Report generation method and device, equipment, storage medium and program product

Abstract

The invention provides a report generation method and device, equipment, a storage medium and a program product, relates to the field of computer application, can efficiently analyze enterprises, and can improve the accuracy of analysis results. According to the specific technical scheme, a first model is trained based on analysis report samples of multiple enterprises in a first industry, and a first analysis entry point of the first industry is obtained; correcting the first analysis entry point based on a second analysis entry point of the first industry determined by a preset user to obtain a target analysis entry point; inputting the enterprise information of the to-be-analyzed enterprise, the target analysis entry point and the analysis configuration information into a second model, and operating the second model to output an analysis conclusion of the to-be-analyzed enterprise in each analysis key point and a fact basis used by the analysis conclusion; and based on a serial verbal skill, integrating the plurality of analysis conclusions corresponding to the plurality of analysis key points and the plurality of facts to obtain an analysis report of the to-be-analyzed enterprise. The scheme can be used for evaluating and analyzing enterprises.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of computer application, and particularly, to a report generation method and device, equipment, storage medium and program product. BACKGROUND

[0002] Before providing corresponding services for an enterprise, the enterprise is analyzed, and then a matching service is provided based on the analysis result, which can improve the accuracy of the service. In the prior art, a large amount of enterprise information is usually collected manually and then analyzed.

[0003] The prior art has at least the following defects: first, the efficiency is low, and the analysis result cannot be obtained quickly, so that the corresponding service cannot be provided for the enterprise in time; second, due to the limitation of manual analysis, the accuracy of the analysis result is usually affected, so that the accurate service cannot be provided for the enterprise. SUMMARY

[0004] Embodiments of the present application provide a report generation method, device, equipment, storage medium and program product, which can efficiently analyze the enterprise compared with the manual analysis method, and can improve the accuracy of the analysis result, so that the corresponding service can be provided for the enterprise in time and accurately.

[0005] To achieve the above object, embodiments of the present application adopt the following technical solutions: In a first aspect, a report generation method is provided, which includes: training a first model based on analysis report samples of a plurality of enterprises in a first industry to obtain a first analysis entry point of the first industry; wherein the first model has the ability to summarize the analysis entry point. Based on a second analysis entry point of the first industry determined by a preset user, the first analysis entry point is corrected to obtain a target analysis entry point, which is used for data analysis of the first industry. The enterprise information of the enterprise to be analyzed, the target analysis entry point and the analysis configuration information are input into a second model, and the second model is run to output the analysis conclusion of the enterprise to be analyzed at each analysis point and the fact basis for using the analysis conclusion; wherein the second model has the ability to analyze and generate the analysis conclusion and the fact basis, the enterprise to be analyzed belongs to the first industry, one target analysis entry point corresponds to one or more analysis points, and the analysis configuration information indicates the analysis strategy. Based on the serial dialogue, a plurality of analysis conclusions corresponding to a plurality of analysis points and a plurality of fact bases are integrated to obtain an analysis report of the enterprise to be analyzed.

[0006] Therefore, by using the present application, the preliminary first analysis entry point can be obtained through training of the analysis report sample, and then the target analysis entry point is obtained by combining the second analysis entry point given by the expert, so that the analysis entry point in the analysis report sample and the experience of the expert can be considered, and the rationality of the obtained analysis entry point is improved. Subsequently, the analysis conclusion and factual basis of the analysis points corresponding to the target analysis entry point are obtained by analyzing the enterprise to be analyzed by the second model, which can not only be intelligent analysis, but also obtain refined analysis conclusion and factual basis. Finally, the analysis conclusion and factual basis are integrated to form an analysis report, so that the analysis report is obtained. In this way, compared with the manual analysis mode, the analysis efficiency can be greatly improved, and the limitations of manual analysis can be avoided, and the accuracy of the analysis result is improved.

[0007] In a possible implementation of the first aspect, before the first model is trained based on the analysis report samples of the plurality of enterprises in the first industry to obtain the first analysis entry point of the first industry, the method further includes: constructing a knowledge base of the first industry, the knowledge base including the analysis report samples, and the knowledge base further including at least one of the following: industry research reports, analysis system information, and annual reports of listed companies; and training the first model based on the analysis report samples of the plurality of enterprises in the first industry includes training the first model based on the knowledge base.

[0008] By using the present implementation, the first model can be trained based on the professional knowledge of the first industry, so that the first model can summarize the first analysis entry point from the professional knowledge, thereby improving the accuracy of the obtained first analysis entry point.

[0009] In a possible implementation of the first aspect, before the analysis report of the enterprise to be analyzed is obtained by integrating the plurality of analysis conclusions corresponding to the plurality of analysis points and the plurality of factual bases based on the string connection rhetoric, the method further includes: generating the string connection rhetoric based on the knowledge base.

[0010] By using the present implementation, the string connection rhetoric is summarized from the professional knowledge of the first industry, so that the obtained string connection rhetoric is more professional, and a more professional description of the first industry can be generated in the subsequent analysis report.

[0011] In a possible implementation of the first aspect, after the analysis report of the enterprise to be analyzed is obtained, the method further includes: in a case where the analysis report meets a preset condition, adding the analysis report to the knowledge base and retraining the first model to obtain an updated first analysis entry point.

[0012] By using the present design, the knowledge base can be continuously updated, and the first model can be retrained, so that a new first analysis entry point can be obtained, and the first analysis entry point is more accurate.

[0013] In a possible implementation manner of the first aspect, the analysis configuration information comprises at least one of the following: a first correspondence relationship between each target analysis entry point and a plurality of analysis titles, and a second correspondence relationship between each analysis title and a plurality of analysis points; data sources required by each analysis title; and analysis rules of one or more analysis points.

[0014] By using the design manner, the correspondence relationship between the analysis entry point and the analysis point can be indicated by analyzing the analysis configuration information, and the data used for analysis and the analysis rules can be indicated, so that the analysis conclusion meeting the analysis rules can be obtained by performing analysis on the refined analysis point and using the corresponding data.

[0015] In a possible implementation manner of the first aspect, the running the second model outputs the analysis conclusion of the enterprise to be analyzed at each analysis point and the factual basis used by the analysis conclusion, comprising: running the second model, for each target analysis entry point, determining a plurality of target analysis titles corresponding to the target analysis entry point according to the first correspondence relationship; for any target analysis title, determining a plurality of target analysis points corresponding to the target analysis title according to the second correspondence relationship, and determining target source data corresponding to the target analysis title according to the data sources required by each analysis title; for any target analysis point, determining whether the target analysis point has a corresponding target analysis rule; if the target analysis point has the target analysis rule, performing analysis on the target source data corresponding to the target analysis title to which the target analysis point belongs according to the target analysis rule, to obtain the analysis conclusion of the target analysis point and the factual basis used by the analysis conclusion; if the target analysis point does not have the target analysis rule, performing analysis on the target source data corresponding to the target analysis title to which the target analysis point belongs, to obtain the analysis conclusion of the target analysis point and the factual basis used by the analysis conclusion.

[0016] In a possible implementation manner of the first aspect, the target analysis entry point comprises a main business market positioning; the target analysis title corresponding to the main business market positioning comprises at least two of the following: main business and market positioning description; project undertaking ability, development ability and management ability description; and market regional share and industry status.

[0017] In a second aspect, a report generation apparatus is provided, which comprises: a training module configured to train a first model based on analysis report samples of a plurality of enterprises in a first industry to obtain a first analysis entry point of the first industry; wherein the first model has the ability to summarize the analysis entry point; a correction module configured to correct the first analysis entry point based on a second analysis entry point of the first industry determined by a preset user to obtain a target analysis entry point, which is used for data analysis of the first industry; an analysis module configured to input enterprise information of a to-be-analyzed enterprise, the target analysis entry point, and analysis configuration information into a second model, and run the second model to output analysis conclusions of the to-be-analyzed enterprise at each analysis point and factual basis used by the analysis conclusions; wherein the second model has the ability to analyze and generate the analysis conclusions and the factual basis, the to-be-analyzed enterprise belongs to the first industry, one target analysis entry point corresponds to one or more analysis points, and the analysis configuration information indicates an analysis strategy; and an integration module configured to integrate a plurality of analysis conclusions and a plurality of factual bases corresponding to the plurality of analysis points based on a stringing speech to obtain an analysis report of the to-be-analyzed enterprise.

[0018] In a third aspect, an electronic device is provided, which comprises: a memory and at least one processor. The memory is in communication connection with the processor. The memory is configured to store computer program code, which comprises computer instructions. When the processor executes the computer instructions, the electronic device performs the method according to the first aspect and any possible implementation manner thereof.

[0019] In a fourth aspect, a computer readable storage medium is provided, which stores computer instructions. When the computer instructions are executed by a processor, the method according to the first aspect and any possible implementation manner thereof is implemented.

[0020] In a fifth aspect, a computer program product is provided, which, when running on a computer / being executed by a processor of a computer, implements the method according to the first aspect and any possible implementation manner thereof. The computer can be the electronic device according to the second aspect and any possible implementation manner thereof.

[0021] It can be understood that the report generation apparatus according to the second aspect, the electronic device according to the third aspect, the computer readable storage medium according to the fourth aspect, and the computer program product according to the fifth aspect can achieve the beneficial effects as described in the first aspect and any possible implementation manner thereof, which will not be described herein again. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1One of flowcharts of a report generation method provided by an embodiment of the present application; Figure 2 Another one of flowcharts of a report generation method provided by an embodiment of the present application; Figure 3 Still another one of flowcharts of a report generation method provided by an embodiment of the present application; Figure 4 A structural schematic diagram of a report generation apparatus provided by an embodiment of the present application; Figure 5 A structural schematic diagram of another report generation apparatus provided by an embodiment of the present application; Figure 6 A structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0023] Hereinafter, the terms "first" and "second" are only used for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more of the features. In the description of the embodiments, the meaning of "a plurality of" is two or more, unless otherwise specified.

[0024] The exemplary embodiments will be described in detail below with reference to the accompanying drawings. In the following description, the same numbers refer to the same or similar elements unless otherwise specified. The embodiments described in the following exemplary embodiments do not represent all the embodiments consistent with the present application. Rather, they are merely examples of devices and methods consistent with some aspects of the present application, as detailed in the appended claims.

[0025] In the technical solutions provided by the embodiments of the present application, the collection, storage, use, processing, transmission, provision and disclosure of information such as financial data or user data, etc. all comply with the relevant legal regulations and do not violate public order and good customs.

[0026] It should be noted that in the embodiments of the present application, some industry existing solutions, components, models, etc. may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0027] The embodiment of the present application provides a report generation method, which can be used in a scenario where enterprises need to be analyzed. A typical scenario is as follows: a certain platform needs to provide services for each enterprise, and accordingly, the platform can analyze each enterprise, and then provide accurate services to the enterprise based on the analysis result. Of course, there can be some platforms that specialize in enterprise analysis business, and these platforms can also analyze each enterprise. Hereinafter, the platform that analyzes the enterprise is collectively referred to as an analysis device.

[0028] By using the report generation method provided in the embodiment of the present application, the enterprise can be automatically analyzed in combination with big data and the experience of professionals, and compared with the manual analysis mode, the efficiency and accuracy of the analysis can be improved. Referring to Figure 1 The method comprises the following steps: S101, training a first model based on analysis report samples of a plurality of enterprises in a first industry to obtain a first analysis entry point of the first industry.

[0029] The first model has the ability to summarize the analysis entry point.

[0030] The plurality of enterprises can include enterprises of various scales in the first industry, such as large enterprises, medium-sized enterprises, and small enterprises; the plurality of enterprises can also include enterprises of various operating conditions, such as high-quality enterprises with good operating conditions, ordinary enterprises with normal operating conditions, and high-risk enterprises with abnormal operating conditions. Accordingly, the analysis report samples of the plurality of enterprises can cover the analysis report samples of various enterprises in the first industry.

[0031] The analysis device can use the analysis report samples of the plurality of enterprises to train the first model in a unsupervised training manner, so that the first model summarizes the analysis entry point in the analysis report samples to obtain the first analysis entry point of the first industry. The analysis entry point refers to the analysis dimension when analyzing the enterprise.

[0032] It should be noted that the analysis entry points used in different analysis report samples can be different. By training the first model, all analysis entry points involved in the plurality of analysis report samples or the analysis entry points with more usage times can be summarized to obtain the first analysis entry point.

[0033] The first analysis entry point obtained above is used as the initial analysis entry point of the first industry, and the final version of the analysis entry point can be further obtained by combining expert opinions, as shown in the following S102: S102, using a second analysis entry point of the first industry determined by a preset user to correct the first analysis entry point to obtain a target analysis entry point, and the target analysis entry point is used for data analysis of the first industry.

[0034] The preset user is an expert in the field of enterprise analysis. That is, the first analysis entry point given by the expert can be used to make up for the deficiency of the first analysis entry point summarized by the first model.

[0035] The first analysis entry point can be a union of the first analysis entry point and the second analysis entry point, or an intersection of the first analysis entry point and the second analysis entry point, and the like, which is not limited in the embodiments of the present application.

[0036] The target analysis entry point obtained by modifying the first analysis entry point by the analysis device can be used for subsequent enterprise analysis, which is more accurate because it takes into account the characteristics of the analysis report sample and the experience of the expert.

[0037] It should be noted that the target analysis entry point obtained by the above S101 and S102 can be used when analyzing each enterprise in the first industry. That is, S101 and S102 can be executed in advance to obtain a target analysis entry point applicable to the first industry, and the target analysis entry point can be used when analyzing each enterprise in the first industry. It can be seen that the above S101 and S102 do not need to be executed when analyzing each enterprise.

[0038] In some embodiments, the analysis device can execute S101 and S102 after obtaining a new analysis report sample or periodically, so that the first model can be updated based on the new analysis report sample, and a new first analysis entry point can be output, and then the new target analysis entry point can be obtained by modifying based on the second analysis entry point.

[0039] In another embodiment, the analysis device can execute S102 once after the expert gives a new second analysis entry point each time, so that the first analysis entry point can be modified based on the new second analysis entry point to obtain a new target analysis entry point.

[0040] S103, inputting the enterprise information of the enterprise to be analyzed, the target analysis entry point and the analysis configuration information into the second model, running the second model to output the analysis conclusion of each analysis point of the enterprise to be analyzed and the factual basis for using the analysis conclusion. The enterprise to be analyzed belongs to the first industry, and the analysis configuration information indicates the analysis strategy.

[0041] The enterprise information of the enterprise to be analyzed can be relevant data collected by the analysis device about the enterprise to be analyzed, which is used to analyze the enterprise to be analyzed. For example, the analysis device can analyze A enterprise, and can collect business data and operating status data of A enterprise.

[0042] One target analysis entry point corresponds to one or more analysis points, and usually multiple analysis points. That is, the analysis device adopts the second model, and for one target analysis entry point, can analyze to obtain analysis conclusions and factual basis of the corresponding one or more analysis points, thereby obtaining refined analysis results.

[0043] The analysis configuration information can include the correspondence between the target analysis entry point and the analysis point, the analysis rule, and the like, thereby indicating the analysis strategy of the analysis device.

[0044] For example, the analysis device can determine the analysis point to be analyzed based on the correspondence.

[0045] For another example, the analysis device can determine the specific data required for the analysis point to be analyzed based on the analysis rule. For example, the analysis point to be analyzed includes the operating conditions of the company in recent years, and accordingly, the analysis rule can indicate listing the data of the last three years by year.

[0046] The factual basis used by the analysis conclusion can be related content in the enterprise information of the enterprise to be analyzed, such as related paragraphs, sentences, charts, and the like.

[0047] The second model has the ability to analyze and generate the analysis conclusion and the factual basis. In some embodiments, the second model can be an existing artificial intelligence (AI) model. That is, the analysis device can access the existing AI model, use the analysis capability of the existing AI model, and analyze and generate the analysis conclusion and the factual basis. In this way, the analysis conclusion and the basis can be generated without training a new model.

[0048] In this embodiment, the existing AI model can also access the knowledge for analysis. For example, the knowledge accessed by the existing AI model includes industry data of the first industry, some public data of the enterprise to be analyzed, and the like. For example, the public data can be sourced from the official website of the enterprise to be analyzed, and data in some enterprise analysis website.

[0049] In this embodiment, the factual basis used by the analysis conclusion can also be related content in the knowledge accessed by the existing AI model.

[0050] Of course, the second model can be a model specially trained to analyze and generate the analysis conclusion and the factual basis, and the embodiments of the present application do not specifically limit this.

[0051] The second model can analyze the analysis conclusion and the factual basis through text analysis, semantic retrieval, natural language processing, and the like.

[0052] The analysis device can generate a plurality of analysis conclusions corresponding to the plurality of analysis points and a plurality of factual bases thereof by using S103. For example, the analysis results obtained by using S103 are shown in Table 1 below: Table 1

[0053] In Table 1 above, the target entry points include entry point 1 and entry point 2. Entry point 1 corresponds to three analysis points, i.e., point 1, point 2, and point 3. The analysis results obtained by using S103 are as follows: point 1 corresponds to conclusion 1, and the factual basis of conclusion 1 is basis 1; point 2 corresponds to conclusion 2, and the factual basis of conclusion 2 is basis 2; and point 3 corresponds to conclusion 3, and the factual basis of conclusion 3 is basis 3. Entry point 2 corresponds to two analysis points, i.e., point 4 and point 5. The analysis results obtained by using S103 are as follows: point 4 corresponds to conclusion 4, and the factual basis of conclusion 4 is basis 4; and point 5 corresponds to conclusion 5, and the factual basis of conclusion 5 is basis 5.

[0054] S104, based on the serial speech, integrates the plurality of analysis conclusions corresponding to the plurality of analysis points and the plurality of factual bases to obtain an analysis report of the enterprise to be analyzed.

[0055] The serial speech includes the speech between the analysis point and the analysis conclusion, and the speech between the analysis conclusion and the factual basis. In this way, the analysis device can connect the analysis point and the analysis conclusion, and connect the analysis conclusion and the factual basis by using the serial speech, to form an analysis report with fluent sentences.

[0056] For example, the serial speech between the analysis point and the analysis conclusion is “for the analysis point, the analysis conclusion obtained is as follows”. For another example, the serial speech between the analysis conclusion and the factual basis is “the analysis conclusion is obtained based on the following factual basis”.

[0057] Further, for different analysis points, the speech between the analysis point and the analysis conclusion can be different, and / or the speech between the analysis conclusion and the factual basis can be different. In this way, the analysis device can use the corresponding serial speech for different analysis points to make the connection of each analysis point more reasonable.

[0058] For example, the speech between point 1 and conclusion 1 in Table 1 above is different from the speech between point 2 and conclusion 2, and the speech between conclusion 1 and basis 1 is different from the speech between conclusion 2 and basis 2.

[0059] Therefore, by using the application, the analysis device can obtain a preliminary first analysis entry point through training on the analysis report sample, and then correct the target analysis entry point by combining the second analysis entry point given by the expert, so that the analysis entry point in the analysis report sample and the expert experience can be considered, and the rationality of the obtained analysis entry point is improved. Subsequently, the analysis device analyzes the enterprise to be analyzed through the second model to obtain the analysis conclusion and factual basis of the analysis points corresponding to the target analysis entry point, which can not only be intelligent analysis, but also obtain refined analysis conclusion and factual basis. Finally, the analysis device integrates the analysis conclusion and factual basis to form an analysis report, thereby obtaining the analysis report. In this way, compared with the manual analysis mode, the analysis efficiency can be greatly improved, and the limitations of manual analysis can be avoided, and the accuracy of the analysis result is improved.

[0060] In some embodiments, the analysis device can construct a knowledge base of the first industry for generating a preliminary first analysis entry point and a final analysis report. Therefore, a more accurate first analysis entry point and analysis report can be obtained. Specifically, see Figure 2 The report generation method includes the following steps: S201, constructing a knowledge base of the first industry, the knowledge base including analysis report samples of a plurality of enterprises.

[0061] The knowledge base can also include at least one of the following information of the first industry: industry research report, analysis system information, and listed company annual report. The embodiments of the application do not make specific limitations on this.

[0062] In a specific implementation, the analysis device can determine the information to be collected and construct the knowledge base based on which aspects of the customer need to be analyzed and what kind of analysis conclusion needs to be obtained.

[0063] The analysis device can collect the information of the first industry described above, and refine professional knowledge, define knowledge classification and structure from the information to construct the knowledge base of the first industry.

[0064] The analysis device can also periodically process the professional knowledge in the knowledge base to filter out outdated or inaccurate knowledge, and ensure the timeliness of the information in the knowledge base.

[0065] S202, training the first model based on the knowledge base to obtain the first analysis entry point of the first industry.

[0066] Similar to S101 described above, the analysis device can use the professional knowledge in the knowledge base, such as the analysis report sample, to train the first model in an unsupervised training manner, so that the first model summarizes the first analysis entry point in the professional knowledge.

[0067] Further, the analysis device can pre-process the sample data before training the first model to ensure data quality and consistency, and then perform data conversion, aggregation or feature extraction to obtain the training input of the first model. The first analysis point can be summarized by repeatedly adjusting the model parameters or trying different model structures.

[0068] In some embodiments, the first analysis point can be summarized from an analysis report sample, and of course can also be summarized from an industry research report, an analysis system information, a listed company annual report, etc.

[0069] The first analysis point can be used as an initial analysis point, and the target analysis point can be obtained by correcting the first analysis point using the second analysis point provided by the expert through S102, and then obtaining the analysis conclusion and the factual basis through S103.

[0070] S102, using a second analysis point of a first industry determined by a preset user, correcting the first analysis point to obtain a target analysis point, the target analysis point being used for data analysis of the first industry.

[0071] S103, inputting enterprise information of a to-be-analyzed enterprise, the target analysis point and analysis configuration information into a second model, and running the second model to output analysis conclusions of the to-be-analyzed enterprise at each analysis point and factual basis used by the analysis conclusions. The to-be-analyzed enterprise belongs to the first industry, and the analysis configuration information indicates an analysis strategy.

[0072] S203, generating a serial speech based on the knowledge base, and integrating a plurality of analysis conclusions corresponding to a plurality of analysis points and a plurality of factual bases using the serial speech to obtain an analysis report of the to-be-analyzed enterprise.

[0073] The analysis device can summarize the serial speech from the professional knowledge in the knowledge base, such as the analysis report sample, and then use it to integrate the analysis report. For example, the analysis device can train a third model based on the knowledge base, so that the third model can summarize the serial speech from the knowledge base, and then use it to generate the analysis report.

[0074] Of course, the analysis device can also use the third model, the third model accesses the knowledge base, and then inputs a plurality of analysis conclusions and a plurality of factual bases into the third model. The third model can perform text processing on the input plurality of analysis conclusions and the plurality of factual bases according to the accessed knowledge base, and integrate to obtain an analysis report conforming to the professional language of the first industry.

[0075] In some embodiments, after generating the analysis report of the enterprise to be analyzed, the analysis device can detect whether the analysis report meets preset conditions. Wherein, if the preset conditions are met, it indicates that the analysis report can be used as an analysis report sample. For example, the preset conditions can be that more than a preset number of users (including experts) confirm that the analysis report is accurate. For another example, the preset conditions can be that the analysis report does not include descriptions in the blacklist.

[0076] In the case where the analysis report meets the preset conditions, the analysis device can add the analysis report to the knowledge base and train an updated first model, and output an updated first analysis entry point using the updated first model. In this way, the analysis device can update the first analysis entry point according to new analysis report samples, and obtain a more accurate first analysis entry point.

[0077] In some embodiments, the analysis configuration information includes at least one of the following: First, a first correspondence relationship between the analysis entry point and the analysis title, and a second correspondence relationship between the analysis title and the analysis point.

[0078] Wherein, the granularity of the analysis entry point is larger, and the granularity of the analysis title is smaller. The analysis entry point is equivalent to a large analysis direction, and the analysis title is equivalent to a refined analysis dimension under the analysis direction.

[0079] After obtaining the target analysis entry point, the analysis device can determine the analysis point corresponding to the target analysis entry point (which can be referred to as the target analysis point) through the first correspondence relationship and the second correspondence relationship. In this way, the analysis device can clearly determine the target analysis point to be analyzed.

[0080] Second, the data sources required by each analysis title.

[0081] Wherein, the granularity of the analysis title is larger, and the granularity of the analysis point is smaller. The analysis point is equivalent to the specific analysis content included in the analysis title.

[0082] After determining the target analysis point, the analysis device can determine the data source for analyzing the target analysis point based on the analysis title to which the target analysis point belongs.

[0083] Of course, the analysis configuration information can also directly include a third correspondence relationship between the analysis entry point and the analysis point, and include the data sources required by each analysis point. The embodiments of the present application do not make specific limitations on this. In this paper, the first and second items are mainly used as examples for illustration.

[0084] Third, the analysis rules of one or more analysis points.

[0085] The analysis rule can indicate the constitution, form, etc. of the analysis conclusion corresponding to the analysis point. Further, the form can be a table, a graph, a text, etc. In this way, the analysis device can determine, based on the analysis rule, which analysis conclusions are to be obtained for a certain analysis point, and in what form the analysis conclusions are to be output.

[0086] Taking a production process with the analysis point as the main business as an example, the analysis rule can be: project acquisition method, including public bidding, invitation bidding, direct contracting, subcontracting, and other five options, and if it is "other", the specific method needs to be specified. That is, the analysis rule indicates that the analysis conclusion needs to include the project acquisition method.

[0087] It should be noted that all or part of the analysis points can have analysis rules, which are not limited in the embodiments of the present application.

[0088] Taking the market positioning of the main business as an example, the analysis configuration information can include the information in Table 2 as follows: Table 2

[0089] The first column and the second column can represent the first correspondence between the analysis entry point and the analysis title, the second column and the fourth column can represent the second correspondence between the analysis title and the analysis point, the second column and the third column can represent the analysis title and the required data source, and the fourth column and the fifth column can represent the analysis point and the analysis rule thereof.

[0090] Referring to Figure 3 , the S103 can include the following steps using the above analysis configuration information: The running of the second model to output the analysis conclusions of the enterprise to be analyzed at each analysis point and the factual basis for using the analysis conclusions can include the following steps: S301, inputting the enterprise information of the enterprise to be analyzed, the target analysis entry point, and the analysis configuration information into the second model.

[0091] S302, running the second model, and determining, according to the first correspondence in the analysis configuration information, a plurality of target analysis titles corresponding to the target analysis entry point for each target analysis entry point.

[0092] Taking the market positioning of the main business as an example, the analysis device can determine, according to the first correspondence shown in the first column and the second column in Table 2, that the target analysis title includes: main business and market situation description; company development project undertaking ability, development ability and management ability description; and market regional share and industry status.

[0093] S303, for any target analysis title, determining a plurality of target analysis points corresponding to the target analysis title according to the second correspondence relationship in the analysis configuration information, and determining target source data corresponding to the target analysis title according to the data source required by each analysis title.

[0094] Taking the target analysis title of main business and market situation description as an example, the analysis device can determine that the target analysis points corresponding to the main business and market situation description include: 1, the overall operation situation, business structure and change situation of the company in recent years, and understand the reasons for the change; 2, combined with the order situation, explain the company's business market sales situation and sustainability, and the matching with the operating income; and 3, other situations that need to be explained, according to the second correspondence relationship between the analysis title and the analysis point shown in the second column and the fourth column in Table 2.

[0095] Continuing to take the target analysis title of main business and market situation description as an example, the analysis device can determine that the data sources required by the main business and market situation description include customer sorted information and network public data, according to the analysis title and the data source required by the analysis title shown in the second column and the third column in Table 2. Among them, the customer sorted information can be understood as the enterprise information of the above-mentioned enterprise to be analyzed, which can be provided by other analyzed enterprises to the analysis device, or can be automatically collected by the analysis device It can be understood that determining the target source data corresponding to the target analysis title is equivalent to determining the data source required for analyzing the target analysis points corresponding to the target analysis title. Still taking the target analysis title of main business and market situation description as an example, the data sources required by the main business and market situation description include customer sorted information and network public data, which can be used to analyze the target analysis points corresponding to the main business and market situation description.

[0096] That is, after S302 and S303, each target analysis point can be subdivided into a target analysis title, and then further subdivided into a target analysis point, so as to facilitate subsequent detailed analysis of each target analysis point. And after S302 and S303, the target data source used for analyzing each target analysis point can also be determined.

[0097] S304, for any target analysis point, determining whether the target analysis point has a corresponding target analysis rule. If yes, execute S305; if no, execute S306.

[0098] Based on Table 2, it is not difficult to see that some analysis points have analysis rules, and some analysis points do not have analysis rules. For these two different cases, corresponding methods can be used for analysis.

[0099] S305, according to the target analysis rule, the target analysis point belongs to the target analysis title corresponding to the target source data analysis, the analysis conclusion of the target analysis point and the analysis conclusion is very practical fact basis.

[0100] For example, the target analysis point is "the overall operation of the company in recent years, business structure and change, and understand the reasons for the change" in Table 2 above. The target analysis point has a corresponding target analysis rule "list the data of the last three years by year". Therefore, when the analysis device analyzes the target analysis point, it needs to analyze the target source data and list the data of the last three years by year according to the target analysis rule, and obtain the analysis conclusion of the target analysis point and its fact basis.

[0101] That is, for the target analysis point with target analysis rule, not only the analysis point analysis target source data, but also follow the target analysis rule, so as to ensure that the analysis conclusion meets the requirements of content and form.

[0102] S306, according to the target analysis rule, the target analysis point belongs to the target analysis title corresponding to the target source data analysis, the analysis conclusion of the target analysis point and the analysis conclusion is very practical fact basis.

[0103] For example, the target analysis point is "the overall operation of the company in recent years, business structure and change, and understand the reasons for the change" in Table 2 above. The target analysis point has a corresponding target analysis rule "list the data of the last three years by year". Therefore, when the analysis device analyzes the target analysis point, it needs to analyze the target source data and list the data of the last three years by year according to the target analysis rule, and obtain the analysis conclusion of the target analysis point and its fact basis.

[0104] Figure 4 A structure diagram of a report generation device provided by an embodiment of the present application is shown in FIG. 4. As shown in FIG. 4, the report generation device includes a training module 401, a correction module 402, an analysis module 403, and an integration module 404. Figure 4

[0105] The training module 401 is configured to train a first model based on analysis report samples of a plurality of enterprises in a first industry, to obtain a first analysis entry point of the first industry. The first model has the ability to summarize the analysis entry point. The correction module 402 is configured to correct the first analysis entry point based on a second analysis entry point of the first industry determined by a preset user, to obtain a target analysis entry point, which is used for data analysis of the first industry. ​The analysis module 403 is configured to input the enterprise information of the enterprise to be analyzed, the target analysis entry point, and analysis configuration information into a second model, and run the second model to output analysis conclusions of the enterprise to be analyzed at each analysis point and factual basis used by the analysis conclusions; the second model has the capability of analyzing and generating analysis conclusions and factual basis, the enterprise to be analyzed belongs to the first industry, one target analysis entry point corresponds to one or more analysis points, and the analysis configuration information indicates an analysis strategy. The integration module 404 is configured to integrate a plurality of analysis conclusions and a plurality of factual basis corresponding to the plurality of analysis points based on the serial dialogue to obtain an analysis report of the enterprise to be analyzed.

[0106] In some other embodiments, as shown in Figure 5 The report generation apparatus can further include a construction module 501 configured to construct a knowledge base of the first industry, the knowledge base including analysis report samples, and the knowledge base further including at least one of the following: an industry research report, analysis system information, and an annual report of a listed company.

[0107] The training module 401 can be further configured to train the first model based on the knowledge base.

[0108] In some other embodiments, as shown in Figure 5 The report generation apparatus can further include a generation module 502 configured to generate the serial dialogue based on the knowledge base.

[0109] In some other embodiments, the construction module 501 can be further configured to add the analysis report as an analysis report sample when the analysis report meets a preset condition, and the training module 401 can be further configured to retrain the first model to obtain an updated first analysis entry point.

[0110] In some other embodiments, the analysis configuration information includes at least one of the following: a first correspondence relationship between each target analysis entry point and a plurality of analysis titles, and a second correspondence relationship between each analysis title and a plurality of analysis points; a data source required by each analysis title; and an analysis rule of one or more analysis points.

[0111] In other embodiments, the analysis module 403 may also be used to: run the second model, for each target analysis entry point, determine multiple target analysis titles corresponding to the target analysis entry point according to the first correspondence; for any target analysis title, determine multiple target analysis points corresponding to the target analysis title according to the second correspondence, and determine target source data corresponding to the target analysis title according to the data sources required for each analysis title; for any target analysis point, determine whether the target analysis point has a corresponding target analysis rule; if the target analysis point has a target analysis rule, analyze the target source data corresponding to the target analysis title to which the target analysis point belongs according to the target analysis rule to obtain the analysis conclusion of the target analysis point and the factual basis for the practicality of the analysis conclusion; if the target analysis point does not have a target analysis rule, analyze the target source data corresponding to the target analysis title to which the target analysis point belongs to obtain the analysis conclusion of the target analysis point and the factual basis for the practicality of the analysis conclusion.

[0112] In other embodiments, the target analysis entry point includes the market positioning of the main business; the target analysis title corresponding to the market positioning of the main business includes at least two of the following: description of the main business and market positioning; description of project undertaking capabilities, development capabilities and management capabilities; and market share and industry position.

[0113] The report generation apparatus provided in this application embodiment can execute the method shown in the above method embodiment. Its implementation principle and beneficial effects can be referred to the relevant description in the method embodiment, and will not be repeated here.

[0114] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. This electronic device can be the device analyzed above. For example... Figure 6 As shown, the electronic device includes: a memory 601, a transceiver 602, and at least one processor 603.

[0115] The transceiver 602 is used to interact with other devices to send and receive data. For example, in this embodiment, the transceiver 602 can specifically be used to receive a second analysis entry point determined by a preset user, professional knowledge of the first industry, and enterprise information of the company to be analyzed. The memory 601 is configured to store computer program code including computer instructions. The computer instructions run in the electronic device described above to implement the method shown in the method embodiments described above. For example, the memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), such as at least one disk memory, and can also be a U disk, a mobile hard disk, a read-only memory, a magnetic disk or an optical disk, etc.

[0116] The processor 603 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The processor 503 can also be other general-purpose processors. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0117] The memory 601, the transceiver 602 and the processor 603 are communicatively connected. For example, the memory 601, the transceiver 602 can be connected with the processor 603 through a system bus and complete mutual communication. The system bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, an industry standard architecture (ISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or only one type of bus.

[0118] Optionally, the memory 601 can be independent or integrated with the processor 603. When the memory 601 is independently arranged, the memory 601 and the processor 603 are connected through a system bus.

[0119] The embodiments of the present application also provide a chip for running instructions, which is used to execute the technical solutions of the report generation method in the above embodiments.

[0120] The embodiment of the present application further provides a computer readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the technical solution of the report generation method in the above embodiment is implemented. Specifically, when the computer instructions are executed by the processor, the electronic device can execute the technical solution of the report generation method in the above embodiment.

[0121] The embodiment of the present application further provides a computer program product, which comprises a computer program stored in a computer readable storage medium. At least one processor can read the computer program from the computer readable storage medium, and when the at least one processor executes the computer program, the technical solution of the report generation method in the above embodiment can be implemented. The computer readable storage medium can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The computer readable storage medium can be any available medium that can be accessed by a general or special purpose computer. An exemplary computer readable storage medium is coupled to the processor, so that the processor can read information from the storage medium and write information to the computer readable storage medium. Of course, the computer readable storage medium can also be an integral part of the processor. The processor and the computer readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the computer readable storage medium can also exist as discrete components in an electronic control unit or a host device, and the embodiment of the present application does not limit this.

[0122] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. For example, the division of the above-described device embodiments is merely a logical function division, and there can be another division manner for the actual implementation, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different modules can be indirect couplings or communication connections through some interfaces, devices or modules, and can be electrical, mechanical or in other forms.

[0123] The modules illustrated as separated components can or can not be physically separated, and the components illustrated as modules can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to implement the embodiments of the present application.

[0124] In addition, each functional module in the embodiments of the present application can be integrated in one processing unit, or each module can be physically present separately, or two or more modules can be integrated in one unit. The above-mentioned units can be implemented in the form of hardware, or in the form of hardware plus software functional units.

[0125] The integrated modules implemented in the form of software functional modules can be stored in a computer readable storage medium. The software functional modules stored in the storage medium include a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the embodiments of the present application.

[0126] It should be understood that the steps of the method disclosed in the embodiments of the present application can be directly embodied as hardware processor execution, or executed by a combination of hardware and software modules in the processor.

[0127] Those of ordinary skill in the art can understand that all or part of the above-mentioned method embodiments can be completed by program instructions related to hardware. The foregoing program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the foregoing storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0128] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A report generation method characterized by, The method comprises: training a first model based on analysis report samples of a plurality of enterprises in a first industry to obtain a first analysis entry point of the first industry; wherein the first model has the ability to summarize the analysis entry point; correcting the first analysis entry point based on a second analysis entry point of the first industry determined by a preset user to obtain a target analysis entry point, which is used for data analysis of the first industry; inputting enterprise information of an enterprise to be analyzed, the target analysis entry point, and analysis configuration information into a second model, and running the second model to output analysis conclusions of the enterprise to be analyzed at each analysis point and factual basis for using the analysis conclusions; wherein the second model has the ability to analyze and generate analysis conclusions and factual basis, the enterprise to be analyzed belongs to the first industry, one target analysis entry point corresponds to one or more analysis points, and the analysis configuration information indicates an analysis strategy; integrating a plurality of analysis conclusions and a plurality of factual basis corresponding to the plurality of analysis points based on a serial speech to obtain an analysis report of the enterprise to be analyzed.

2. The method of claim 1, wherein, Before the training of the first model based on the analysis report samples of the plurality of enterprises in the first industry to obtain the first analysis entry point of the first industry, the method further comprises: constructing a knowledge base of the first industry, wherein the knowledge base includes analysis report samples, and the knowledge base further includes at least one of the following: industry research reports, analysis system information, and annual reports of listed companies; the training of the first model based on the analysis report samples of the plurality of enterprises in the first industry comprises: training the first model based on the knowledge base.

3. The method of claim 2, wherein, Before the integration of the plurality of analysis conclusions and the plurality of factual basis corresponding to the plurality of analysis points based on the serial speech to obtain the analysis report of the enterprise to be analyzed, the method further comprises: generating the serial speech based on the knowledge base.

4. The method of claim 2, wherein, After the integration to obtain the analysis report of the enterprise to be analyzed, the method further comprises: in the case that the analysis report meets a preset condition, adding the analysis report to the knowledge base and retraining the first model to obtain an updated first analysis entry point.

5. The method according to any one of claims 1-4, characterized in that, The analysis configuration information includes at least one of the following: a first correspondence relationship between each target analysis entry point and a plurality of analysis titles, and a second correspondence relationship between each analysis title and a plurality of analysis points; data sources required by each analysis title; and analysis rules of one or more analysis points.

6. The method of claim 5, wherein, The running of the second model to output the analysis conclusions of the enterprise to be analyzed at each analysis point and the factual basis for using the analysis conclusions comprises: running the second model to determine, for each target analysis entry point, a plurality of target analysis titles corresponding to the target analysis entry point according to the first correspondence relationship; for any target analysis title, determining a plurality of target analysis points corresponding to the target analysis title according to the second correspondence relationship, and determining target source data corresponding to the target analysis title according to the data sources required by each analysis title; for any target analysis point, determining whether the target analysis point has a corresponding target analysis rule; If the target analysis point has a target analysis rule, the target source data corresponding to a target analysis title to which the target analysis point belongs is analyzed according to the target analysis rule, to obtain an analysis conclusion of the target analysis point and a fact basis for practicality of the analysis conclusion; If the target analysis point does not have a target analysis rule, the target source data corresponding to a target analysis title to which the target analysis point belongs is analyzed, to obtain an analysis conclusion of the target analysis point and a fact basis for practicality of the analysis conclusion.

7. The method of claim 5, wherein, The target analysis entry point includes a main business market positioning. The target analysis title corresponding to the main business market positioning includes at least two of the following: main business and market positioning description; project undertaking ability, development ability and management ability description; and market regional share and industry status.

8. A report generating apparatus characterized by comprising: Comprise: The training module is configured to train a first model based on analysis report samples of a plurality of enterprises in a first industry, to obtain a first analysis entry point of the first industry; wherein the first model has the ability to summarize the analysis entry point; The correction module is configured to correct the first analysis entry point based on a second analysis entry point of the first industry determined by a preset user, to obtain a target analysis entry point, which is used for data analysis of the first industry; The analysis module is configured to input enterprise information of a to-be-analyzed enterprise, the target analysis entry point and analysis configuration information into a second model, and run the second model to output analysis conclusions of the to-be-analyzed enterprise at each analysis point and fact bases for use of the analysis conclusions; wherein the second model has the ability to analyze and generate the analysis conclusions and the fact bases, the to-be-analyzed enterprise belongs to the first industry, one target analysis entry point corresponds to one or more analysis points, and the analysis configuration information indicates an analysis strategy; The integration module is configured to integrate a plurality of analysis conclusions and a plurality of fact bases corresponding to the plurality of analysis points based on a serial speech to obtain an analysis report of the to-be-analyzed enterprise.

9. An electronic device, comprising: Comprise: A memory and at least one processor; the memory is in communication connection with the processor; the memory is used to store computer program code, the computer program code comprises computer instructions; when the processor executes the computer instructions, the electronic device executes the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are executed by the processor to implement the method of any one of claims 1-7.

11. A computer program product, characterised in that, When the computer program product is running on the computer / being executed by the processor of the computer, the method of any one of claims 1-7 is implemented.

Images