresponding feedback and obtain the interaction response results. Based on the relevant instructions, the user can supplement, modify, and deepen the report to generate the target research report.

[0067] In the technical solution of this application embodiment, an initial research report is first generated based on core viewpoint data. Then, in response to receiving an interactive instruction for the initial research report, data processing and analysis are performed on at least one of the initial research report, data analysis results, and core viewpoint data to obtain an interactive response result. Finally, a target research report is generated based on the interactive response result and the initial research report. Through multiple human agents, effective task decomposition and collaboration are carried out, and information collection, information analysis, and writing are performed autonomously. At the same time, an interactive report interface is provided, allowing users to ask follow-up questions, modify, and deepen the report based on the AI-generated draft through natural language dialogue, thus achieving an efficient combination of human and machine intelligence.

[0068] For example, the generated target research report can be further processed, such as updating the target research report at preset intervals and marking the updated content for output.

[0069] Specifically, the system will automatically run the research task (i.e., the methodology described above) periodically based on the user's periodic update needs, and push the updated data and viewpoints in the report to the user in the form of highlights or summaries, thereby generating periodic reports (targeted research reports), which users can subscribe to. At the same time, the system allows enterprises to upload past internal research reports, meeting minutes, etc. as a private knowledge base for AI. When generating new reports, AI will prioritize searching for information and viewpoints from these more valuable private data.

[0070] In the technical solution of this application embodiment, the target survey report is updated at preset intervals, and the updated content is marked and output, thereby generating a periodic target survey report, automatically presenting the latest effective information to the user, and realizing a high degree of intelligence.

[0071] Figure 3 A schematic diagram of a survey report generation apparatus provided for an embodiment of this application.

[0072] like Figure 3 As shown, the survey report generation device 300 includes: The first acquisition module 310 is used to acquire survey instructions; The second acquisition module 320 is used to acquire multi-source information based on survey instructions; The first processing module 330 is used to perform data processing and analysis on multi-source information to obtain data analysis results; The second processing module 340 is used to perform natural speech understanding processing on the data analysis results to generate core viewpoint data. Module 350 is used to generate target research reports based on core viewpoint data.

[0073] For example, the generation module 350 is also used to: process data analysis results and core viewpoint data based on a preset report template to generate a target research report.

[0074] For example, the generation module 350 is further configured to: generate an initial research report based on core viewpoint data; in response to receiving an interaction instruction for the initial research report, perform data processing and analysis on at least one of the initial research report, data analysis results, and core viewpoint data to obtain an interaction response result, wherein the interaction response result includes at least one of report supplementation results, report modification results, and report deepening processing results; and generate a target research report based on the interaction response result and the initial research report.

[0075] For example, the second acquisition module 320 is also configured to: acquire multi-source information in parallel from at least one external data source and / or at least one internal data source based on the survey instructions.

[0076] For example, the second acquisition module 320 is further configured to: identify at least one of the survey topic, survey area range, and survey time information in the survey instruction to obtain an intent recognition result; based on the intent recognition result, decompose the survey instruction into multiple sub-tasks to be processed; and execute the multiple sub-tasks to be processed to obtain multi-source information.

[0077] For example, the device 300 also includes an update module for: updating the target survey report at preset intervals and marking the updated content for output.

[0078] For example, the first processing module 330 is further configured to: preprocess the multi-source information to obtain a preprocessing result, wherein the preprocessing includes at least one of information cleaning, information integration, and information statistical analysis; perform chart processing on the multi-source information and / or the preprocessing result to obtain a chart processing result, wherein the chart processing result includes at least one of chart data and tabular data; and obtain a data analysis result based on the preprocessing result and the chart processing result.

[0079] It is understandable that the specific functions of the survey report generation device 300 can be referred to the survey report generation method above, and will not be repeated here.

[0080] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0081] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0082] Figure 4 A block diagram of an electronic device provided in an embodiment of this application.

[0083] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method in any of the above embodiments.

[0084] like Figure 4 As shown, for ease of understanding, embodiments of this application illustrate a specific electronic device 400.

[0085] Electronic device 400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 400 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0086] like Figure 4As shown, device 400 includes a computing unit 401, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 402 or a computer program loaded from storage unit 408 into random access memory (RAM) 403. RAM 403 may also store various programs and data required for the operation of electronic device 400. The computing unit 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 404 is also connected to bus 404.

[0087] Multiple components in electronic device 400 are connected to I / O interface 404. These components include: input unit 406, such as a keyboard or mouse; output unit 407, such as various types of displays or speakers; storage unit 408, such as a disk or optical disk; and communication unit 409, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 409 allows electronic device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0088] The computing unit 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 401 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0089] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0090] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0091] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0092] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0093] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0094] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0095] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for generating a survey report, characterized in that, The method includes: Receive survey instructions; Based on the aforementioned survey instructions, information from multiple sources is obtained; The multi-source information is processed and analyzed to obtain data analysis results; The data analysis results are processed using natural language understanding to generate core viewpoint data; Based on the aforementioned core viewpoints and data, a target research report is generated.

2. The method according to claim 1, characterized in that, The generation of the target research report based on the core viewpoint data includes: Based on a preset report template, the data analysis results and the core viewpoint data are processed to generate the target research report.

3. The method according to claim 1 or 2, characterized in that, The generation of the target research report based on the core viewpoint data includes: Based on the aforementioned core viewpoints and data, an initial research report was generated; In response to receiving an interactive instruction for the initial survey report, data processing and analysis are performed on at least one of the initial survey report, the data analysis results, and the core viewpoint data to obtain an interactive response result, wherein the interactive response result includes at least one of the report supplementation result, the report modification result, and the report deepening processing result; Based on the interactive response results and the initial survey report, the target survey report is generated.

4. The method according to claim 1, characterized in that, The process of obtaining multi-source information based on the survey instructions includes: Based on the survey instructions, the multi-source information is obtained in parallel from at least one external data source and / or at least one internal data source.

5. The method according to claim 1, characterized in that, The process of obtaining multi-source information based on the survey instructions includes: Identify at least one of the survey topic, survey area, and survey time information in the survey instruction to obtain the intent recognition result; Based on the intent recognition results, the survey instruction is decomposed into multiple sub-tasks to be processed. The multiple subtasks to be processed are executed to obtain the multi-source information.

6. The method according to claim 1, 2, 4, or 5, characterized in that, Also includes: The target survey report is updated at preset intervals, and the updated content is marked and output.

7. The method according to claim 1, 2, 4, or 5, characterized in that, The process of processing and analyzing the multi-source information to obtain data analysis results includes: The multi-source information is preprocessed to obtain a preprocessing result, wherein the preprocessing includes at least one of information cleaning, information integration, and information statistical analysis; The multi-source information and / or the preprocessing results are subjected to chart processing to obtain chart processing results, wherein the chart processing results include at least one of chart data and tabular data; Based on the preprocessing results and the chart processing results, the data analysis results are obtained.

8. A survey report generation device, characterized in that, The device includes: The first acquisition module is used to acquire survey instructions; The second acquisition module is used to acquire multi-source information based on the survey instructions; The first processing module is used to perform data processing and analysis on the multi-source information to obtain data analysis results; The second processing module is used to perform natural speech understanding processing on the data analysis results to generate core viewpoint data. The generation module is used to generate a target research report based on the core viewpoint data.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images