Investment reference information generation method and related device

Abstract

The invention discloses an investment reference information generation method and a related device, relates to the field of artificial intelligence, and accurately identifies a target intention type corresponding to an investment demand and a key parameter set corresponding to the target intention type by using a large model obtained by training an investment demand sample marked with an intention type. And calling a target tool group corresponding to the target intention type in the quantification strategy tool library, carrying out quantitative analysis on the investment demand in the target data range according to the key parameter set corresponding to the target intention type to obtain a structured output result, and aggregating the structured output result by using a large model to obtain a structured output result. And more stable and accurate investment reference information can be obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to an investment reference information generation method and related device. BACKGROUND

[0002] With the development of artificial intelligence technology, large models are increasingly widely used. In the investment field, investment consultants often need to query and organize individual stock related market information, research reports, information and other materials during investment analysis, and then input the organized materials into a large model to generate investment reference information.

[0003] In the process of generating investment reference information, the large model analyzes and summarizes in combination with the input materials, online search and model built-in parameters to generate investment reference information. However, since the data from online search comes from the Internet, some information on the Internet is unverified and may contain errors, and some information is not updated in time and has a lag, so the investment reference information generated by the large model based on these information is unstable and inaccurate. SUMMARY

[0004] In view of the above problems, the present application provides an investment reference information generation method and related device to achieve the purpose of generating stable and accurate investment reference information. The specific scheme is as follows:

[0005] The first aspect of the present application provides an investment reference information generation method, comprising:

[0006] inputting an investment demand and a first instruction prompt into a large model, performing intent recognition on the investment demand by using the large model, obtaining a target intent type corresponding to the investment demand and a key parameter set corresponding to the target intent type, wherein the large model is trained by using investment demand samples of the labeled intent type and the first instruction prompt;

[0007] calling a target tool group corresponding to the target intent type in a quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to the key parameter set corresponding to the target intent type, and obtaining a structured output result, wherein the target data range includes data published by an institution that has passed reliability authentication;

[0008] inputting the structured output result and a second instruction prompt into the large model, performing aggregation on the structured output result by using the large model, and outputting investment reference information corresponding to the investment demand.

[0009] In a possible implementation, the calling the target tool group corresponding to the target intention type in the quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and obtaining a structured output result, include:

[0010] According to the pre-arranged workflow, the target intention type is identified, and the target tool group corresponding to the target intention type is called to perform quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and obtain a structured output result.

[0011] In a possible implementation, the calling the target tool group corresponding to the target intention type in the quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and obtaining a structured output result, include:

[0012] In the case where the target intention type is stock selection, a stock selection tool group is called to perform quantitative analysis on a key parameter set corresponding to stock selection in a target data range, and a target stock list is obtained.

[0013] In a possible implementation, the calling the target tool group corresponding to the target intention type in the quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and obtaining a structured output result, include:

[0014] The stock selection tool group is called to parse a stock selection condition parameter in the key parameter set corresponding to the stock selection, and at least one stock selection condition is obtained.

[0015] Stocks meeting the at least one stock selection condition are filtered from a database to obtain the target stock list, wherein the database stores stock data corresponding to each stock selection factor collected from the target data range, and a stock selection condition includes a stock selection factor.

[0016] In a possible implementation, the calling the target tool group corresponding to the target intention type in the quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and obtaining a structured output result, include:

[0017] In the case where the target intention type is stock diagnosis, a stock diagnosis tool group is called to perform quantitative analysis on the investment demand in a target data range according to a key parameter set corresponding to stock diagnosis, and a diagnosis object dimension score and an analysis result are obtained.

[0018] In a possible implementation, the calling the diagnosis tool set performs multi-dimensional analysis on the diagnosis object in the target data range in the set of key parameters corresponding to the diagnosis, to obtain the scores and analysis results of each dimension of the diagnosis object.

[0019] The calling the diagnosis tool set performs multi-dimensional analysis on the diagnosis object in the target data range in the set of key parameters corresponding to the diagnosis, to obtain the scores and analysis results of each dimension of the diagnosis object.

[0020] In a possible implementation, the investment reference information corresponding to the investment demand further includes traceability information of the investment reference information.

[0021] The traceability information includes at least one of a name of the target tool set and a data source used to generate the investment reference information.

[0022] The second aspect of the present application provides an investment reference information generation apparatus, comprising:

[0023] The recognition unit is configured to input the investment demand and a first instruction prompt into a large model, perform intent recognition on the investment demand by using the large model, and obtain a target intent type corresponding to the investment demand and a set of key parameters corresponding to the target intent type, wherein the large model is trained by using investment demand samples of the labeled intent type and the first instruction prompt.

[0024] The analysis unit is configured to call a target tool set corresponding to the target intent type in a quantitative strategy tool library, perform quantitative analysis on the investment demand in a target data range according to the set of key parameters corresponding to the target intent type, and obtain a structured output result, wherein the target data range includes data published by an institution that has passed reliability authentication.

[0025] The output unit is configured to input the structured output result and a second instruction prompt into the large model, aggregate the structured output result by using the large model, and output investment reference information corresponding to the investment demand.

[0026] The third aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0027] The memory is configured to store a computer program.

[0028] The processor is configured to execute the computer program, so that the electronic device can implement the investment reference information generation method of the first aspect or any implementation manner of the first aspect.

[0029] The fourth aspect of the present application provides a computer program product comprising computer readable instructions which, when executed on an electronic device, cause the electronic device to implement the investment reference information generation method of the first aspect or any implementation manner of the first aspect.

[0030] The fifth aspect of the present application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, can cause the electronic device to implement the investment reference information generation method of the first aspect or any implementation manner of the first aspect.

[0031] By the above technical solution, the investment reference information generation method provided by the present application uses the large model trained by the labeled attention graph type investment demand sample to accurately identify the target intent type corresponding to the investment demand and the key parameter set corresponding to the target intent type, thereby calling the target tool group corresponding to the target intent type in the quantitative strategy tool library, performing quantitative analysis on the investment demand in the target data range according to the key parameter set corresponding to the target intent type, obtaining a structured output result, and then using the large model to aggregate the structured output result to output investment reference information. Since the target data range includes data published by an institution that has passed reliability authentication, the reliability of the structured output result output by the target tool group is guaranteed, and on this basis, the structured output result output by the target tool group of the large model is aggregated, which ensures that the investment reference information output by the large model after aggregating the structured output result is more stable and accurate. BRIEF DESCRIPTION OF DRAWINGS

[0032] The above and other features, advantages, and aspects of the embodiments of the present disclosure will become more apparent by describing in detail the following specific embodiments with reference to the attached drawings. Throughout the drawings, the same or similar reference numerals refer to the same or similar elements. It should be understood that the drawings are schematic, and the original and elements are not necessarily drawn according to the scale.

[0033] Figure 1 A system architecture diagram is provided for the embodiments of the present application;

[0034] Figure 2 A flowchart of an investment reference information generation method is provided for the embodiments of the present application;

[0035] Figure 3 A flowchart of another investment reference information generation method is provided for the embodiments of the present application;

[0036] Figure 4 A structural diagram of an investment reference information generation device is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0037] The embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation part of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0038] The embodiments of the present application will be described below in conjunction with the drawings. It is known to those skilled in the art that as technology develops and new scenarios appear, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0039] The terms "first", "second", and the like in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances, and this is only a way of distinguishing the objects with the same attributes in the description of the embodiments of the present application. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, so that the processes, methods, systems, products or equipment containing a series of units do not have to be limited to those units, but can include other units not clearly listed or inherent to these processes, methods, products or equipment.

[0040] Referring to Figure 1 , Figure 1 A system architecture diagram is shown. The system can include a terminal 100 and a server 200. The server 200 can provide the investment reference information generation method provided by the embodiments of the present application for one or more terminals.

[0041] Among them, the application program can be installed on the terminal 100, the above-mentioned application program can provide an interface, the terminal 100 can receive the investment demand input by the user on the interface, and send the above-mentioned investment demand to the server 200, the server 200 can perform intent recognition on the investment demand based on the received investment demand using a large model, obtain the target intent type corresponding to the investment demand and the key parameter set corresponding to the target intent type, call the target tool group corresponding to the target intent type in the quantitative strategy tool library, perform quantitative analysis on the investment demand in the target data range according to the key parameter set corresponding to the target intent type, obtain the structured output result, input the structured output result and the second instruction prompt into the large model, aggregate the structured output result using the large model, output the investment reference information corresponding to the investment demand, and return the investment reference information to the terminal 100. It can be seen that for the user, only the investment demand needs to be input, such as "please analyze the stock A", "help me recommend 3 stocks with a price change of 0%-5% and a turnover rate less than 1%", without collecting and sorting materials, and without complex operation, the feedback investment reference information can be obtained, and the user experience is effectively improved.

[0042] It should be understood that, in some optional implementations, the terminal 100 can also complete the action of obtaining the investment reference information based on the received investment demand by itself without the cooperation of the server, and the embodiments of the present application are not limited thereto.

[0043] Next, the product form of the terminal 100 is described. Figure 1

[0044] The terminal 100 in the embodiments of the present application can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiments of the present application do not make any limitation thereto.

[0045] The terminal 100 can include a radio frequency unit, a memory, an input unit, a display unit, a camera (optional), an audio circuit (optional), a speaker (optional), a microphone (optional), a headset jack (optional), a processor, an external interface, a power supply, etc. Those skilled in the art can understand that the above components are only examples and do not constitute a limitation on the terminal or the multifunctional device, and more or fewer components can be included, or some components can be combined or different components can be included.

[0046] The input unit can be used to receive inputted digital or character information, and to generate key signal input related to user settings and function control of the portable multifunctional device. Specifically, the input unit can include a touch screen (optional) and / or other input devices. Specifically, the other input devices can include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, on / off keys, etc.), trackballs, mice, joysticks, etc.

[0047] Wherein, the input device can receive inputted data, etc.

[0048] The display unit can be used to display information inputted by the user or information provided to the user, various menus of the terminal, interactive interfaces, file display, and / or playing of any kind of multimedia files. In the embodiments of the present application, the display unit can be used to display an input interface of the investment demand and an interface of display of the investment reference information, etc.

[0049] Wherein, the memory can be used to store software codes related to the investment reference information generation method, and the processor can execute the steps of the investment reference information generation method, and can also dispatch other units (such as the above input unit and display unit) to realize corresponding functions.​

[0050] The radio frequency unit (optional) can be used for receiving and sending signals in the process of information or conversation.

[0051] In the embodiments of the present application, the radio frequency unit can send data to the server 200 and receive the processing result sent by the server 200.

[0052] It should be understood that the radio frequency unit is optional, which can be replaced by other communication interfaces, for example, it can be a network interface.

[0053] The terminal 100 also includes a power supply (such as a battery) for powering various components.

[0054] The terminal 100 also includes an external interface, which can be a standard Micro USB interface, or a multi-pin connector, which can be used to connect the terminal 100 to other devices for communication, or to connect a charger to charge the terminal 100.

[0055] The server 200 includes a bus, a processor, a communication interface and a memory. The processor, the memory and the communication interface communicate through the bus.

[0056] The memory can be used to store software codes related to the investment reference information generation method, and the processor can execute the steps of the investment reference information generation method of the chip, or can dispatch other units to realize the corresponding functions.

[0057] The embodiments of the present application provide an investment reference information generation method. The investment reference information generation method of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0058] Referring to Figure 2 , Figure 2 The flowchart of the investment reference information generation method provided by the embodiments of the present application is shown in FIG. 1. The investment reference information generation method provided by the embodiments of the present application can include steps 201 to 203, which will be described in detail below. Figure 2

[0059] 201: input the investment demand and the first instruction prompt into the large model, use the large model to perform intent recognition on the investment demand, obtain the target intent type corresponding to the investment demand and the key parameter set corresponding to the target intent type;

[0060] The investment demand is described in natural language, such as "please analyze the stock A", "help me recommend 3 stocks with a price change of 0%-5% and a turnover rate less than 1%" and the like.

[0061] ​The instruction prompt (Prompt) is used to prompt the large model to perform corresponding processing based on the input data. The first instruction prompt is used to prompt the large model to identify the intention type of the input investment demand and the key parameter set corresponding to the intention type.

[0062] The large model is trained using investment demand samples with labeled intention types and the first instruction prompt. The large model is trained to recognize intentions in the investment consultant domain using investment demand samples with labeled intention types and the first instruction prompt. This improves the large model's ability to recognize intentions in the investment consultant scenario, allowing it to accurately identify the target intention type of the investment demand and the key parameter set corresponding to the target intention type.

[0063] Different intention types correspond to different key parameter sets.

[0064] For example, the intention types include stock selection and stock diagnosis.

[0065] The key parameter set corresponding to stock selection includes stock selection condition parameters. For example, if the investment demand is "Help me recommend 3 stocks with a price increase of 0%-5% and a turnover rate less than 1%", the stock selection condition parameters can be represented as an array: { (price increase, 0%-5%), (turnover rate, less than 1%)}.

[0066] The key parameter set corresponding to stock diagnosis includes the diagnosis object. For example, if the investment demand is "Please analyze stock A", the diagnosis object is stock A.

[0067] 202: Call the target tool group corresponding to the target intention type in the quantitative strategy tool library, and perform quantitative analysis on the investment demand in the target data range based on the key parameter set corresponding to the target intention type, to obtain a structured output result;

[0068] If the target intention type is stock selection, the target tool group is the stock selection tool group. If the target intention type is stock diagnosis, the target tool group is the stock diagnosis tool group.

[0069] It should be noted that the target data range includes data published by reliable institutions. Reliability certification can be in various forms, such as official certification, regulatory agency certification, legal supervision certification, etc. That is, when the target tool group performs quantitative analysis on the investment demand, the data source it references is the data source within the target data range, such as the annual report published by listed companies, official data from stock exchanges, and other authoritative data sources, ensuring the reliability of the structured output result output by the target tool group.

[0070] The target tool group output result is a structured output result, such as JSON format data, which is convenient for the large model to aggregate.

[0071] 203: input the structured output result and the second instruction prompt into the large model, aggregate the structured output result by the large model, and output investment reference information corresponding to the investment demand.

[0072] The second instruction prompt is used to prompt the large model to aggregate the structured output result to generate investment reference information.

[0073] Taking the investment demand as "help me recommend 3 stocks with a price fluctuation range of 0%-5% and a turnover rate less than 1%", and the target tool group as the stock selection tool group as an example, the structured output result includes a target stock list meeting the stock selection conditions, and the large model aggregates and analyzes each stock in the target stock list, such as calculating a comprehensive score of each stock according to the parameters of each stock, sorting the stocks according to the comprehensive scores from high to low according to the comprehensive scores of each stock, and outputting the basic information of the stocks as investment reference information according to the sorting. Further, the investment reference information can also include more rich information, such as the investment weight of the stock (the weight can be determined according to the comprehensive score), investment value analysis, risk prompt, and yield characteristics, etc.

[0074] Taking the investment demand as "please analyze stock A", and the target tool group as the stock diagnosis tool group as an example, the structured output result includes the dimension scores and analysis results of stock A, and the large model aggregates and analyzes the dimension scores and analysis results of stock A, such as using a weighted average method to calculate a comprehensive score of stock A according to the dimension scores of stock A, and taking the dimension scores, the comprehensive score, and the risk prompt of stock A as investment reference information.

[0075] In a possible implementation, the investment reference information corresponding to the investment demand can also include traceability information of the investment reference information, the traceability information including at least one of a name of the target tool group and a data source used to generate the investment reference information, so as to facilitate the user to understand the data source of the investment reference information, such as which tool group and which data source (such as which listed company publishes an annual report, which stock exchange, etc.) cited by the tool group for quantitative analysis of the investment demand, and further, the user can also click the traceability information link to directly jump to the data source, such as directly jumping to the webpage of the listed company publishing the annual report.

[0076] The investment reference information generation method provided in the embodiment utilizes a large model trained by investment demand samples of a labeled attention graph type, accurately identifies a target intention type corresponding to the investment demand and a key parameter set corresponding to the target intention type, thereby calling a target tool group corresponding to the target intention type in a quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to the key parameter set corresponding to the target intention type, obtaining a structured output result, and further utilizing the large model to aggregate the structured output result to output investment reference information. Since the target data range includes data published by an institution that has passed reliability authentication, the reliability of the structured output result output by the target tool group is ensured, and on this basis, the structured output result output by the target tool group of the large model is aggregated, so that the investment reference information output by the large model after aggregating the structured output result is more stable and accurate.

[0077] In a possible implementation, the implementation of step 202 in the above embodiment is specifically as follows: according to a pre-arranged workflow, a target intention type is identified, a target tool group corresponding to the target intention type is called, quantitative analysis is performed on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and a structured output result is obtained.

[0078] In the embodiment, the workflow is adopted, the target intention type output by the large model is identified by using a workflow engine, a target tool group corresponding to the target intention type is called according to a pre-set correspondence between intention types and tool groups, quantitative analysis is performed on the investment demand in a target data range according to a key parameter set corresponding to the target intention type, and a structured output result is obtained. The whole process does not need human participation and is completely automated, thereby improving the processing efficiency.

[0079] In a possible implementation, the stock selection tool group provides a factor list query and a conditional stock selection execution interface, and supports intelligent stock screening based on multi-dimensional factors. The stock selection tool group is called to analyze stock selection condition parameters in a key parameter set corresponding to the stock selection, to obtain at least one stock selection condition, and then to screen stocks meeting the at least one stock selection condition from a database to obtain a target stock list. The database stores stock data corresponding to each stock selection factor collected from a target data range, and the stock selection condition includes a stock selection factor.

[0080] An exemplary implementation of the factor list query function includes the following steps A1-A2:

[0081] A1: Factor data collection and arrangement;

[0082] (1) Financial performance factor

[0083] Extract data from company financial statements, including but not limited to operating revenue, net profit, gross profit margin, net profit margin, debt-to-equity ratio, current ratio, and quick ratio. This data can be obtained from listed companies' periodic reports (such as annual and quarterly reports), specifically through web scraping or by collaborating with financial data providers.

[0084] For some complex financial indicators, such as free cash flow, it needs to be calculated using the formula: Free Cash Flow = (Net Cash Flow from Operating Activities - Capital Expenditures). At the same time, the data needs to be cleaned to remove outliers (such as obviously erroneous data) and missing values ​​(handled through interpolation or deletion).

[0085] (2) Transaction situation factor

[0086] Collect stock trading data, including daily trading volume, turnover, turnover rate, and average transaction price. This data can be obtained from the official data interface of the stock exchange or financial data platforms. For example, by calling the stock exchange's API interface, you can obtain detailed transaction data for each trading day.

[0087] (3) Market factors

[0088] Market factors include stock market capitalization (total market capitalization = stock price × total shares outstanding, free float market capitalization = stock price × free float), industry classification, PE (price-to-earnings ratio), PB (price-to-book ratio), etc., which can be obtained from financial data platforms or listed companies' official websites.

[0089] For example, the formula for calculating PE is: PE = (Stock Price / Earnings Per Share). Earnings per share can be obtained from the company's annual report, or the trailing PE ratio (Stock Price / Earnings Per Share over the past 12 months) can be used to more accurately reflect the current valuation.

[0090] A2: Factor storage;

[0091] The collected factor data is stored in a database. Relational databases (such as MySQL or PostgreSQL) or non-relational databases (such as MongoDB) can be used.

[0092] Taking a relational database as an example, a data table is created for each factor. For example, the "Financial Performance Factor Table" can have fields such as stock code, date, operating revenue, and net profit; the "Transaction Status Factor Table" can have fields such as stock code, date, trading volume, trading value, and turnover rate. Each table has a primary key (such as a combination of stock code and date) to uniquely identify a record.

[0093] Based on this, by calling the conditional stock selection execution interface, the stocks meeting the stock selection conditions can be queried in the database. The stock selection conditions can be a combination of multi-dimensional factors, for example, the stock selection conditions are "net profit greater than 100 million yuan, and turnover rate greater than 2%, and price-earnings ratio less than 20 times".

[0094] Therefore, the stock selection conditions include the following stock selection factors:

[0095] Stock selection factor 1: net profit, condition: greater than, threshold: 100 million yuan;

[0096] Stock selection factor 2: turnover rate, condition: greater than, threshold: 2%;

[0097] Stock selection factor 3: price-earnings ratio, condition: less than, threshold: 20 times;

[0098] According to the above stock selection factors (corresponding to the factors in the database), conditions and thresholds, the stocks meeting the stock selection conditions are screened from the database.

[0099] Through the implementation of the above factor list query and conditional stock selection execution interface, the stock selection tool group can support users to intelligently select stocks based on multi-dimensional factors, and improve the efficiency and accuracy of stock selection.

[0100] In one possible implementation, the stock diagnosis tool group includes a plurality of depth analysis interfaces for individual stocks. Exemplarily, the plurality of depth analysis interfaces for individual stocks cover eight key dimensions of basic market information, comprehensive score, technical analysis, main force trend, market heat, theme perspective, financial valuation and related theme mining, to realize comprehensive diagnosis and evaluation of target stocks. By calling the plurality of depth analysis interfaces for individual stocks in the stock diagnosis tool, the diagnosis objects are analyzed respectively to obtain the scores and analysis results of the diagnosis objects in each dimension.

[0101] Please refer to Figure 3The illustrated investment reference information generation method flowchart shows that the user inputs a query (i.e., an investment demand), a workflow engine inputs the investment demand and a first instruction prompt into a large model, calls the large model to perform intent recognition on the investment demand, and the large model determines whether the intent corresponding to the investment demand is clear. If the intent is not clear, the user is prompted to re-input the investment demand, and the investment demand can be clarified through multiple rounds of interaction until the intent is clear. If the intent is clear, an intent type (stock diagnosis, stock selection, or other questions) is output. If the intent type is stock diagnosis, a set of key parameters in the investment demand, such as a stock name or code, is extracted, a stock diagnosis API corresponding to a stock diagnosis tool group is called, and a structured output result is obtained by performing quantitative analysis on the investment demand in a target data range according to the set of key parameters. If the intent type is stock selection, a set of key parameters in the investment demand, such as a selection factor, is extracted, a stock selection API corresponding to a stock selection tool group is called, and a structured output result is obtained by performing quantitative analysis on the investment demand in a target data range according to the set of key parameters. The workflow engine inputs the structured output result output by the tool group (the stock diagnosis tool group or the stock selection tool group) and a second instruction prompt into the large model, aggregates the structured output result by using the large model, and generates investment reference information rhetoric. If the intent type is other questions, a pre-set bottom-up rhetoric is displayed.

[0102] The above introduces an investment reference information generation method provided by an embodiment of the present application. The following will introduce a device for executing the investment reference information generation method.

[0103] Please refer to Figure 4 , Figure 4 FIG. 4 is a structural schematic diagram of an investment reference information generation device provided by an embodiment of the present application. As shown in Figure 4 FIG. 4, the investment reference information generation device 400 includes:

[0104] An identification unit 401 is configured to input an investment demand and a first instruction prompt into a large model, perform intent recognition on the investment demand by using the large model, and obtain a target intent type corresponding to the investment demand and a set of key parameters corresponding to the target intent type, wherein the large model is trained by using investment demand samples of the intent type and the first instruction prompt.

[0105] An analysis unit 402 is configured to call a target tool group corresponding to the target intent type in a quantitative strategy tool library, perform quantitative analysis on the investment demand in a target data range according to the set of key parameters corresponding to the target intent type, and obtain a structured output result, wherein the target data range includes data published by an institution that has passed reliability authentication.

[0106] The output unit 403 is configured to input the structured output result and the second instruction prompt into the large model, aggregate the structured output result by using the large model, and output investment reference information corresponding to the investment demand.

[0107] In a possible implementation, the analysis unit 402 is specifically configured to identify the target intent type according to a pre-arranged workflow, call a target tool group corresponding to the target intent type, and perform quantitative analysis on the investment demand according to a key parameter set corresponding to the target intent type, to obtain a structured output result.

[0108] In a possible implementation, when the target intent type is stock selection, the analysis unit 402 is configured to call a stock selection tool group to perform quantitative analysis on the investment demand in the target data range according to a key parameter set corresponding to the stock selection, to obtain a target stock list.

[0109] In a possible implementation, when the target intent type is stock selection, the analysis unit 402 is configured to call the stock selection tool group to analyze a stock selection condition parameter in the key parameter set corresponding to the stock selection, to obtain at least one stock selection condition, and filter stocks meeting the at least one stock selection condition from a database to obtain the target stock list, wherein the database stores stock data corresponding to each stock selection factor collected from the target data range, and the stock selection condition includes a stock selection factor.

[0110] In a possible implementation, when the target intent type is stock diagnosis, the analysis unit 402 is configured to call a stock diagnosis tool group to perform quantitative analysis on the investment demand in the target data range according to a key parameter set corresponding to the stock diagnosis, to obtain a score of each dimension of a stock diagnosis object and an analysis result.

[0111] In a possible implementation, when the target intent type is stock diagnosis, the analysis unit 402 is configured to call the stock diagnosis tool group to perform multi-dimensional analysis on the stock diagnosis object in the key parameter set corresponding to the stock diagnosis in the target data range, to obtain the score of each dimension of the stock diagnosis object and the analysis result.

[0112] In a possible implementation, the investment reference information corresponding to the investment demand further includes traceability information of the investment reference information, and the traceability information includes at least one of a name of the target tool group and a data source used to generate the investment reference information.

[0113] The investment reference information generation device provided in the embodiment accurately identifies a target intention type corresponding to an investment demand and a key parameter set corresponding to the target intention type, so as to call a target tool group corresponding to the target intention type in a quantitative strategy tool library, performs quantitative analysis on the investment demand according to the key parameter set corresponding to the target intention type, obtains a structured output result, and then aggregates the structured output result by using the large model to output investment reference information. Since only the large model is used to identify the target intention type corresponding to the investment demand and aggregate the structured output result output by the target tool group, but the target tool group is called to perform quantitative analysis on the investment demand in a target data range according to the key parameter set corresponding to the target intention type, and since the target data range includes data published by an institution that has passed reliability authentication, the reliability of the structured output result output by the target tool group is ensured, so that the investment reference information output after the large model aggregates the structured output result is more stable and accurate.

[0114] The electronic device provided in the embodiment includes at least one processor and a memory connected with the processor, and the memory is configured to store a computer program.

[0115] The memory is configured to store a computer program.

[0116] The processor is configured to execute the computer program, so that the electronic device can implement any investment reference information generation method provided in the embodiments.

[0117] The electronic device can be the terminal 100 or the server 200 shown in the above. Figure 1 The electronic device can be the terminal 100 or the server 200 shown in the above.

[0118] The computer program product includes computer readable instructions, and when the computer readable instructions run on the electronic device, the electronic device can implement any investment reference information generation method provided in the embodiments.

[0119] The computer readable storage medium carries one or more computer programs, and when the one or more computer programs are executed by the electronic device, the electronic device can implement any investment reference information generation method provided in the embodiments.

[0120] It should be noted that the apparatus embodiments described above are merely illustrative, and the units described as separate units can or can not be physically separate, and the units displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment. In addition, the connection relationship between the modules in the apparatus embodiment provided in the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines.

[0121] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and the necessary general hardware, and of course can also be realized by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. Generally, functions completed by computer programs can be easily realized by corresponding hardware, and the specific hardware structure for realizing the same function can also be various, such as analog circuit, digital circuit or special circuit, etc. However, for the present application, software program implementation is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products, which are stored in readable storage media, such as computer floppy disks, U disks, mobile hard disks, ROM, RAM, magnetic or optical disks, etc., including a plurality of instructions for making a computer device (which can be a personal computer, a training device, or a network device, etc.) execute the methods described in various embodiments of the present application.

[0122] In the above embodiments, all or part can be realized by software, hardware, firmware or any combination thereof. When realized by software, it can be realized in the form of a computer program product in whole or in part.

[0123] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, training device or data center to another website, computer, training device or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be stored by the computer or a data storage device such as a training device, a data center, etc. integrated with one or more available media sets. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)), etc.

Claims

1. An investment reference information generation method characterized by comprising: The method comprises the following steps: inputting investment demand and a first instruction prompt into a large model, performing intent recognition on the investment demand by using the large model, obtaining a target intent type corresponding to the investment demand and a key parameter set corresponding to the target intent type, wherein the large model is trained by using investment demand samples of the labeled intent type and the first instruction prompt; calling a target tool group corresponding to the target intent type in a quantitative strategy tool library, performing quantitative analysis on the investment demand in a target data range according to the key parameter set corresponding to the target intent type, and obtaining a structured output result, wherein the target data range includes data published by an institution that has passed reliability authentication; inputting the structured output result and a second instruction prompt into the large model, performing aggregation on the structured output result by using the large model, and outputting investment reference information corresponding to the investment demand.

2. The investment reference information generation method according to claim 1, characterized by, The calling of the target tool group corresponding to the target intent type in the quantitative strategy tool library, the quantitative analysis of the investment demand in the target data range according to the key parameter set corresponding to the target intent type, and the obtaining of the structured output result comprise: identifying the target intent type according to a pre-arranged workflow, calling the target tool group corresponding to the target intent type, and performing quantitative analysis on the investment demand in the target data range according to the key parameter set corresponding to the target intent type to obtain the structured output result.

3. The investment reference information generation method according to claim 1, characterized by, The calling of the target tool group corresponding to the target intent type in the quantitative strategy tool library, the quantitative analysis of the investment demand in the target data range according to the key parameter set corresponding to the target intent type, and the obtaining of the structured output result comprise: in a case where the target intent type is stock selection, calling a stock selection tool group to perform quantitative analysis on the investment demand in the target data range according to a key parameter set corresponding to stock selection to obtain a target stock list.

4. The investment reference information generation method according to claim 3, characterized by, The calling of the stock selection tool group to perform quantitative analysis on the investment demand in the target data range according to the key parameter set corresponding to stock selection to obtain the target stock list comprises: calling the stock selection tool group to parse a stock selection condition parameter in the key parameter set corresponding to stock selection to obtain at least one stock selection condition; filtering stocks meeting the at least one stock selection condition from a database to obtain the target stock list, wherein the database stores stock data corresponding to each stock selection factor collected from the target data range, and a stock selection condition includes a stock selection factor.

5. The investment reference information generation method according to claim 1, characterized by, The calling of the target tool group corresponding to the target intent type in the quantitative strategy tool library, the quantitative analysis of the investment demand in the target data range according to the key parameter set corresponding to the target intent type, and the obtaining of the structured output result comprise: in a case where the target intent type is stock diagnosis, calling a stock diagnosis tool group to perform quantitative analysis on the investment demand in the target data range according to a key parameter set corresponding to stock diagnosis to obtain a diagnosis object dimension score and an analysis result.

6. The investment reference information generation method according to claim 5, characterized by, The calling diagnosis tool group diagnoses the corresponding key parameter set of the target data range to quantitatively analyze the investment demand, obtains the diagnosis object dimension score and analysis result, including: The calling diagnosis tool group diagnoses the corresponding key parameter set of the target data range to quantitatively analyze the investment demand, obtains the diagnosis object dimension score and analysis result, including:

7. The investment reference information generation method according to any one of claims 1 to 6, characterized by, The investment reference information corresponding to the investment demand further includes traceability information of the investment reference information; The traceability information includes at least one of the name of the target tool group and the data source used to generate the investment reference information.

8. An investment reference information generation device characterized by comprising: Including: The identification unit inputs the investment demand and the first instruction prompt into the large model, and uses the large model to identify the intention of the investment demand, to obtain the target intention type corresponding to the investment demand and the key parameter set corresponding to the target intention type, wherein the large model is trained using investment demand samples of labeled intention types and first instruction prompts; The analysis unit calls the target tool group corresponding to the target intention type in the quantitative strategy tool library, and quantitatively analyzes the investment demand in the target data range according to the key parameter set corresponding to the target intention type, to obtain a structured output result, wherein the target data range includes data published by institutions that have passed reliability authentication; The output unit inputs the structured output result and the second instruction prompt into the large model, and uses the large model to aggregate the structured output result, to output the investment reference information corresponding to the investment demand.

9. An electronic device, comprising: Including at least one processor and a memory connected to the processor, wherein: The memory is used to store a computer program; The processor is used to execute the computer program, so that the electronic device can implement the investment reference information generation method according to any one of claims 1 to 7.

10. A computer program product, characterised in that, The computer readable instructions, when executed on an electronic device, cause the electronic device to implement the investment reference information generation method according to any one of claims 1 to 7.

11. A computer storage medium, characterized in that The storage medium carries one or more computer programs, which, when executed by an electronic device, can enable the electronic device to implement the investment reference information generation method according to any one of claims 1 to 7.

Images