CN114996530A - A method and device for analyzing data of company research report based on knowledge distillation - Google Patents

Abstract

The invention discloses a method and a device for analyzing company research and report data based on knowledge distillation. Wherein, the method comprises the following steps: acquiring first business data; subdividing the first business data to obtain Lagrange target operators with preset level number; converting the second business data according to the Lagrange target operator to obtain report data; and inputting the newspaper data into an input layer of a newspaper analysis model to obtain a newspaper analysis result. The invention solves the technical problem that in the prior art, in company research and report data extraction, the business data in a company report is only directly used for corresponding analysis of a multivariate matrix or a binary matrix, and the analysis result is output as a data rational output point, so that the research and report data cannot be generated through different business data and diversified influence factors, and the research and report analysis result is finally generated.

Description

A method and device for analyzing data of company research report based on knowledge distillation

technical field

The invention relates to the field of intelligent extraction of company data, in particular to a method and device for analyzing data of company research reports based on knowledge distillation.

Background technique

With the continuous development of intelligent technology, more and more intelligent equipment is used in people's life, work and study. The use of intelligent technology has improved the quality of people's life and increased the efficiency of people's study and work.

At present, when major companies or platforms conduct data extraction and data statistics, they usually use various types of data in the company's research report to conduct research report analysis, and use the analysis results as the company's output report for further decision analysis. However, in the data extraction of the company's research report in the prior art, the business data in the company's report is directly used for the response analysis of the multivariate matrix or the binary matrix, and the analysis result is output as the data output point. It is impossible to generate research report data through different business data and diversified impact factors, so as to finally generate research report analysis results.

For the above problems, no effective solution has been proposed yet.

SUMMARY OF THE INVENTION

The embodiments of the present invention provide a method and device for analyzing data of a company research report based on knowledge distillation, so as to at least solve the problem that in the data extraction of a company research report in the prior art, only the business data in the company report is directly used in a multivariate matrix Or the response analysis of the binary matrix, and the analysis results are output as the data should be output. It is impossible to generate the research report data through different business data and diversified influence factors, so as to finally generate the technical problem of the research report analysis results. .

According to an aspect of the embodiments of the present invention, a method for analyzing data of a company research report based on knowledge distillation is provided, including: obtaining first business data; Grange target operator; convert the second business data according to the Lagrangian target operator, and obtain research report data; input the research report data into the input layer of the research report analysis model to obtain the research report Analyze the results.

Optionally, before acquiring the first business data, the method further includes: generating the first business data according to historical research report data and research report analysis requirements.

Optionally, the step of subdividing the first business data to obtain a Lagrangian target operator with a preset number of levels includes: splitting the first business data into vector segments, and using the first business data The volume of the business data calculates the split interval value, and obtains the split business data; substitute the split business data into the Lagrangian inverse function to obtain the Lagrangian of the preset number of levels a target operator, used to perform Lagrangian conversion on the second business data.

Optionally, before the research report data is input into the input layer of the research report analysis model and the research report analysis result is obtained, the method further includes: training the research report analysis model through historical data, and applying the model to the input layer. Perform a correction operation.

According to another aspect of the embodiments of the present invention, a device for analyzing data of company research reports based on knowledge distillation is also provided, including: an acquisition module for acquiring first business data; a subdivision module for The business data is subdivided to obtain a preset number of Lagrangian target operators; the conversion module is used to convert the second business data according to the Lagrangian target operators, and obtain research report data; input The module is used to input the research report data into the input layer of the research report analysis model to obtain the research report analysis result.

Optionally, the device further includes: a generating module, configured to generate the first business data according to the historical data of the research report and the analysis requirements of the research report.

Optionally, the subdivision module includes: a splitting unit, configured to split the first business data into vector segments, calculate the value of the split interval based on the volume of the first business data, and obtain the split interval value. The divided business data; the computing unit is used for substituting the divided business data into the inverse Lagrangian function to obtain the Lagrangian target operator of the preset number of levels, which is used for the calculation of the first 2. Lagrangian conversion of operating data.

Optionally, the device further includes: a training correction module, configured to train the research report analysis model through historical data, and perform a correction operation on the model.

According to another aspect of the embodiments of the present invention, a non-volatile storage medium is further provided, and the non-volatile storage medium includes a stored program, wherein the program controls the location of the non-volatile storage medium when running. The device implements a knowledge distillation-based method for analyzing company research report data.

According to another aspect of the embodiments of the present invention, an electronic device is also provided, including a processor and a memory; the memory stores computer-readable instructions, and the processor is configured to execute the computer-readable instructions, wherein, The computer-readable instruction executes a method for analyzing data of a company's research report based on knowledge distillation.

In the embodiment of the present invention, the first business data is acquired; the first business data is subdivided to obtain a Lagrangian target operator with a preset number of levels; according to the Lagrangian target operator, the The second business data is converted, and the research report data is obtained; the method of inputting the research report data into the input layer of the research report analysis model to obtain the research report analysis result solves the problem of extracting the company's research report data in the prior art. , just use the business data in the company's report to directly use the multivariate matrix or binary matrix for coping analysis, and output the analysis results as the data output points. The generation of the report data, so as to finally generate the technical problems of the analysis results of the research report.

Description of drawings

The accompanying drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the attached image:

Fig. 1 is the flow chart of a kind of company research report data analysis method based on knowledge distillation according to an embodiment of the present invention;

FIG. 2 is a structural block diagram of an apparatus for analyzing data of a company research report based on knowledge distillation according to an embodiment of the present invention.

Detailed ways

In order to make those skilled in the art better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only Embodiments are part of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms "first", "second" and the like in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used may be interchanged under appropriate circumstances such that the embodiments of the invention described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having" and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those expressly listed Rather, those steps or units may include other steps or units not expressly listed or inherent to these processes, methods, products or devices.

According to an embodiment of the present invention, a method embodiment of a method for analyzing data of a company's research report based on knowledge distillation is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a set of computer executable instructions, for example. is performed in a computer system and, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that herein.

Example 1

Fig. 1 is a flow chart of a method for analyzing data of a company's research report based on knowledge distillation according to an embodiment of the present invention. As shown in Fig. 1 , the method comprises the following steps:

Step S102, acquiring first business data.

Optionally, before acquiring the first business data, the method further includes: generating the first business data according to historical research report data and research report analysis requirements.

Specifically, the embodiment of the present invention analyzes and extracts the company's research report data based on the knowledge distillation technology, and uses the analysis result for the company's strategy deployment, which solves the problem of the company's research report data extraction in the prior art. It is to directly use the business data in the company's report for the response analysis of the multivariate matrix or binary matrix, and output the analysis results as the data output point. The technical problem of generating the analysis results of the research report and finally generating the analysis results of the research report requires first obtaining the first business data of the company. Obtaining the first business data is to generate the first business data according to the historical data of the research report and the analysis requirements of the research report, thereby The first business data has the analysis characteristics of historical research report data and integrates the research report analysis needs, and accurately collects and obtains the first business data according to the company's research report analysis needs.

Step S104: Subdivide the first business data to obtain Lagrangian target operators with a preset number of levels.

Optionally, the step of subdividing the first business data to obtain a Lagrangian target operator with a preset number of levels includes: splitting the first business data into vector segments, and using the first business data The volume of the business data calculates the split interval value, and obtains the split business data; substitute the split business data into the Lagrangian inverse function to obtain the Lagrangian of the preset number of levels a target operator, used to perform Lagrangian conversion on the second business data.

Specifically, in order to subdivide the first business data obtained in the above embodiments of the present invention, so as to accurately generate the analysis results of the research report data in the future, the first business data needs to be divided into vector segments, so as to The volume of the first business data calculates the value of the split interval, and obtains the split business data; and substitutes the split business data into the Lagrangian inverse function to obtain the pull of the preset number of levels. A Grange target operator, used to perform Lagrangian conversion on the second business data. Among them, the first business data is classified by the Lagrangian operator, and the n-^1n matrix is generated according to the Lagrangian N column coefficients of the power to obtain the final Lagrangian operator, and The obtained target operator is used to convert the second business data.

Step S106, convert the second business data according to the Lagrangian target operator, and obtain research report data.

Specifically, in order to obtain the research report data, it is necessary to use the latest local business data of the company as the second business data, and use the above Lagrangian target operator, that is, the parameter conversion formula of the first business data generated according to the historical research report data. As the basis for generating the second business data, to generate accurate second business data, and then according to the convolution of the second business data and the Lagrangian operator, the partial derivative generates the research report data.

Step S108, the research report data is input into the input layer of the research report analysis model to obtain the research report analysis result.

Optionally, before the research report data is input into the input layer of the research report analysis model and the research report analysis result is obtained, the method further includes: training the research report analysis model through historical data, and applying the model to the input layer. Perform a correction operation.

Specifically, after training the research report analysis model through historical data and correcting the model, it is equivalent to training a mature research report analysis model, and the research report analysis model needs to be used to input the research report data into the research report. The input layer of the report analysis model is used to obtain the analysis results of the research report.

Through the above-mentioned embodiments, in the data extraction of the company's research report in the prior art, the business data in the company's report is only directly used for the response analysis of the multivariate matrix or the binary matrix, and the analysis result is used as the data output point. For output, it is impossible to generate research report data through different business data and diversified impact factors, so as to finally generate the technical problem of research report analysis results.

Embodiment 2

Fig. 2 is a structural block diagram of a company research report data analysis device based on knowledge distillation according to an embodiment of the present invention. As shown in Fig. 2, the device includes:

The obtaining module 20 is used for obtaining the first business data.

Optionally, the device further includes: a generating module, configured to generate the first business data according to the historical data of the research report and the analysis requirements of the research report.

Specifically, the embodiment of the present invention analyzes and extracts the company's research report data based on the knowledge distillation technology, and uses the analysis result for the company's strategy deployment, which solves the problem of the company's research report data extraction in the prior art. It is to directly use the business data in the company's report for the response analysis of the multivariate matrix or binary matrix, and output the analysis results as the data output point. The technical problem of generating the analysis results of the research report and finally generating the analysis results of the research report requires first obtaining the first business data of the company. Obtaining the first business data is to generate the first business data according to the historical data of the research report and the analysis requirements of the research report, thereby The first business data has the analysis characteristics of historical research report data and integrates the research report analysis needs, and accurately collects and obtains the first business data according to the company's research report analysis needs.

The subdivision module 22 is configured to subdivide the first business data to obtain a Lagrangian target operator with a preset number of levels.

Optionally, the subdivision module includes: a splitting unit, configured to split the first business data into vector segments, calculate the value of the split interval based on the volume of the first business data, and obtain the split interval value. The divided business data; the computing unit is used for substituting the divided business data into the inverse Lagrangian function to obtain the Lagrangian target operator of the preset number of levels, which is used for the calculation of the first 2. Lagrangian conversion of operating data.

Specifically, in order to subdivide the first business data obtained in the above embodiments of the present invention, so as to accurately generate the analysis results of the research report data in the future, the first business data needs to be divided into vector segments, so as to The volume of the first business data calculates the value of the split interval, and obtains the split business data; and substitutes the split business data into the Lagrangian inverse function to obtain the pull of the preset number of levels. A Grange target operator, used to perform Lagrangian conversion on the second business data. Among them, the first business data is classified by the Lagrangian operator, and the n-^1n matrix is generated according to the Lagrangian N column coefficients of the power to obtain the final Lagrangian operator, and The obtained target operator is used to convert the second business data.

The conversion module 24 is configured to convert the second business data according to the Lagrangian target operator, and obtain research report data.

Specifically, in order to obtain the research report data, it is necessary to use the latest local business data of the company as the second business data, and use the above Lagrangian target operator, that is, the parameter conversion formula of the first business data generated according to the historical research report data. As the basis for generating the second business data, to generate accurate second business data, and then according to the convolution of the second business data and the Lagrangian operator, the partial derivative generates the research report data.

The input module 26 is used for inputting the research report data into the input layer of the research report analysis model to obtain the research report analysis result.

Optionally, the device further includes: a training correction module, configured to train the research report analysis model through historical data, and perform a correction operation on the model.

Specifically, after training the research report analysis model through historical data and correcting the model, it is equivalent to training a mature research report analysis model, and the research report analysis model needs to be used to input the research report data into the research report. The input layer of the report analysis model is used to obtain the analysis results of the research report.

According to another aspect of the embodiments of the present invention, a non-volatile storage medium is further provided, and the non-volatile storage medium includes a stored program, wherein the program controls the location of the non-volatile storage medium when running. The device implements a knowledge distillation-based method for analyzing company research report data.

According to another aspect of the embodiments of the present invention, an electronic device is also provided, including a processor and a memory; the memory stores computer-readable instructions, and the processor is configured to execute the computer-readable instructions, wherein, The computer-readable instruction executes a method for analyzing data of a company's research report based on knowledge distillation.

Through the above-mentioned embodiments, in the data extraction of the company's research report in the prior art, the business data in the company's report is only directly used for the response analysis of the multivariate matrix or the binary matrix, and the analysis result is used as the data output point. For output, it is impossible to generate research report data through different business data and diversified impact factors, so as to finally generate the technical problem of research report analysis results.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

In the above-mentioned embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are only illustrative, for example, the division of the units may be a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components may be combined or Integration into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of units or modules, and may be in electrical or other forms.

The units described as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other various storage media that can store program codes. medium.

The above are only the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (10)

1. a company research report data analysis method based on knowledge distillation, is characterized in that, comprises: Obtain the first business data; Subdividing the first business data to obtain a Lagrangian target operator with a preset number of levels; Convert the second business data according to the Lagrangian target operator, and obtain research report data; The research report data is input into the input layer of the research report analysis model to obtain the research report analysis result. 2 . The method according to claim 1 , wherein before acquiring the first business data, the method further comprises: 2 . The first business data is generated according to the historical data of the research report and the analysis requirements of the research report. 3. The method according to claim 1, wherein the step of subdividing the first business data to obtain a Lagrangian target operator with a preset number of levels comprises: Splitting the first business data into vector segments, calculating the split interval value based on the volume of the first business data, and obtaining the split business data; Substitute the split business data into an inverse Lagrangian function to obtain the Lagrangian target operator of the preset number of levels, which is used to perform Lagrangian conversion on the second business data. 4. The method according to claim 1, characterized in that, before the research report data is input into the input layer of the research report analysis model, and before the research report analysis result is obtained, the method further comprises: The research report analysis model is trained through historical data, and the model is corrected. 5. a company research report data analysis device based on knowledge distillation, is characterized in that, comprises: an acquisition module for acquiring the first business data; a subdivision module, configured to subdivide the first business data to obtain a Lagrangian target operator with a preset number of levels; a conversion module, configured to convert the second business data according to the Lagrangian target operator, and obtain research report data; The input module is used for inputting the research report data into the input layer of the research report analysis model to obtain the research report analysis result. 6. The apparatus according to claim 5, wherein the apparatus further comprises: The generating module is configured to generate the first business data according to the historical data of the research report and the analysis requirements of the research report. 7. The device according to claim 5, wherein the subdivision module comprises: a splitting unit, configured to split the first business data into vector segments, calculate the value of the split interval based on the volume of the first business data, and obtain the split business data; a calculation unit, configured to substitute the split business data into an inverse Lagrangian function to obtain a Lagrangian target operator of the preset number of levels, which is used to perform a Lagrangian operation on the second business data Long day conversion. 8. The apparatus according to claim 5, wherein the apparatus further comprises: A training correction module is used to train the research report analysis model through historical data, and perform a correction operation on the model. 9 . A non-volatile storage medium, characterized in that the non-volatile storage medium comprises a stored program, wherein, when the program is run, the program controls a device where the non-volatile storage medium is located to execute claims 1 to 10 . The method of any one of 4. 10. An electronic device, comprising a processor and a memory; wherein computer-readable instructions are stored in the memory, and the processor is configured to execute the computer-readable instructions, wherein the computer-readable instructions run while performing the method of any one of claims 1 to 4.

Images