Industrial prosperity determination method, device, equipment, medium and product

Abstract

The embodiment of the invention discloses an industry prosperity determination method and device, equipment, a medium and a product. The method comprises the steps of obtaining an industry analysis text of a target object; inputting the industry analysis text into a pre-trained large-scale language model, and executing the large-scale language model to obtain an industry consistency expected index output by the large-scale language model; obtaining an industry profitability index, an industry growth ability index and an industry debt paying ability index; according to the industry consistency expected index, the industry profit ability index, the industry growth ability index and the industry debt paying ability index, determining an industry profit degree standby value; obtaining a fund flow trend index in the industry, and determining a correction value according to the fund flow trend index; and correcting the industry prosperity spare value based on the correction value, and determining the final industry prosperity. According to the technical scheme, the industry prosperity can be determined more accurately.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data analysis, and in particular, to a method, apparatus, device, medium and product for determining industry prosperity Background Art

[0002] Prosperity is a description of the rise and fall of the economic or business cycle. High prosperity means that the economic or business cycle is in the upward stage, while low prosperity means that the economic or business cycle is in the downward stage. Therefore, the prosperity of an industry is an indicator reflecting the development trend of the industry and is a barometer of the industry's development status. At present, the prosperity of an industry is mostly determined by institutions or analysts based on their own experience, which is not accurate. Therefore, there is an urgent need for an objective industry prosperity analysis scheme to accurately determine the prosperity of an industry. Summary of the Invention

[0003] To solve the problems in the related art, embodiments of the present disclosure provide a method, apparatus, device, medium and product for determining industry prosperity.

[0004] In a first aspect, embodiments of the present disclosure provide a method for determining industry prosperity.

[0005] Specifically, the method for determining industry prosperity includes:

[0006] Obtain the industry analysis text of the target object;

[0007] Input the industry analysis text into a pre-trained large language model, execute the large language model, and obtain the industry consensus expectation index output by the large language model;

[0008] Obtain industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators;

[0009] Determine the standby value of industry prosperity according to the industry consensus expectation index, industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators;

[0010] Obtain the capital flow direction index in the industry, and determine the correction value according to the capital flow direction index;

[0011] Based on the correction value, correct the standby value of industry prosperity to determine the final industry prosperity.

[0012] In a second aspect, embodiments of the present disclosure provide a device for determining industry prosperity, including:

[0013] A text acquisition module, configured to obtain the industry analysis text of the target object;

[0014] An extraction module, configured to input the industry analysis text into a pre-trained large language model, execute the large language model, and obtain industry consistent expected indicators output by the large language model;

[0015] An indicator acquisition module, configured to acquire industry profitability indicators, industry growth ability indicators, and industry debt paying ability indicators;

[0016] A backup value determination module, configured to determine a backup value of industry prosperity according to the industry consistent expected indicators, industry profitability indicators, industry growth ability indicators, and industry debt paying ability indicators;

[0017] A correction value determination module, configured to acquire a capital flow direction indicator within the industry, and determine a correction value according to the capital flow direction indicator;

[0018] A correction module, configured to correct the backup value of industry prosperity based on the correction value to determine the final industry prosperity.

[0019] In a third aspect, an embodiment of the present disclosure provides an electronic device, including a memory and a processor, wherein the memory is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the method according to any one of the first aspects.

[0020] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, on which computer instructions are stored, and when the computer instructions are executed by a processor, the method according to any one of the first aspects is implemented.

[0021] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including computer instructions, and when the computer instructions are executed by a processor, the method steps according to any one of the first aspects are implemented.

[0022] According to the technical solution provided by the embodiment of the present disclosure, industry consistent expected indicators can be extracted from the current object's industry analysis text, and combined with industry profitability indicators, industry growth ability indicators, and industry debt paying ability indicators to determine a backup value of industry prosperity. At the same time, a capital flow direction indicator is used to determine a correction value, and the backup value of industry prosperity is corrected based on the correction value to determine the final industry prosperity; by introducing a capital flow direction indicator to judge the general decline pattern of the market and accordingly correct the backup value of industry prosperity to determine the final industry prosperity, the industry prosperity can be determined more accurately.

[0023] It should be understood that the above general description and subsequent detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Description of the Drawings

[0024] In conjunction with the accompanying drawings, other features, objects, and advantages of the present disclosure will become more apparent through the following detailed description of non-limiting embodiments. In the drawings:

[0025] Figure 1 A flowchart showing a method for determining the industry prosperity according to an embodiment of the present disclosure;

[0026] Figure 2 A structural block diagram showing an apparatus for determining the industry prosperity according to an embodiment of the present disclosure;

[0027] Figure 3 A structural block diagram showing an electronic device according to an embodiment of the present disclosure;

[0028] Figure 4 A schematic structural diagram showing a computer system suitable for implementing the method according to an embodiment of the present disclosure. Detailed Embodiments

[0029] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. In addition, for clarity, parts unrelated to the description of the exemplary embodiments are omitted in the drawings.

[0030] In the present disclosure, it should be understood that terms such as "including" or "having" are intended to indicate the presence of features, numbers, steps, actions, components, parts, or combinations thereof disclosed in this specification, and are not intended to exclude the possibility of the presence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0031] In addition, it should be noted that, without conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other. The present disclosure will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0032] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of the relevant data need to comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding operation entrances are provided for the user to select to authorize or refuse.

[0033] Figure 1 A flowchart showing the determination of the industry prosperity according to an embodiment of the present disclosure. As Figure 1 shown, the method for determining the industry prosperity includes the following steps S101 - S106:

[0034] In step S101, obtain the industry analysis text of the target object;

[0035] In step S102, input the industry analysis text into a pre-trained large language model, execute the large language model, and obtain the industry consensus expectation indicators output by the large language model;

[0036] In step S103, obtain the industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators;

[0037] In step S104, determine the standby value of the industry prosperity based on the industry consensus expectation indicators, industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators;

[0038] In step S105, obtain the capital flow trend indicators in the industry, and determine the correction value according to the capital flow trend indicators;

[0039] In step S106, correct the standby value of the industry prosperity based on the correction value to determine the final industry prosperity.

[0040] In a possible implementation manner, this method for determining the industry prosperity is applicable to devices such as computers, computing devices, servers, and server clusters that can execute the determination of the industry prosperity.

[0041] In a possible implementation manner, the target object refers to an analyst or an analysis institution, and the industry analysis text can be an industry analysis report of the analyst or the analysis institution, and can be collected from the websites of industry analysis report publishing institutions, research institutions, or relevant professional websites according to the type and scope of the industry analysis report input by the user using an API (Application Programming Interface).

[0042] In a possible implementation manner, a large language model (LLM) is a deep learning algorithm that can execute various natural language processing tasks. For example, it can perform various tasks such as recognition, translation, prediction, text generation, or other content.

[0043] In a possible implementation, the large language model is mainly used to extract industry consensus expectation indicators from industry analysis texts. The consensus expectation refers to an expectation in which multiple analysis institutions or analysts reach a consensus on the future performance of the industry. The industry consensus expectation indicators can be the consensus expected cash flow per share. Or, the industry consensus expectation indicators can also be the expected return on equity (ROE) for the next year, the ROE expectation for the next two years, the expected net asset per share for the next two years, the expected cash flow per share for the next two years, and so on. The industry analysis text can be input into the large language model, and using the super strong language understanding ability of the large language model, the content related to the industry consensus expectation indicators in the industry analysis text can be identified, and the industry consensus expectation indicators can be output accordingly.

[0044] In a possible implementation, the industry profitability indicators refer to a category of indicators that can reflect the profitability of the industry. For example, they can be the original values of ROE, net profit, earnings per share, ROA (Return on Assets), and their year-on-year growth rates, month-on-month growth rates, year-on-year and month-on-month growth rates, etc. of the enterprises in the industry. The industry growth indicators refer to a category of indicators that can reflect the growth ability of the industry. For example, they can be the operating income of the enterprises in the industry. The debt repayment ability indicators refer to a category of indicators that can reflect the debt repayment ability of the enterprises in the industry. For example, they can be the current ratio, etc.

[0045] In a possible implementation, according to the predetermined evaluation rules, based on the industry consensus expectation indicators, industry profitability indicators, industry growth indicators, and industry debt repayment ability indicators, the current prosperity of the industry can be estimated to obtain a standby value of industry prosperity. For example, according to the predetermined scoring rules, based on the indicator values corresponding to the industry consensus expectation indicators, industry profitability indicators, industry growth indicators, and industry debt repayment ability indicators, these four categories of indicators can be scored, and the scores of these four categories of indicators can be weighted and averaged to obtain the standby value of industry prosperity; or, a scoring model can also be used for scoring. The industry consensus expectation indicators, industry profitability indicators, industry growth indicators, and industry debt repayment ability indicators can be input into a pre-trained scoring model, and the scoring model can be executed to obtain the standby value of industry prosperity output by the scoring model. The scoring model can be trained using training data, and the training data can be the industry consensus expectation indicators, industry profitability indicators, industry growth indicators, and industry debt repayment ability indicators within a historical time period and the quantitative values of historical real industry prosperity.

[0046] In a possible implementation, it is found through research that in the scenario of a general decline in the industry, the backup value of the industry prosperity may not be accurate as the industry prosperity. To solve this problem, it is considered to introduce an indicator of the capital flow direction within the industry. By judging the general decline pattern of the market through the withdrawal of the capital flow within the industry in the market, the backup value of the industry prosperity is corrected to determine the final industry prosperity.

[0047] In a possible implementation, the indicator of the capital flow direction refers to an indicator that can reflect the capital flow direction within the industry. For example, it can be the indicator of the northbound capital flow direction or the indicator of the main capital flow direction. The northbound capital flow refers to the capital flow from Hong Kong into the mainland. The main capital flow refers to the total transaction value of large orders and super-large orders in the market, which is the capital flow that can affect the stock market and even control the short-term trend of the stock market. The direction of the main capital flow includes main net inflow and main net outflow. Generally, the flow ratio of the main capital flow or the northbound capital flow, and the flow direction of the year-on-year incremental revenue can be used as the indicator of the capital flow direction. Or, the indicator of the capital flow direction can also be the proportion of the main capital circulation, and the proportion of the main capital circulation = the net inflow of the main capital / the free float market capitalization.

[0048] In a possible implementation, the scheme for correcting the backup value of the industry prosperity according to the indicator of the capital flow direction can be: obtaining a correction value according to the indicator of the capital flow direction, and weighted averaging the backup value of the industry prosperity and the correction value to obtain the final industry prosperity, that is, the final industry prosperity = α1 * the backup value of the industry prosperity + α2 * the correction value. For example, α1 can take a value of 1, and α2 can be searched within the range from 0 to 2 with a step size of 0.05 to determine the value of α2 when the backtest effect is the best.

[0049] In a possible implementation, using the industry prosperity can quickly analyze and judge the development stability, risks, and future development trends of the industry, and can also measure the operating conditions of enterprises through the industry prosperity, and obtain the guiding basis for the production, operation, and investment of enterprises. Therefore, the industry prosperity can be applied to fields such as industry operation monitoring and risk warning, enterprise production, operation, and investment, enterprise loans, and customer analysis.

[0050] This implementation can extract the industry consensus expectation indicator from the current industry analysis text of the object, and combine the industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators to determine the backup value of the industry prosperity. At the same time, use the indicator of the capital flow direction to determine the correction value, and correct the backup value of the industry prosperity based on the correction value to determine the final industry prosperity; by introducing the indicator of the capital flow direction to judge the general decline pattern of the market, and accordingly correct the backup value of the industry prosperity to determine the final industry prosperity, the industry prosperity can be determined more accurately.

[0051] In a possible implementation, the method further includes:

[0052] Obtain the historical industry analysis texts of each object;

[0053] Extract the historical market data proposed by each object from the historical industry analysis texts of each object;

[0054] Obtain the real market data corresponding to the historical market data proposed by each object;

[0055] Determine the target object from each object by comparing the historical market data proposed by each object and its corresponding real market data.

[0056] In this implementation, in order to ensure the accuracy of the industry prosperity, the industry consensus expectation index can be extracted from high-quality industry analysis texts, where the high-quality industry analysis texts refer to the industry analysis reports made by the target object, and the high target object refers to the object that can make high-quality industry analysis reports.

[0057] In this implementation, first, the industry analysis texts made by multiple objects within a historical time period, that is, the historical industry analysis texts, can be obtained. The historical market data proposed by each object can be extracted from the historical industry analysis texts of each object. A large language model can be used for extraction. These historical market data are proposed by each object after its own industry analysis.

[0058] In this implementation, after this historical time period, the real market data corresponding to these historical market data can be obtained, and the historical market data analyzed by each object is compared with its corresponding real market data. For example, the similarity between all the historical market values proposed by the object and its corresponding real market values can be calculated. The higher the similarity, the more accurate the analysis. Each object can be sorted according to this similarity, and the higher the similarity, the higher the ranking. The object with a higher ranking can be used as the target object. For example, the cash flow per share in 2022 analyzed by each object can be extracted from the industry analysis texts of each object in 2020, and then the real cash flow per share in 2022 can be obtained. The cash flow per share in 2022 analyzed by each object is compared with the real cash flow per share in 2022, and each object can be sorted according to the similarity degree. The higher the similarity, the higher the ranking. The objects ranked in the top 60% can be used as the target objects.

[0059] In this implementation, the target object is screened by comparing the analysis data of each object with the real data that has occurred. In this way, high-quality objects can be screened, and then high-quality industry analysis texts can be obtained to extract relatively accurate industry consensus expectation indexes, thereby making the final industry prosperity more accurate.

[0060] In a possible implementation manner, determining a target object from the respective objects by comparing the historical market data proposed by the respective objects and their corresponding true market data includes:

[0061] Comparing the historical market data proposed by the respective objects and their corresponding true market data to determine the difference situation between the historical market data proposed by the respective objects and their corresponding true market data;

[0062] For each object, determining the analysis accuracy of the object according to the difference situation corresponding to the object;

[0063] Determining the analysis independence of the object according to the difference situation corresponding to the object and the difference situations corresponding to other objects;

[0064] Determining a target object from the respective objects according to the analysis accuracy and analysis independence of the respective objects.

[0065] In this implementation manner, the historical market data and its corresponding true market data can be compared to obtain the difference situation between the historical market data proposed by the respective objects and their corresponding true market data. This difference situation can be the difference between each historical market value and the true market value, or the difference in sentiment such as optimism or pessimism about the market, etc.

[0066] In this implementation manner, for each object, the analysis accuracy of the object can be determined according to the difference situation corresponding to the object. For example, if the difference between a certain historical market value proposed by the object and its corresponding true market value is within a predetermined error range (for example, the absolute value of the difference is within 10% of the true market value), it indicates that this historical market value is accurate. Thus, count whether the various historical market values proposed by the object are accurate and calculate the accuracy of the object.

[0067] In this implementation manner, if the industry view (optimistic / pessimistic) and the profit forecast direction of a certain analyst always follow the views of other industry analysts, it indicates that the analyst does not have his own analysis logic and judgment basis and does not belong to a high-quality analyst. Therefore, in addition to the above accuracy, whether each object has its own unique view in market judgment also belongs to the determination conditions of the target object.

[0068] In this implementation manner, the difference situation of the object can be compared with the difference situations of other objects. If, for a certain item of data, the difference situation corresponding to the object is small and conforms to the true market performance, but the difference situations corresponding to most other objects are large, it indicates that the object holds a few correct views on this item of data. The proportion of the items of data for which the object holds a few correct views in all the historical market data proposed by the object can be counted, and this proportion can be used as the analysis independence of the object.

[0069] In this embodiment, the target object can be determined from each object based on the analysis accuracy and analysis independence of each object; for example, for each object, the analysis accuracy and analysis independence of the object can be weighted averaged to obtain an evaluation value of the object, and the objects are sorted from high to low according to the evaluation value, and the objects with the highest ranking (such as 60%) are determined as target objects.

[0070] This embodiment determines the target object based on the analysis accuracy and analysis independence of each object, and can select a better target object.

[0071] In a possible implementation, inputting the industry analysis text into a pre-trained large-scale language model, executing the large-scale language model, and obtaining the industry consensus expected index output by the large-scale language model includes:

[0072] Inputting the industry analysis text into the pre-trained large-scale language model, and extracting the original consistent expected index from the industry analysis text by the large-scale language model;

[0073] Acquire, from the industry analysis text, sentiment texts whose distance from the original consistent expectation index is within a certain range through the large-scale language model;

[0074] Determining fine-tuning parameters according to the emotional text;

[0075] The original consistent expectation index is fine-tuned based on the fine-tuning parameters to obtain the industry consistent expectation index output by the large-scale language model.

[0076] In this implementation, when each target object proposes a consistent expectation index in the industry analysis text, it will be affected by the market sentiment of the target object itself. For example, if the market sentiment of the target object is optimistic, the proposed consistent expectation index will be high, and if the market sentiment of the target object is pessimistic, the proposed consistent expectation index will be low. Therefore, it is necessary to fine-tune the consistent expectation index originally recorded in the industry analysis text to obtain accurate industry consistent expectation index.

[0077] In this embodiment, after inputting the industry analysis text into the pre-trained large language model, the large language model can extract the original consensus expectation indicators in the industry analysis text. At this time, the large language model can also identify sentiment texts within a certain range of distance from the original consensus expectation indicators in the industry analysis text. For example, "The annual performance growth is good. It is recommended to actively layout and be optimistic about the current market situation" or "The industry leader makes steady progress, promotes nationalization with high quality, and maintains a neutral view", etc. The fine-tuning parameters can be determined based on these sentiment texts. For example, if the sentiment text is overly optimistic, the fine-tuning parameter is a parameter to be decreased, which is used to decrease the original consensus expectation indicator; if the sentiment text is overly pessimistic (with a high degree of pessimism), the fine-tuning parameter is a parameter to be increased, which is used to increase the original consensus expectation indicator, and so on. After fine-tuning the original consensus expectation indicators based on the fine-tuning parameters, the industry consensus expectation indicators output by the large language model can be obtained. At this time, the output industry consensus expectation indicators will be more accurate, and thus the industry prosperity can be accurately determined based on this.

[0078] In a possible implementation manner, the method further includes:

[0079] Obtain a preset corpus database, where the corpus database stores corpus vectors that conform to the consensus expectation characteristics;

[0080] The step of inputting the industry analysis text into the pre-trained large language model and extracting the original consensus expectation indicators from the industry analysis text by the large language model includes:

[0081] Input the industry analysis text into a large language model based on retrieval-augmented generation. Through the large language model based on retrieval-augmented generation, perform similarity search in the industry analysis text based on the corpus vectors in the corpus database to obtain consensus expectation key information;

[0082] Extract the original consensus expectation indicators from the consensus expectation key information through the large language model.

[0083] In this embodiment, in order to improve the accuracy of the large language model in extracting consensus expectation indicators, the method of manual annotation and screening can be adopted to select the corpus that conforms to the consensus expectation characteristics, and store these corpus in the form of vectors in the corpus database to provide data support for subsequent identification work.

[0084] In this embodiment, the large language model is a large language model based on Retrieval Augmented Generation (RAG). The industry analysis text can be input into the large language model based on Retrieval Augmented Generation. The large language model will use the industry analysis text as part of the prompt, and indicate in the prompt that the large language model performs text reading and filters out the key information of the consistent expectation. In the process of generating an answer, the large language model can call the corpus database to perform a similarity search on the industry analysis text, and identify the paragraph text that meets the characteristics of the consistent expectation as the key information of the consistent expectation.

[0085] For example, the identified key information of the consistent expectation can be as follows:

[0086] "Company A: Adjust the profit forecast according to the performance forecast. We expect the operating revenues of SF Holding in 2022 - 2024 to be 277.384 billion yuan, 300.074 billion yuan, and 340.005 billion yuan respectively, and the net profits attributable to the parent company to be 6.183 billion yuan, 9.048 billion yuan, and 12.130 billion yuan respectively. The corresponding PEs are 46.61 times, 31.85 times, and 23.76 times respectively. SF Holding is the leading enterprise in direct-operated comprehensive logistics. In 2023, the company's operating environment will be optimized, and growth is expected to accelerate. Maintain the 'Buy' rating."

[0087] "Company B: We expect the company's revenues in 2022 - 2024 to be 8.38 / 10.25 / 12.39 billion yuan respectively, and the net profits attributable to the parent company to be 1.36 / 1.76 / 2.21 billion yuan respectively. The corresponding EPS in 2022 - 2024 are 3.4 / 4.4 / 5.5 yuan per share respectively. Referring to the valuation of comparable companies in 2023 at 37x, considering that the company is in the leading position in the domestic energy drink industry and the current valuation has fully reflected the future growth potential, give the company a PE of 40x in 2023, corresponding to a target price of 176 yuan. The first coverage gives a 'Neutral' rating."

[0088] In this embodiment, through the powerful language understanding ability of the large language model, the original consistent expectation indicators can be extracted from the key information of the consistent expectation. Here, it is necessary to ensure that the time and the expected values correspond. For example, assuming that the identified key information of the consistent expectation is as shown above, the extracted original consistent expectation indicators can be as shown in Table 1 and Table 2 below:

[0089]

[0090] Table 1

[0091]

[0092] Table 2

[0093] In this embodiment, by constructing a corpus database to support a large-scale language model for consistent expectation information recognition, the accuracy of the large-scale language model in extracting consistent expectation metrics can be improved.

[0094] Figure 2 The structural block diagram of an industry prosperity determination device according to an embodiment of the present disclosure is shown. Among them, the device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. As Figure 3 shown, the industry prosperity determination device includes:

[0095] A text acquisition module 201, configured to acquire industry analysis text of a target object;

[0096] An extraction module 202, configured to input the industry analysis text into a pre-trained large-scale language model, execute the large-scale language model, and obtain industry consistent expectation metrics output by the large-scale language model;

[0097] An index acquisition module 203, configured to acquire industry profitability metrics, industry growth ability metrics, and industry debt repayment ability metrics;

[0098] A backup value determination module 204, configured to determine a backup value of industry prosperity according to the industry consistent expectation metrics, industry profitability metrics, industry growth ability metrics, and industry debt repayment ability metrics;

[0099] A correction value determination module 205, configured to acquire a capital flow trend index within the industry and determine a correction value according to the capital flow trend index;

[0100] A correction module 206, configured to correct the backup value of industry prosperity based on the correction value to determine the final industry prosperity.

[0101] In a possible implementation manner, the device further includes:

[0102] A target object determination module, configured to acquire historical industry analysis text of each object; extract historical market data proposed by each object from the historical industry analysis text of each object; acquire true market data corresponding to the historical market data proposed by each object; and determine a target object from the objects by comparing the historical market data proposed by each object and its corresponding true market data.

[0103] In a possible implementation manner, the part in the target object determination module that determines a target object from the objects by comparing the historical market data proposed by each object and its corresponding true market data is configured to:

[0104] Compare the historical market data proposed by each of the objects with the corresponding true market data, and determine the differences between the historical market data proposed by each object and the corresponding true market data;

[0105] For each object, determine the analysis accuracy of the object according to the corresponding difference situation of the object;

[0106] Determine the analysis independence of the object according to the difference situation corresponding to the object and the difference situations corresponding to other objects;

[0107] Determine the target object from each object according to the analysis accuracy and analysis independence of each object.

[0108] In a possible implementation manner, the extraction module is configured to:

[0109] Input the industry analysis text into the pre-trained large language model, and extract the original consensus expectation indicators from the industry analysis text through the large language model;

[0110] Obtain sentiment text from the industry analysis text through the large language model, where the distance between the sentiment text and the original consensus expectation indicators is within a certain range;

[0111] Determine the fine-tuning parameters according to the sentiment text;

[0112] Fine-tune the original consensus expectation indicators based on the fine-tuning parameters to obtain the industry consensus expectation indicators output by the large language model.

[0113] In a possible implementation manner, the device further includes:

[0114] A database acquisition module, configured to acquire a preset corpus database, where the corpus database stores corpus vectors conforming to the consensus expectation characteristics;

[0115] The part in the extraction module that inputs the industry analysis text into the pre-trained large language model and extracts the original consensus expectation indicators from the industry analysis text through the large language model is configured to:

[0116] Input the industry analysis text into a large language model based on retrieval-augmented generation, and through the large language model based on retrieval-augmented generation, perform similarity search in the industry analysis text based on the corpus vectors in the corpus database to obtain consensus expectation key information;

[0117] Extract the original consensus expectation indicators from the consensus expectation key information through the large language model.

[0118] The technical terms and technical features mentioned in the embodiments of the present device are the same as or similar to those mentioned in the above method embodiments. For the explanations and descriptions of the technical terms and technical features involved in the present device, reference can be made to the explanations and descriptions of the above method embodiments, which will not be elaborated here.

[0119] The present disclosure also discloses an electronic device, Figure 3 showing a structural block diagram of an electronic device according to an embodiment of the present disclosure.

[0120] As Figure 3 shown, the electronic device 300 includes a memory 301 and a processor 302. Among them, the memory 301 is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor 302 to implement the method according to the embodiment of the present disclosure.

[0121] Figure 4 showing a structural schematic diagram of a computer system suitable for implementing the method according to an embodiment of the present disclosure.

[0122] As Figure 4 shown, the computer system 400 includes a processing unit 401, which can execute various processes in the above embodiments according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage section 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the computer system 400 are also stored. The processing unit 401, the ROM 402, and the RAM 403 are connected to each other through a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0123] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read from it can be installed into the storage section 408 as needed. Among them, the processing unit 401 can be implemented as a processing unit such as a CPU, a GPU, a TPU, an FPGA, an NPU, etc.

[0124] In particular, according to an embodiment of the present disclosure, the method described above can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product that includes computer instructions that, when executed by a processor, implement the method steps described above. In such an embodiment, the computer program product can be downloaded and installed from a network via the communication section 409, and / or installed from the removable medium 411.

[0125] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0126] The units or modules described in the embodiments of the present disclosure can be implemented in software or by programmable hardware. The described units or modules can also be provided in a processor, and the names of these units or modules do not, in some cases, constitute a limitation on the units or modules themselves.

[0127] On the other hand, the present disclosure also provides a computer-readable storage medium, which can be the computer-readable storage medium included in the electronic device or computer system in the above embodiments; or it can exist separately and be a computer-readable storage medium not assembled into the device. The computer-readable storage medium stores one or more programs, and the one or more programs are used by one or more processors to execute the method described in the present disclosure.

[0128] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, technical solutions formed by mutually replacing the above features with technical features having similar functions (but not limited to) disclosed in the present disclosure.

Claims

1. A method for determining the prosperity degree of an industry, characterized in that, Including: Obtain the industry analysis text of the target object; Input the industry analysis text into a pre-trained large language model, execute the large language model, and obtain the industry consensus expectation index output by the large language model; Obtain industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators; Determine the standby value of industry prosperity according to the industry consensus expectation index, industry profitability indicators, industry growth ability indicators, and industry debt repayment ability indicators; Obtain the capital flow direction indicator within the industry, and determine the correction value according to the capital flow direction indicator; Based on the correction value, correct the standby value of industry prosperity to determine the final industry prosperity.

2. The method according to claim 1, wherein The method further includes: Obtain the historical industry analysis text of each object; Extract the historical market data proposed by each object from the historical industry analysis text of each object; Obtain the real market data corresponding to the historical market data proposed by each object; Determine the target object from each object by comparing the historical market data proposed by each object and its corresponding real market data.

3. The method according to claim 2, wherein The determining the target object from each object by comparing the historical market data proposed by each object and its corresponding real market data includes: Compare the historical market data proposed by each object and its corresponding real market data to determine the difference between the historical market data proposed by each object and its corresponding real market data; For each object, determine the analysis accuracy of the object according to the difference corresponding to the object; Determine the analysis independence of the object according to the difference corresponding to the object and the differences corresponding to other objects; Determine the target object from each object according to the analysis accuracy and analysis independence of each object.

4. The method according to claim 1, characterized in that, The inputting the industry analysis text into a pre-trained large language model, executing the large language model, and obtaining the industry consensus expectation index output by the large language model includes: Input the industry analysis text into the pre-trained large language model to extract the original consensus expectation index from the industry analysis text by the large language model; Obtain the sentiment text within a certain range of distance from the original consensus expectation index from the industry analysis text by the large language model; Determine the fine-tuning parameter according to the sentiment text; Based on the fine-tuning parameter, fine-tune the original consensus expectation index to obtain the industry consensus expectation index output by the large language model.

5. The method according to claim 4, wherein The method further includes: Obtain a preset corpus database, and the corpus database stores corpus vectors conforming to the consensus expectation characteristics; The inputting the industry analysis text into the pre-trained large language model to extract the original consensus expectation index from the industry analysis text by the large language model includes: Input the industry analysis text into a large-scale language model based on retrieval-augmented generation. Through the large-scale language model based on retrieval-augmented generation, perform similarity search in the industry analysis text based on the corpus vectors in the corpus database to obtain the consensus expected key information; Through the large-scale language model, extract the original consensus expected metrics from the consensus expected key information.

6. An industry prosperity determination device, characterized in that, It includes: A text acquisition module configured to acquire the industry analysis text of the target object; An extraction module configured to input the industry analysis text into a pre-trained large-scale language model, execute the large-scale language model, and obtain the industry consensus expected metrics output by the large-scale language model; A metric acquisition module configured to acquire industry profitability metrics, industry growth ability metrics, and industry debt repayment ability metrics; A standby value determination module configured to determine the standby value of industry prosperity according to the industry consensus expected metrics, industry profitability metrics, industry growth ability metrics, and industry debt repayment ability metrics; A correction value determination module configured to acquire the capital flow trend metrics within the industry and determine the correction value according to the capital flow trend metrics; A correction module configured to correct the standby value of industry prosperity based on the correction value to determine the final industry prosperity.

7. The device according to claim 6, characterized in that, The device further includes: A target object determination module configured to acquire the historical industry analysis text of each object; extract the historical market data proposed by each object from the historical industry analysis text of each object; acquire the real market data corresponding to the historical market data proposed by each object; determine the target object from each object by comparing the historical market data proposed by each object and its corresponding real market data.

8. The device according to claim 7, characterized in that The part in the target object determination module that determines the target object from each object by comparing the historical market data proposed by each object and its corresponding real market data is configured to: Compare the historical market data proposed by each object and its corresponding real market data to determine the difference situation between the historical market data proposed by each object and its corresponding real market data; For each object, determine the analysis accuracy of the object according to the difference situation corresponding to the object; Determine the analysis independence of the object according to the difference situation corresponding to the object and the difference situations corresponding to other objects; Determine the target object from each object according to the analysis accuracy and analysis independence of each object.

9. The device according to claim 6, characterized in that The extraction module is configured to: Input the industry analysis text into the pre-trained large-scale language model, and extract the original consensus expected metrics from the industry analysis text through the large-scale language model; Obtain the sentiment text within a certain range of distance from the original consensus expected metrics from the industry analysis text through the large-scale language model; Determine the fine-tuning parameters according to the sentiment text; Fine-tune the original consensus expected metrics based on the fine-tuning parameters to obtain the industry consensus expected metrics output by the large-scale language model.

10. The device according to claim 9, characterized in that, The device further includes: A database acquisition module, configured to acquire a preset corpus database, where the corpus database stores corpus vectors that conform to consistent expected features; In the extraction module, the industry analysis text is input into the pre-trained large language model, and the part of the original consistent expected indicators extracted from the industry analysis text by the large language model is configured to: Input the industry analysis text into the large language model based on retrieval-augmented generation. Through the large language model based on retrieval-augmented generation, perform similarity search in the industry analysis text based on the corpus vectors in the corpus database to obtain consistent expected key information; Extract the original consistent expected indicators from the consistent expected key information through the large language model.

11. An electronic device, comprising a memory and a processor; wherein, The memory is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the method steps described in any one of claims 1 to 5.

12. A computer-readable storage medium having computer instructions stored thereon, wherein, When the computer instructions are executed by the processor, the method described in any one of claims 1-5 is implemented.

13. A computer program product, including computer instructions, which implement the method steps described in any one of claims 1 to 5 when executed by a processor.

Images